Merge branch 'main' into fix/fd_1828

* put removeListner in a try/catch

* typo

.gitignore

@@ -2,6 +2,9 @@
 logs
 *.log
 
+.claude/
+CLAUDE.md 
+
 # Runtime data
 pids
 *.pid

app.js

@@ -29,6 +29,12 @@ const {LifeCycleEvents, FS_UUID_SET_NAME, SystemState, FEATURE_SERVER} = require
 const installSrfLocals = require('./lib/utils/install-srf-locals');
 const createHttpListener = require('./lib/utils/http-listener');
 const healthCheck = require('@jambonz/http-health-check');
+const ProcessMonitor = require('./lib/utils/process-monitor');
+const monitor = new ProcessMonitor(logger);
+
+// Log startup
+monitor.logStartup();
+monitor.setupSignalHandlers();
 
 logger.on('level-change', (lvl, _val, prevLvl, _prevVal, instance) => {
   if (logger !== instance) {

data/example-voicemail-greetings.json

@@ -174,5 +174,61 @@
     "non è raggiungibile",
     "lascia pure un messaggio",
     "puoi lasciare un messaggio"
+  ],
+  "ja-JP": [
+    "この通話は留守番電話に転送されました",
+    "発信先は現在電話に出ることができません",
+    "発信音の後でメッセージを録音してください",
+    "録音を完了したら電話を切ることができます",
+    "只今電話に出ることができません",
+    "ただ今電話に出ることができません",
+    "ただいま電話に出ることができません",
+    "ピーという発信音の後にお名前とご用件をお話しください",
+    "ファックスを送られる方はスタートボタンを押してください",
+    "FAXを送られる方はスタートボタンを押してください",
+    "おかけになった電話をお呼びしましたが",
+    "お出になりません",
+    "おでになりません",
+    "お掛けになった電話番号は",
+    "おかけになった電話番号は",
+    "お掛けになった電話は",
+    "おかけになった電話は",
+    "現在使われておりません",
+    "番号をお確かめになって",
+    "お掛け直し下さい",
+    "おかけ直し下さい",
+    "おかけ直しください",
+    "こちらはNTTドコモです",
+    "こちらはエーユーです",
+    "こちらはソフトバンクです",
+    "電波の届かない",
+    "電源が入っていない",
+    "掛かりません",
+    "かかりません",
+    "お繋ぎすることが出来ません",
+    "お繋ぎ出来ません",
+    "お繋ぎすることができません",
+    "お繋ぎできません",
+    "おつなぎすることができません",
+    "おつなぎできません",
+    "メッセージを録音",
+    "留守番電話",
+    "お留守番サービス",
+    "留守番",
+    "留守電",
+    "留守",
+    "接続します",
+    "合図の音",
+    "ピーと",
+    "発信音",
+    "ご用件",
+    "伝言",
+    "お話しください",
+    "ファックス",
+    "FAX",
+    "終了",
+    "終了しました",
+    "終了いたしました",
+    "営業時間"
   ]
 }

lib/config.js

@@ -119,7 +119,7 @@ const ENCRYPTION_SECRET = process.env.ENCRYPTION_SECRET;
 const HTTP_POOL = process.env.HTTP_POOL && parseInt(process.env.HTTP_POOL);
 const HTTP_POOLSIZE = parseInt(process.env.HTTP_POOLSIZE, 10) || 10;
 const HTTP_PIPELINING = parseInt(process.env.HTTP_PIPELINING, 10) || 1;
-const HTTP_TIMEOUT = 10000;
+const HTTP_TIMEOUT = parseInt(process.env.JAMBONES_HTTP_TIMEOUT, 10) || 10000;
 const HTTP_PROXY_IP = process.env.JAMBONES_HTTP_PROXY_IP;
 const HTTP_PROXY_PORT = process.env.JAMBONES_HTTP_PROXY_PORT;
 const HTTP_PROXY_PROTOCOL = process.env.JAMBONES_HTTP_PROXY_PROTOCOL || 'http';
@@ -130,7 +130,7 @@ const OPTIONS_PING_INTERVAL = parseInt(process.env.OPTIONS_PING_INTERVAL, 10) ||
 const JAMBONZ_RECORD_WS_BASE_URL = process.env.JAMBONZ_RECORD_WS_BASE_URL || process.env.JAMBONES_RECORD_WS_BASE_URL;
 const JAMBONZ_RECORD_WS_USERNAME = process.env.JAMBONZ_RECORD_WS_USERNAME || process.env.JAMBONES_RECORD_WS_USERNAME;
 const JAMBONZ_RECORD_WS_PASSWORD = process.env.JAMBONZ_RECORD_WS_PASSWORD || process.env.JAMBONES_RECORD_WS_PASSWORD;
-const JAMBONZ_DISABLE_DIAL_PAI_HEADER = process.env.JAMBONZ_DISABLE_DIAL_PAI_HEADER || false;
+const JAMBONZ_DIAL_PAI_HEADER = process.env.JAMBONZ_DIAL_PAI_HEADER || false;
 const JAMBONES_DISABLE_DIRECT_P2P_CALL = process.env.JAMBONES_DISABLE_DIRECT_P2P_CALL || false;
 
 const JAMBONES_EAGERLY_PRE_CACHE_AUDIO = parseInt(process.env.JAMBONES_EAGERLY_PRE_CACHE_AUDIO, 10) || 0;
@@ -139,6 +139,11 @@ const JAMBONES_USE_FREESWITCH_TIMER_FD = process.env.JAMBONES_USE_FREESWITCH_TIM
 const JAMBONES_DIAL_SBC_FOR_REGISTERED_USER = process.env.JAMBONES_DIAL_SBC_FOR_REGISTERED_USER || false;
 const JAMBONES_MEDIA_TIMEOUT_MS = process.env.JAMBONES_MEDIA_TIMEOUT_MS || 0;
 const JAMBONES_MEDIA_HOLD_TIMEOUT_MS = process.env.JAMBONES_MEDIA_HOLD_TIMEOUT_MS || 0;
+const JAMBONES_WEBHOOK_ERROR_RETURN = parseInt(process.env.JAMBONES_WEBHOOK_ERROR_RETURN, 10) || 480;
+
+/* say / tts */
+const JAMBONES_SAY_CHUNK_SIZE = parseInt(process.env.JAMBONES_SAY_CHUNK_SIZE, 10) || 900;
+
 // jambonz
 const JAMBONES_TRANSCRIBE_EP_DESTROY_DELAY_MS =
   process.env.JAMBONES_TRANSCRIBE_EP_DESTROY_DELAY_MS;
@@ -225,11 +230,13 @@ module.exports = {
   JAMBONZ_RECORD_WS_BASE_URL,
   JAMBONZ_RECORD_WS_USERNAME,
   JAMBONZ_RECORD_WS_PASSWORD,
-  JAMBONZ_DISABLE_DIAL_PAI_HEADER,
+  JAMBONZ_DIAL_PAI_HEADER,
   JAMBONES_DISABLE_DIRECT_P2P_CALL,
   JAMBONES_USE_FREESWITCH_TIMER_FD,
   JAMBONES_DIAL_SBC_FOR_REGISTERED_USER,
   JAMBONES_MEDIA_TIMEOUT_MS,
   JAMBONES_MEDIA_HOLD_TIMEOUT_MS,
-  JAMBONES_TRANSCRIBE_EP_DESTROY_DELAY_MS
+  JAMBONES_SAY_CHUNK_SIZE,
+  JAMBONES_TRANSCRIBE_EP_DESTROY_DELAY_MS,
+  JAMBONES_WEBHOOK_ERROR_RETURN
 };

lib/http-routes/api/create-call.js

@@ -13,10 +13,11 @@ const WsRequestor = require('../../utils/ws-requestor');
 const RootSpan = require('../../utils/call-tracer');
 const dbUtils = require('../../utils/db-utils');
 const { decrypt } = require('../../utils/encrypt-decrypt');
-const { mergeSdpMedia, extractSdpMedia } = require('../../utils/sdp-utils');
+const { mergeSdpMedia, extractSdpMedia, removeVideoSdp } = require('../../utils/sdp-utils');
 const { createCallSchema, customSanitizeFunction } = require('../schemas/create-call');
 const { selectHostPort } = require('../../utils/network');
 const { JAMBONES_DIAL_SBC_FOR_REGISTERED_USER } = require('../../config');
+const { createMediaEndpoint } = require('../../utils/media-endpoint');
 
 const removeNullProperties = (obj) => (Object.keys(obj).forEach((key) => obj[key] === null && delete obj[key]), obj);
 const removeNulls = (req, res, next) => {
@@ -67,7 +68,7 @@ router.post('/',
       const {
         lookupAppBySid
       }  = srf.locals.dbHelpers;
-      const {getSBC, getFreeswitch} = srf.locals;
+      const {getSBC} = srf.locals;
       let sbcAddress = getSBC();
       if (!sbcAddress) throw new Error('no available SBCs for outbound call creation');
       const target = restDial.to;
@@ -146,7 +147,7 @@ router.post('/',
 
       // find handling sbc sip for called user
       if (JAMBONES_DIAL_SBC_FOR_REGISTERED_USER && target.type === 'user') {
-        const { registrar }  = srf.locals.dbHelpers;
+        const { registrar}  = srf.locals.dbHelpers;
         const reg = await registrar.query(target.name);
         if (reg) {
           sbcAddress = selectHostPort(logger, reg.sbcAddress, 'tcp')[1];
@@ -158,7 +159,9 @@ router.post('/',
      * trunk isn't specified,
      * check if from-number matches any existing numbers on Jambonz
      *  */
-      if (target.type === 'phone' && !target.trunk) {
+      const { lookupLcrByAccount}  = srf.locals.dbHelpers;
+      const lcrs = await lookupLcrByAccount(req.body.account_sid);
+      if (target.type === 'phone' && !target.trunk && lcrs.length == 0) {
         const str = restDial.from || '';
         const callingNumber = str.startsWith('+') ? str.substring(1) : str;
         const voip_carrier_sid = await lookupCarrierByPhoneNumber(req.body.account_sid, callingNumber);
@@ -170,9 +173,7 @@ router.post('/',
       }
 
       /* create endpoint for outdial */
-      const ms = getFreeswitch();
-      if (!ms) throw new Error('no available Freeswitch for outbound call creation');
-      const ep = await ms.createEndpoint();
+      const ep = await createMediaEndpoint(srf, logger);
       logger.debug(`createCall: successfully allocated endpoint, sending INVITE to ${sbcAddress}`);
 
       /* launch outdial */
@@ -181,10 +182,14 @@ router.post('/',
       let localSdp = ep.local.sdp;
 
       if (req.body.dual_streams) {
-        dualEp = await ms.createEndpoint();
+        dualEp = await createMediaEndpoint(srf, logger);
         localSdp = mergeSdpMedia(localSdp, dualEp.local.sdp);
       }
-
+      if (process.env.JAMBONES_VIDEO_CALLS_ENABLED_IN_FS) {
+        logger.debug('createCall: removing video sdp');
+        localSdp = removeVideoSdp(localSdp);
+        ep.modify(localSdp);
+      }
       const connectStream = async(remoteSdp) => {
         if (remoteSdp !== sdp) {
           sdp = remoteSdp;
@@ -235,6 +240,7 @@ router.post('/',
         if (app.call_hook.url === app.call_status_hook?.url  || !app.call_status_hook?.url) {
           logger.debug('reusing websocket for call status hook');
           app.notifier = app.requestor;
+          app.call_status_hook = app.call_hook;
         }
       }
       else {
@@ -285,7 +291,7 @@ router.post('/',
             }, {
               ...(account.enable_debug_log && {level: 'debug'})
             });
-            app.requestor.logger = app.notifier.logger = sipLogger;
+            app.requestor.logger = app.notifier.logger = restDial.logger = sipLogger;
             const callInfo = new CallInfo({
               direction: CallDirection.Outbound,
               req: inviteReq,

lib/middleware.js

@@ -12,7 +12,8 @@ const RootSpan = require('./utils/call-tracer');
 const listTaskNames = require('./utils/summarize-tasks');
 const {
   JAMBONES_MYSQL_REFRESH_TTL,
-  JAMBONES_DISABLE_DIRECT_P2P_CALL
+  JAMBONES_DISABLE_DIRECT_P2P_CALL,
+  JAMBONES_WEBHOOK_ERROR_RETURN
 } = require('./config');
 const { createJambonzApp } = require('./dynamic-apps');
 const { decrypt } = require('./utils/encrypt-decrypt');
@@ -112,6 +113,14 @@ module.exports = function(srf, logger) {
         req.locals.callingNumber = sipURIs[1];
       }
     }
+
+    // Feature server INVITE request pipelines taking time to finish,
+    // while connecting and fetch application from db and invoking webhook.
+    // call can be canceled without any handling, so we add a listener here
+    req.once('cancel', (sipMsg) => {
+      logger.info(`${callId} got CANCEL request`);
+      req.locals.canceled = true;
+    });
     next();
   }
 
@@ -362,13 +371,14 @@ module.exports = function(srf, logger) {
       });
       // if transferred call contains callInfo, let update original data to newly created callInfo in this instance.
       if (app.transferredCall && app.callInfo) {
-        const {direction, callerName, from, to, originatingSipIp, originatingSipTrunkName} = app.callInfo;
+        const {direction, callerName, from, to, originatingSipIp, originatingSipTrunkName, customerData} = app.callInfo;
         req.locals.callInfo.direction = direction;
         req.locals.callInfo.callerName = callerName;
         req.locals.callInfo.from = from;
         req.locals.callInfo.to = to;
         req.locals.callInfo.originatingSipIp = originatingSipIp;
         req.locals.callInfo.originatingSipTrunkName = originatingSipTrunkName;
+        if (customerData) req.locals.callInfo.customerData = customerData;
         delete app.callInfo;
       }
       next();
@@ -471,7 +481,7 @@ module.exports = function(srf, logger) {
         message: `${err?.message}`.trim()
       }).catch((err) => this.logger.info({err}, 'Error generating alert for parsing application'));
       logger.info({err}, `Error retrieving or parsing application: ${err?.message}`);
-      res.send(480, {headers: {'X-Reason': err?.message || 'unknown'}});
+      res.send(JAMBONES_WEBHOOK_ERROR_RETURN, {headers: {'X-Reason': err?.message || 'unknown'}});
       app.requestor.close(WS_CLOSE_CODES.GoingAway);
     }
   }

lib/session/call-session.js

@@ -10,8 +10,9 @@ const {
   RecordState,
   AllowedSipRecVerbs,
   AllowedConfirmSessionVerbs,
-  TtsStreamingEvents
-} = require('../utils/constants');
+  TtsStreamingEvents,
+  ListenStatus
+} = require('../utils/constants.json');
 const moment = require('moment');
 const assert = require('assert');
 const sessionTracker = require('./session-tracker');
@@ -29,10 +30,6 @@ const {
   JAMBONES_INJECT_CONTENT,
   JAMBONES_EAGERLY_PRE_CACHE_AUDIO,
   AWS_REGION,
-  JAMBONES_USE_FREESWITCH_TIMER_FD,
-  JAMBONES_MEDIA_TIMEOUT_MS,
-  JAMBONES_MEDIA_HOLD_TIMEOUT_MS,
-  JAMBONES_TRANSCRIBE_EP_DESTROY_DELAY_MS
 } = require('../config');
 const bent = require('bent');
 const BackgroundTaskManager = require('../utils/background-task-manager');
@@ -40,7 +37,8 @@ const dbUtils = require('../utils/db-utils');
 const BADPRECONDITIONS = 'preconditions not met';
 const CALLER_CANCELLED_ERR_MSG = 'Response not sent due to unknown transaction';
 const { NonFatalTaskError} = require('../utils/error');
-const { sleepFor } = require('../utils/helpers');
+const { createMediaEndpoint } = require('../utils/media-endpoint');
+const SttLatencyCalculator = require('../utils/stt-latency-calculator');
 const sqlRetrieveQueueEventHook = `SELECT * FROM webhooks 
 WHERE webhook_sid = 
 (
@@ -150,6 +148,30 @@ class CallSession extends Emitter {
     this.conversationTurns = [];
     this.on('userSaid', this._onUserSaid.bind(this));
     this.on('botSaid', this._onBotSaid.bind(this));
+    /**
+     * Support STT latency
+     */
+    this.sttLatencyCalculator = new SttLatencyCalculator({
+      logger,
+      cs: this
+    });
+    this.on('transcribe-start', () => {
+      this.sttLatencyCalculator.resetTime();
+    });
+    this.on('on-transcription', () => {
+      this.sttLatencyCalculator.onTranscriptionReceived();
+    });
+    this.on('transcribe-stop', () => {
+      this.sttLatencyCalculator.onTranscribeStop();
+    });
+  }
+
+  get notifySttLatencyEnabled() {
+    return this._notifySttLatencyEnabled || false;
+  }
+
+  set notifySttLatencyEnabled(enabled) {
+    this._notifySttLatencyEnabled = enabled;
   }
 
   /**
@@ -482,7 +504,12 @@ class CallSession extends Emitter {
   }
 
   get isTtsStreamEnabled() {
-    return this.backgroundTaskManager.isTaskRunning('ttsStream');
+    // 1st background tts stream
+    return this.backgroundTaskManager.isTaskRunning('ttsStream') ||
+    // 2nd current task streaming tts
+    TaskName.Say === this.currentTask?.name && this.currentTask?.isStreamingTts ||
+    // 3rd nested verb is streaming tts
+    TaskName.Gather === this.currentTask?.name && this.currentTask.sayTask?.isStreamingTts;
   }
 
   get isListenEnabled() {
@@ -636,6 +663,15 @@ class CallSession extends Emitter {
     }
   }
 
+  // disableTtsCache
+  get disableTtsCache() {
+    return this._disableTtsCache || false;
+  }
+
+  set disableTtsCache(d) {
+    this._disableTtsCache = d;
+  }
+
   getTsStreamingVendor() {
     let v;
     if (this.currentTask?.isStreamingTts) {
@@ -688,7 +724,7 @@ class CallSession extends Emitter {
   }
 
 
-  hasGlobalSttPunctuation() {
+  get hasGlobalSttPunctuation() {
     return this._globalSttPunctuation !== undefined;
   }
 
@@ -720,10 +756,25 @@ class CallSession extends Emitter {
     return this._fillerNoise;
   }
 
+  async pauseOrResumeBackgroundListenIfRequired(action, silence = false) {
+    if ((action == 'pauseCallRecording' || action == 'resumeCallRecording') &&
+    this.backgroundTaskManager.isTaskRunning('record')) {
+      this.logger.debug({action, silence}, 'CallSession:pauseOrResumeBackgroundListenIfRequired');
+      const backgroundListenTask = this.backgroundTaskManager.getTask('record');
+      const status = action === 'pauseCallRecording' ? ListenStatus.Pause : ListenStatus.Resume;
+      backgroundListenTask.updateListen(
+        status,
+        silence
+      );
+    }
+  }
+
   async notifyRecordOptions(opts) {
-    const {action} = opts;
+    const {action, silence} = opts;
     this.logger.debug({opts}, 'CallSession:notifyRecordOptions');
 
+    this.pauseOrResumeBackgroundListenIfRequired(action, silence);
+
     /* if we have not answered yet, just save the details for later */
     if (!this.dlg) {
       if (action === 'startCallRecording') {
@@ -881,7 +932,7 @@ class CallSession extends Emitter {
         this.logger.debug('CallSession:enableBackgroundTtsStream - ttsStream enabled');
       } else {
         this.logger.debug(
-          'CallSession:enableBackgroundTtsStream - ignoring request as call does not have required conditions');
+          'CallSession:enableBackgroundTtsStream - ignoring request; conditions not met (probably not using ws api)');
       }
     }  catch (err) {
       this.logger.info({err, say}, 'CallSession:enableBackgroundTtsStream - Error creating background tts stream task');
@@ -895,15 +946,25 @@ class CallSession extends Emitter {
     }
   }
   clearTtsStream() {
-    this.requestor?.request('tts:streaming-event', '/streaming-event', {event_type: 'user_interruption'})
-      .catch((err) => this.logger.info({err}, 'CallSession:clearTtsStream - Error sending user_interruption'));
-    this.ttsStreamingBuffer?.clear();
+    if (this.isTtsStreamEnabled) {
+      this.requestor?.request('tts:streaming-event', '/streaming-event', {event_type: 'user_interruption'})
+        .catch((err) => this.logger.info({err}, 'CallSession:clearTtsStream - Error sending user_interruption'));
+      this.ttsStreamingBuffer?.clear();
+    }
   }
 
   startTtsStream() {
     this.ttsStreamingBuffer?.start();
   }
 
+  stopTtsStream() {
+    if (this.isTtsStreamEnabled) {
+      this.requestor?.request('tts:streaming-event', '/streaming-event', {event_type: 'stream_closed'})
+        .catch((err) => this.logger.info({err}, 'CallSession:clearTtsStream - Error sending user_interruption'));
+      this.ttsStreamingBuffer?.stop();
+    }
+  }
+
   async enableBotMode(gather, autoEnable) {
     try {
       let task;
@@ -927,7 +988,7 @@ class CallSession extends Emitter {
       task.sticky = autoEnable;
       // listen to the bargein-done from background manager
       this.backgroundTaskManager.on('bargeIn-done', () => {
-        if (this.requestor instanceof WsRequestor) {
+        if (this.appIsUsingWebsockets) {
           try {
             this.kill(true);
           } catch (err) {}
@@ -980,8 +1041,6 @@ class CallSession extends Emitter {
         (type === 'tts' && credential.use_for_tts) ||
         (type === 'stt' && credential.use_for_stt)
       )) {
-        this.logger.debug(
-          `${type}: ${credential.vendor} ${credential.label ? `, label: ${credential.label}` : ''}  `);
         if ('google' === vendor) {
           if (type === 'tts' && !credential.tts_tested_ok ||
             type === 'stt' && !credential.stt_tested_ok) {
@@ -1051,6 +1110,13 @@ class CallSession extends Emitter {
             deepgram_stt_use_tls: credential.deepgram_stt_use_tls
           };
         }
+        else if ('gladia' === vendor) {
+          return {
+            speech_credential_sid: credential.speech_credential_sid,
+            api_key: credential.api_key,
+            region: credential.region,
+          };
+        }
         else if ('soniox' === vendor) {
           return {
             speech_credential_sid: credential.speech_credential_sid,
@@ -1082,6 +1148,7 @@ class CallSession extends Emitter {
           return {
             api_key: credential.api_key,
             model_id: credential.model_id,
+            api_uri: credential.api_uri,
             options: credential.options
           };
         }
@@ -1097,6 +1164,7 @@ class CallSession extends Emitter {
           return {
             api_key: credential.api_key,
             model_id: credential.model_id,
+            stt_model_id: credential.stt_model_id,
             embedding: credential.embedding,
             options: credential.options
           };
@@ -1108,10 +1176,40 @@ class CallSession extends Emitter {
             options: credential.options
           };
         }
+        else if ('resemble' === vendor) {
+          return {
+            api_key: credential.api_key,
+            resemble_tts_use_tls: credential.resemble_tts_use_tls,
+            resemble_tts_uri: credential.resemble_tts_uri,
+          };
+        }
+        else if ('inworld' === vendor) {
+          return {
+            api_key: credential.api_key,
+            model_id: credential.model_id,
+            options: credential.options
+          };
+        }
         else if ('assemblyai' === vendor) {
           return {
             speech_credential_sid: credential.speech_credential_sid,
-            api_key: credential.api_key
+            api_key: credential.api_key,
+            service_version: credential.service_version
+          };
+        }
+        else if ('houndify' === vendor) {
+          return {
+            speech_credential_sid: credential.speech_credential_sid,
+            client_id: credential.client_id,
+            client_key: credential.client_key,
+            user_id: credential.user_id,
+            houndify_server_uri: credential.houndify_server_uri
+          };
+        }
+        else if ('deepgramflux' === vendor) {
+          return {
+            speech_credential_sid: credential.speech_credential_sid,
+            api_key: credential.api_key,
           };
         }
         else if ('voxist' === vendor) {
@@ -1157,9 +1255,10 @@ class CallSession extends Emitter {
       }
       else {
         writeAlerts({
-          alert_type: AlertType.STT_NOT_PROVISIONED,
+          alert_type: type === 'tts' ? AlertType.TTS_NOT_PROVISIONED : AlertType.STT_NOT_PROVISIONED,
           account_sid: this.accountSid,
           vendor,
+          label,
           target_sid: this.callSid
         }).catch((err) => this.logger.error({err}, 'Error writing tts alert'));
       }
@@ -1190,6 +1289,7 @@ class CallSession extends Emitter {
       this.ttsStreamingBuffer.on(TtsStreamingEvents.Pause, this._onTtsStreamingPause.bind(this));
       this.ttsStreamingBuffer.on(TtsStreamingEvents.Resume, this._onTtsStreamingResume.bind(this));
       this.ttsStreamingBuffer.on(TtsStreamingEvents.ConnectFailure, this._onTtsStreamingConnectFailure.bind(this));
+      this.ttsStreamingBuffer.on(TtsStreamingEvents.Connected, this._onTtsStreamingConnected.bind(this));
     }
     else {
       this.logger.info(`CallSession:exec - not a normal call session: ${this.constructor.name}`);
@@ -1248,7 +1348,7 @@ class CallSession extends Emitter {
       }
 
       if (0 === this.tasks.length &&
-        this.requestor instanceof WsRequestor &&
+        this.appIsUsingWebsockets &&
         !this.requestor.closedGracefully &&
         !this.callGone &&
         !this.isConfirmCallSession
@@ -1358,7 +1458,11 @@ class CallSession extends Emitter {
       }
       else {
         if (this.req && !this.dlg) {
-          this.req.cancel();
+          try {
+            this.req.cancel();
+          } catch (err) {
+            this.logger.error({err}, 'CallSession:_lccCallStatus error cancelling request');
+          }
           this._callReleased();
         }
       }
@@ -1421,6 +1525,14 @@ class CallSession extends Emitter {
         method: 'POST'
       };
     }
+
+    /* if given a relative url then send to same base url as current session */
+    if (this.requestor._isRelativeUrl(opts.call_hook?.url)) {
+      opts.call_hook.url = `${this.requestor.baseUrl}${opts.call_hook?.url}`;
+    }
+    if (this.requestor._isRelativeUrl(opts.call_status_hook?.url)) {
+      opts.call_status_hook.url = `${this.requestor.baseUrl}${opts.call_status_hook?.url}`;
+    }
   }
 
   /**
@@ -1648,8 +1760,8 @@ Duration=${duration} `
   async _lccWhisper(opts, callSid) {
     const {whisper} = opts;
     let tasks;
-    const b3 = this.b3;
-    const httpHeaders = b3 && {b3};
+    //const b3 = this.b3;
+    //const httpHeaders = b3 && {b3};
 
     // this whole thing requires us to be in a Dial verb
     const task = this.currentTask;
@@ -1658,12 +1770,15 @@ Duration=${duration} `
     }
 
     // allow user to provide a url object, a url string, an array of tasks, or a single task
-    if (typeof whisper === 'string' || (typeof whisper === 'object' && whisper.url)) {
-      // retrieve a url
-      const json = await this.requestor(opts.call_hook, this.callInfo.toJSON(), httpHeaders);
-      tasks = normalizeJambones(this.logger, json).map((tdata) => makeTask(this.logger, tdata));
-    }
-    else if (Array.isArray(whisper)) {
+    // Disable passing a URL as not functional Sam Machin - 17/07/2025
+    //if (typeof whisper === 'string' || (typeof whisper === 'object' && whisper.url && !whisper.verb)) {
+    //  // retrieve a url
+    //  const json = await this.requestor(opts.call_hook, this.callInfo.toJSON(), httpHeaders);
+    //  tasks = normalizeJambones(this.logger, json).map((tdata) => makeTask(this.logger, tdata));
+    //}
+    //else if (Array.isArray(whisper)) {
+
+    if (Array.isArray(whisper)) {
       // an inline array of tasks
       tasks = normalizeJambones(this.logger, whisper).map((tdata) => makeTask(this.logger, tdata));
     }
@@ -1794,7 +1909,7 @@ Duration=${duration} `
       return;
     }
     else if (tokens === undefined) {
-      this.logger.info({opts}, 'CallSession:_lccTtsTokens - invalid command since id is missing');
+      this.logger.info({opts}, 'CallSession:_lccTtsTokens - invalid command since tokens is missing');
       return this.requestor.request('tts:tokens-result', '/tokens-result', {
         id,
         status: 'failed',
@@ -1900,6 +2015,17 @@ Duration=${duration} `
       }
     } catch (err) {
       this.logger.info({err, opts, callSid}, 'CallSession:updateCall - error updating call');
+      const {writeAlerts} = this.srf.locals;
+      try {
+        writeAlerts({
+          alert_type: 'error-updating-call',
+          account_sid: this.accountSid,
+          message: err.message,
+          target_sid: callSid
+        });
+      } catch (err) {
+        this.logger.error({err}, 'Error writing error-updating-call alert');
+      }
     }
   }
 
@@ -2301,23 +2427,22 @@ Duration=${duration} `
 
     // need to allocate an endpoint
     try {
-      if (!this.ms) this.ms = this.getMS();
-      const ep = await this.ms.createEndpoint({
+      const ep = await this._createMediaEndpoint({
         headers: {
           'X-Jambones-Call-ID': this.callId,
+          'X-Call-Sid': this.callSid,
         },
         remoteSdp: this.req.body
       });
       //ep.cs = this;
       this.ep = ep;
       this.logger.info(`allocated endpoint ${ep.uuid}`);
-      this._configMsEndpoint();
 
       this.ep.on('destroy', () => {
         this.logger.debug(`endpoint was destroyed!! ${this.ep.uuid}`);
       });
 
-      if (this.direction === CallDirection.Inbound) {
+      if (this.direction === CallDirection.Inbound || this.application?.transferredCall) {
         if (task.earlyMedia && !this.req.finalResponseSent) {
           this.res.send(183, {body: ep.local.sdp});
           return {ep};
@@ -2382,9 +2507,6 @@ Duration=${duration} `
       this.logger.error('CallSession:replaceEndpoint cannot be called without stable dlg');
       return;
     }
-    // When this call kicked out from conference, session need to replace endpoint
-    // but this.ms might be undefined/null at this case.
-    this.ms = this.ms || this.getMS();
     // Destroy previous ep if it's still running.
     if (this.ep?.connected) this.ep.destroy();
 
@@ -2409,8 +2531,7 @@ Duration=${duration} `
      * This prevents call failures during media renegotiation.
      */
 
-    this.ep = await this.ms.createEndpoint();
-    this._configMsEndpoint();
+    this.ep = await this._createMediaEndpoint();
 
     const sdp = await this.dlg.modify(this.ep.local.sdp);
     await this.ep.modify(sdp);
@@ -2453,7 +2574,9 @@ Duration=${duration} `
     this.backgroundTaskManager.stopAll();
     this.clearOrRestoreActionHookDelayProcessor().catch((err) => {});
 
-    this.ttsStreamingBuffer?.stop();
+    this.stopTtsStream();
+
+    this.sttLatencyCalculator?.stop();
   }
 
   /**
@@ -2605,7 +2728,7 @@ Duration=${duration} `
    */
   _onRefer(req, res) {
     const task = this.currentTask;
-    const sd = task.sd;
+    const sd = task?.sd;
     if (task && TaskName.Dial === task.name && sd && task.referHook) {
       task.handleRefer(this, req, res);
     }
@@ -2670,15 +2793,8 @@ Duration=${duration} `
   async createOrRetrieveEpAndMs() {
     if (this.ms && this.ep) return {ms: this.ms, ep: this.ep};
 
-    // get a media server
-    if (!this.ms) {
-      const ms = this.srf.locals.getFreeswitch();
-      if (!ms) throw new Error('no available freeswitch');
-      this.ms = ms;
-    }
     if (!this.ep) {
-      this.ep = await this.ms.createEndpoint({remoteSdp: this.req.body});
-      this._configMsEndpoint();
+      this.ep = await this._createMediaEndpoint({remoteSdp: this.req.body});
     }
     return {ms: this.ms, ep: this.ep};
   }
@@ -2831,8 +2947,7 @@ Duration=${duration} `
   async reAnchorMedia(currentMediaRoute = MediaPath.PartialMedia) {
     assert(this.dlg && this.dlg.connected && !this.ep);
 
-    this.ep = await this.ms.createEndpoint({remoteSdp: this.dlg.remote.sdp});
-    this._configMsEndpoint();
+    this.ep = await this._createMediaEndpoint({remoteSdp: this.dlg.remote.sdp});
     await this.dlg.modify(this.ep.local.sdp, {
       headers: {
         'X-Reason': 'anchor-media'
@@ -2907,72 +3022,26 @@ Duration=${duration} `
     }
   }
 
-  _configMsEndpoint() {
-    this._enableInbandDtmfIfRequired(this.ep);
-    this.ep.once('destroy', this._handleMediaTimeout.bind(this));
-    const opts = {
-      ...(this.onHoldMusic && {holdMusic: `shout://${this.onHoldMusic.replace(/^https?:\/\//, '')}`}),
-      ...(JAMBONES_USE_FREESWITCH_TIMER_FD && {timer_name: 'timerfd'}),
-      ...(JAMBONES_MEDIA_TIMEOUT_MS && {media_timeout: JAMBONES_MEDIA_TIMEOUT_MS}),
-      ...(JAMBONES_MEDIA_HOLD_TIMEOUT_MS && {media_hold_timeout: JAMBONES_MEDIA_HOLD_TIMEOUT_MS})
-    };
-    if (Object.keys(opts).length > 0) {
-      this.ep.set(opts);
-    }
-
-    const origDestroy = this.ep.destroy.bind(this.ep);
-    this.ep.destroy = async() => {
-      try {
-        if (this.currentTask?.name === TaskName.Transcribe && JAMBONES_TRANSCRIBE_EP_DESTROY_DELAY_MS) {
-          // transcribe task is being used, wait for some time before destroy
-          // if final transcription is received but endpoint is already closed,
-          // freeswitch module will not be able to send the transcription
-
-          this.logger.debug('callSession:_configMsEndpoint -' +
-            ' transcribe task, wait for some time before destroy');
-          await sleepFor(JAMBONES_TRANSCRIBE_EP_DESTROY_DELAY_MS);
-        }
-        await origDestroy();
-      } catch (err) {
-        this.logger.error(err, 'callSession:_configMsEndpoint - error destroying endpoint');
-      }
-    };
-  }
-
   async _handleMediaTimeout(evt) {
-    if (evt.reason === 'MEDIA_TIMEOUT' && !this.callGone) {
+    if (evt?.reason === 'MEDIA_TIMEOUT' && !this.callGone) {
       this.logger.info('CallSession:_handleMediaTimeout: received MEDIA_TIMEOUT, hangup the call');
       this._jambonzHangup('Media Timeout');
     }
   }
 
-  async _enableInbandDtmfIfRequired(ep) {
-    if (ep.inbandDtmfEnabled) return;
-    // only enable inband dtmf detection if voip carrier dtmf_type === tones
-    if (this.inbandDtmfEnabled) {
-      // https://developer.signalwire.com/freeswitch/FreeSWITCH-Explained/Modules/mod-dptools/6587132/#0-about
-      try {
-        ep.execute('start_dtmf');
-        ep.inbandDtmfEnabled = true;
-      } catch (err) {
-        this.logger.info(err, 'CallSession:_enableInbandDtmf - error enable inband DTMF');
-      }
-    }
-  }
-
   /**
    * notifyTaskError - only used when websocket connection is used instead of webhooks
    */
 
   _notifyTaskError(obj) {
-    if (this.requestor instanceof WsRequestor) {
+    if (this.appIsUsingWebsockets) {
       this.requestor.request('jambonz:error', '/error', obj)
         .catch((err) => this.logger.debug({err}, 'CallSession:_notifyTaskError - Error sending'));
     }
   }
 
   _notifyTaskStatus(task, evt) {
-    if (this.notifyEvents && this.requestor instanceof WsRequestor) {
+    if (this.notifyEvents && this.appIsUsingWebsockets) {
       const obj = {...evt, id: task.id, name: task.name};
       this.requestor.request('verb:status', '/status', obj)
         .catch((err) => this.logger.debug({err}, 'CallSession:_notifyTaskStatus - Error sending'));
@@ -3011,8 +3080,20 @@ Duration=${duration} `
     });
   }
 
+  async _enableInbandDtmfIfRequired(ep) {
+    if (ep.inbandDtmfEnabled) return;
+    // only enable inband dtmf detection if voip carrier dtmf_type === tones
+    if (this.inbandDtmfEnabled) {
+      // https://developer.signalwire.com/freeswitch/FreeSWITCH-Explained/Modules/mod-dptools/6587132/#0-about
+      ep.execute('start_dtmf').catch((err) => {
+        this.logger.info({err}, 'CallSession:_enableInbandDtmfIfRequired - error starting DTMF');
+      });
+      ep.inbandDtmfEnabled = true;
+    }
+  }
+
   _clearTasks(backgroundGather, evt) {
-    if (this.requestor instanceof WsRequestor && !backgroundGather.cleared) {
+    if (this.appIsUsingWebsockets && !backgroundGather.cleared) {
       this.logger.debug({evt}, 'CallSession:_clearTasks on event from background gather');
       try {
         backgroundGather.cleared = true;
@@ -3040,10 +3121,20 @@ Duration=${duration} `
     }
   }
 
+  _onTtsStreamingConnected() {
+    this.requestor?.request('tts:streaming-event', '/streaming-event', {event_type: 'stream_open'})
+      .catch((err) => this.logger.info({err}, 'CallSession:_onTtsStreamingConnected - Error sending'));
+  }
+
   _onTtsStreamingEmpty() {
     const task = this.currentTask;
     if (task && TaskName.Say === task.name) {
       task.notifyTtsStreamIsEmpty();
+    } else if (
+      // If Gather nested say task is streaming
+      task && TaskName.Gather === task.name && task.sayTask && task.sayTask.isStreamingTts) {
+      const sayTask = task.sayTask;
+      sayTask.notifyTtsStreamIsEmpty();
     }
   }
 
@@ -3118,6 +3209,16 @@ Duration=${duration} `
     }
   }
 
+  async _createMediaEndpoint(drachtioFsmrfOptions = {}) {
+    return await createMediaEndpoint(this.srf, this.logger, {
+      activeMs: this.getMS(),
+      drachtioFsmrfOptions,
+      onHoldMusic: this.onHoldMusic,
+      inbandDtmfEnabled: this.inbandDtmfEnabled,
+      mediaTimeoutHandler: this._handleMediaTimeout.bind(this),
+    });
+  }
+
   getFormattedConversation(numTurns) {
     const turns = this.conversationTurns.slice(-numTurns);
     if (turns.length === 0) return null;
@@ -3128,6 +3229,20 @@ Duration=${duration} `
       return `assistant: ${t.text}`;
     }).join('\n');
   }
+
+  startSttLatencyVad() {
+    if (this.notifySttLatencyEnabled) {
+      this.sttLatencyCalculator.start();
+    }
+  }
+
+  stopSttLatencyVad() {
+    this.sttLatencyCalculator.stop();
+  }
+
+  calculateSttLatency() {
+    return this.sttLatencyCalculator.calculateLatency();
+  }
 }
 
 module.exports = CallSession;

lib/session/confirm-call-session.js

@@ -8,7 +8,8 @@ const CallSession = require('./call-session');
 
  */
 class ConfirmCallSession extends CallSession {
-  constructor({logger, application, dlg, ep, tasks, callInfo, accountInfo, memberId, confName, rootSpan, req}) {
+  // eslint-disable-next-line max-len
+  constructor({logger, application, dlg, ep, tasks, callInfo, accountInfo, memberId, confName, rootSpan, req, tmpFiles}) {
     super({
       logger,
       application,
@@ -24,6 +25,7 @@ class ConfirmCallSession extends CallSession {
     this.dlg = dlg;
     this.ep = ep;
     this.req = req;
+    this.tmpFiles = tmpFiles;
   }
 
   /**

lib/session/inbound-call-session.js

@@ -22,6 +22,12 @@ class InboundCallSession extends CallSession {
     this.req = req;
     this.res = res;
 
+    // if the call was canceled before we got here, handle it
+    if (this.req.locals.canceled) {
+      req.locals.logger.info('InboundCallSession: constructor - call was already canceled');
+      this._onCancel();
+    }
+
     req.once('cancel', this._onCancel.bind(this));
 
     this.on('callStatusChange', this._notifyCallStatusChange.bind(this));

lib/session/rest-call-session.js

@@ -27,10 +27,13 @@ class RestCallSession extends CallSession {
     }
 
     this.on('callStatusChange', this._notifyCallStatusChange.bind(this));
-    this._notifyCallStatusChange({
-      callStatus: CallStatus.Trying,
-      sipStatus: 100,
-      sipReason: 'Trying'
+
+    setImmediate(() => {
+      this._notifyCallStatusChange({
+        callStatus: CallStatus.Trying,
+        sipStatus: 100,
+        sipReason: 'Trying'
+      });
     });
   }
 

lib/session/siprec-call-session.js

@@ -45,12 +45,11 @@ class SipRecCallSession extends InboundCallSession {
 
   async answerSipRecCall() {
     try {
-      this.ms = this.getMS();
       let remoteSdp = this.sdp1.replace(/sendonly/, 'sendrecv');
-      this.ep = await this.ms.createEndpoint({remoteSdp});
+      this.ep = await this._createMediaEndpoint({remoteSdp});
       //this.logger.debug({remoteSdp, localSdp: this.ep.local.sdp}, 'SipRecCallSession - allocated first endpoint');
       remoteSdp = this.sdp2.replace(/sendonly/, 'sendrecv');
-      this.ep2 = await this.ms.createEndpoint({remoteSdp});
+      this.ep2 = await this._createMediaEndpoint({remoteSdp});
       //this.logger.debug({remoteSdp, localSdp: this.ep2.local.sdp}, 'SipRecCallSession - allocated second endpoint');
       await this.ep.bridge(this.ep2);
       const combinedSdp = await createSipRecPayload(this.ep.local.sdp, this.ep2.local.sdp, this.logger);

lib/tasks/alert.js

@@ -0,0 +1,31 @@
+const Task = require('./task');
+const {TaskName} = require('../utils/constants');
+
+class TaskAlert extends Task {
+  constructor(logger, opts, parentTask) {
+    super(logger, opts);
+    this.message = this.data.message;
+  }
+
+  get name() { return TaskName.Alert; }
+
+  async exec(cs) {
+    const {srf, accountSid:account_sid, callSid:target_sid, applicationSid:application_sid} = cs;
+    const {writeAlerts, AlertType} = srf.locals;
+    await super.exec(cs);
+    writeAlerts({
+      account_sid,
+      alert_type: AlertType.APPLICATION,
+      detail: `Application SID ${application_sid}`,
+      message: this.message,
+      target_sid
+    }).catch((err) => this.logger.info({err}, 'Error generating alert application'));
+  }
+
+  async kill(cs) {
+    super.kill(cs);
+    this.notifyTaskDone();
+  }
+}
+
+module.exports = TaskAlert;

lib/tasks/conference.js

@@ -49,7 +49,8 @@ class Conference extends Task {
     this.confName = this.data.name;
     [
       'beep', 'startConferenceOnEnter', 'endConferenceOnExit', 'joinMuted',
-      'maxParticipants', 'waitHook', 'statusHook', 'endHook', 'enterHook', 'endConferenceDuration'
+      'maxParticipants', 'waitHook', 'statusHook', 'endHook', 'enterHook',
+      'endConferenceDuration', 'distributeDtmf'
     ].forEach((attr) => this[attr] = this.data[attr]);
     this.record = this.data.record || {};
     this.statusEvents = [];
@@ -356,6 +357,7 @@ class Conference extends Task {
         //https://developer.signalwire.com/freeswitch/FreeSWITCH-Explained/Modules/mod_conference_3965534/
         // mute |	Enter conference muted
         ...((this.joinMuted || this.speakOnlyTo) && {mute: true}),
+        ...(this.distributeDtmf && {'dist-dtmf': true})
       }});
 
       /**
@@ -673,7 +675,8 @@ class Conference extends Task {
         confName: this.confName,
         tasks,
         rootSpan: cs.rootSpan,
-        req: cs.req
+        req: cs.req,
+        tmpFiles: cs.tmpFiles,
       });
       await this._playSession.exec();
       this._playSession = null;

lib/tasks/config.js

@@ -18,12 +18,16 @@ class TaskConfig extends Task {
       'boostAudioSignal',
       'vad',
       'ttsStream',
-      'autoStreamTts'
+      'autoStreamTts',
+      'disableTtsCache'
     ].forEach((k) => this[k] = this.data[k] || {});
 
     if ('notifyEvents' in this.data) {
       this.notifyEvents = !!this.data.notifyEvents;
     }
+    if (this.hasNotifySttLatency) {
+      this.notifySttLatency = !!this.data.notifySttLatency;
+    }
 
     if (this.bargeIn.enable) {
       this.gatherOpts = {
@@ -83,7 +87,9 @@ class TaskConfig extends Task {
   get hasVad() { return Object.keys(this.vad).length; }
   get hasFillerNoise() { return Object.keys(this.fillerNoise).length; }
   get hasReferHook() { return Object.keys(this.data).includes('referHook'); }
+  get hasNotifySttLatency() { return Object.keys(this.data).includes('notifySttLatency'); }
   get hasTtsStream() { return Object.keys(this.ttsStream).length; }
+  get hasDisableTtsCache() { return Object.keys(this.data).includes('disableTtsCache'); }
 
   get summary() {
     const phrase = [];
@@ -112,6 +118,8 @@ class TaskConfig extends Task {
     if (this.hasFillerNoise) phrase.push(`fillerNoise ${this.fillerNoise.enable ? 'on' : 'off'}`);
     if (this.data.amd) phrase.push('enable amd');
     if (this.notifyEvents) phrase.push(`event notification ${this.notifyEvents ? 'on' : 'off'}`);
+    if (this.hasNotifySttLatency) phrase.push(
+      `notifySttLatency ${this.notifySttLatency ? 'on' : 'off'}`);
     if (this.onHoldMusic) phrase.push(`onHoldMusic: ${this.onHoldMusic}`);
     if ('boostAudioSignal' in this.data) phrase.push(`setGain ${this.data.boostAudioSignal}`);
     if (this.hasReferHook) phrase.push('set referHook');
@@ -119,6 +127,7 @@ class TaskConfig extends Task {
       phrase.push(`${this.ttsStream.enable ? 'enable' : 'disable'} ttsStream`);
     }
     if ('autoStreamTts' in this.data) phrase.push(`enable Say.stream value ${this.data.autoStreamTts ? 'on' : 'off'}`);
+    if (this.hasDisableTtsCache) phrase.push(`disableTtsCache ${this.data.disableTtsCache ? 'on' : 'off'}`);
     return `${this.name}{${phrase.join(',')}}`;
   }
 
@@ -130,6 +139,11 @@ class TaskConfig extends Task {
       cs.notifyEvents = !!this.data.notifyEvents;
     }
 
+    if (this.hasNotifySttLatency) {
+      this.logger.debug(`turning notifySttLatency ${this.notifySttLatency ? 'on' : 'off'}`);
+      cs.notifySttLatencyEnabled = this.notifySttLatency;
+    }
+
     if (this.onHoldMusic) {
       cs.onHoldMusic = this.onHoldMusic;
     }
@@ -318,7 +332,10 @@ class TaskConfig extends Task {
         voiceMs: this.vad.voiceMs || 250,
         silenceMs: this.vad.silenceMs || 150,
         strategy: this.vad.strategy || 'one-shot',
-        mode: (this.vad.mode !== undefined && this.vad.mode !== null) ? this.vad.mode : 2
+        mode: (this.vad.mode !== undefined && this.vad.mode !== null) ? this.vad.mode : 2,
+        vendor: this.vad.vendor || 'silero',
+        threshold: this.vad.threshold || 0.5,
+        speechPadMs: this.vad.speechPadMs || 30,
       };
     }
 
@@ -343,6 +360,11 @@ class TaskConfig extends Task {
       this.logger.info('Config: disabling ttsStream');
       cs.disableTtsStream();
     }
+
+    if (this.hasDisableTtsCache) {
+      this.logger.info(`set disableTtsCache = ${this.disableTtsCache}`);
+      cs.disableTtsCache = this.data.disableTtsCache;
+    }
   }
 
   async kill(cs) {

lib/tasks/dial.js

@@ -19,9 +19,9 @@ const parseDecibels = require('../utils/parse-decibels');
 const debug = require('debug')('jambonz:feature-server');
 const {parseUri} = require('drachtio-srf');
 const {ANCHOR_MEDIA_ALWAYS,
-  JAMBONZ_DISABLE_DIAL_PAI_HEADER,
+  JAMBONZ_DIAL_PAI_HEADER,
   JAMBONES_DIAL_SBC_FOR_REGISTERED_USER} = require('../config');
-const { isOnhold, isOpusFirst } = require('../utils/sdp-utils');
+const { isOnhold, isOpusFirst, getLeadingCodec } = require('../utils/sdp-utils');
 const { normalizeJambones } = require('@jambonz/verb-specifications');
 const { selectHostPort } = require('../utils/network');
 const { sleepFor } = require('../utils/helpers');
@@ -109,6 +109,7 @@ class TaskDial extends Task {
     this.tag = this.data.tag;
     this.boostAudioSignal = this.data.boostAudioSignal;
     this._mediaPath = MediaPath.FullMedia;
+    this.forwardPAI = this.data.forwardPAI;
 
     if (this.dtmfHook) {
       const {parentDtmfCollector, childDtmfCollector} = parseDtmfOptions(logger, this.data.dtmfCapture || {});
@@ -157,6 +158,7 @@ class TaskDial extends Task {
 
   get canReleaseMedia() {
     const keepAnchor = this.data.anchorMedia ||
+      this.isTranscoding ||
       this.cs.isBackGroundListen ||
       this.cs.onHoldMusic ||
       ANCHOR_MEDIA_ALWAYS ||
@@ -397,7 +399,11 @@ class TaskDial extends Task {
             ...customHeaders
           }
         }, httpHeaders);
-        if (json && Array.isArray(json)) {
+        res.send(202);
+        this.logger.info('DialTask:handleRefer - sent 202 Accepted');
+
+        const returnedInstructions = !!json && Array.isArray(json);
+        if (returnedInstructions) {
           try {
             const logger = isChild ? this.logger : this.sd.logger;
             const tasks = normalizeJambones(logger, json).map((tdata) => makeTask(this.logger, tdata));
@@ -415,16 +421,21 @@ class TaskDial extends Task {
                 /* need to update the callSid of the child with its own (new) AdultingCallSession */
                 sessionTracker.add(adultingSession.callSid, adultingSession);
               }
+              if (this.ep) this.ep.unbridge();
+
+              /* if we got the REFER on the parent leg, end the dial task after completing the refer */
+              if (!isChild) {
+                logger.info('DialTask:handleRefer - killing dial task after processing REFER on parent leg');
+                cs.currentTask?.kill(cs, KillReason.ReferComplete);
+              }
             }
           } catch (err) {
             this.logger.info(err, 'Dial:handleRefer - error setting new application after receiving REFER');
           }
         }
-        //caller and callee legs are briged together, accept refer with 202 will release callee leg endpoint
-        //that makes freeswitch release endpoint for caller leg.
-        if (this.ep) this.ep.unbridge();
-        res.send(202);
-        this.logger.info('DialTask:handleRefer - sent 202 Accepted');
+        else {
+          this.logger.info('DialTask:handleRefer - no tasks returned from referHook, not setting new application');
+        }
       } catch (err) {
         this.logger.info({err}, 'DialTask:handleRefer - error processing incoming REFER');
         res.send(err.statusCode || 501);
@@ -539,13 +550,14 @@ class TaskDial extends Task {
     let sbcAddress = this.proxy || getSBC();
     const teamsInfo = {};
     let fqdn;
-
+    const forwardPAI = this.forwardPAI ?? JAMBONZ_DIAL_PAI_HEADER; // dial verb overides env var
+    this.logger.debug(forwardPAI, 'forwardPAI value');
     if (!sbcAddress) throw new Error('no SBC found for outbound call');
     this.headers = {
       'X-Account-Sid': cs.accountSid,
       ...(req && req.has('X-CID') && {'X-CID': req.get('X-CID')}),
       ...(direction === 'outbound' && callInfo.sbcCallid && {'X-CID': callInfo.sbcCallid}),
-      ...(!JAMBONZ_DISABLE_DIAL_PAI_HEADER && req && {
+      ...(req && forwardPAI && {
         ...(req.has('P-Asserted-Identity') && {'P-Asserted-Identity': req.get('P-Asserted-Identity')}),
         ...(req.has('Privacy') && {'Privacy': req.get('Privacy')}),
       }),
@@ -564,7 +576,7 @@ class TaskDial extends Task {
       proxy: `sip:${sbcAddress}`,
       callingNumber: this.callerId || fromUri.user,
       ...(this.callerName && {callingName: this.callerName}),
-      opusFirst: isOpusFirst(this.cs.ep.remote.sdp),
+      opusFirst: isOpusFirst(this.cs.ep.local.sdp),
       isVideoCall: this.cs.ep.remote.sdp.includes('m=video')
     };
 
@@ -630,7 +642,9 @@ class TaskDial extends Task {
          * trunk isn't specified,
          * check if number matches any existing numbers
          *  */
-        if (t.type === 'phone' && !t.trunk) {
+        const { lookupLcrByAccount}  = srf.locals.dbHelpers;
+        const lcrs = await lookupLcrByAccount(cs.accountSid);
+        if (t.type === 'phone' && !t.trunk && lcrs.length == 0) {
           const str = this.callerId || req.callingNumber || '';
           const callingNumber = str.startsWith('+') ? str.substring(1) : str;
           const voip_carrier_sid = await lookupCarrierByPhoneNumber(cs.accountSid, callingNumber);
@@ -663,7 +677,8 @@ class TaskDial extends Task {
           rootSpan: cs.rootSpan,
           startSpan: this.startSpan.bind(this),
           dialTask: this,
-          onHoldMusic: this.cs.onHoldMusic
+          onHoldMusic: this.cs.onHoldMusic,
+          tmpFiles: this.cs.tmpFiles,
         });
         this.dials.set(sd.callSid, sd);
 
@@ -758,12 +773,24 @@ class TaskDial extends Task {
   }
 
   async _connectSingleDial(cs, sd) {
+    // start connect with dialed leg, this is the soonest we can identify transcoding
+    if (this.epOther && sd.ep) {
+      const codecA = getLeadingCodec(this.epOther.local.sdp);
+      const codecB = getLeadingCodec(sd.ep.remote.sdp);
+      this.isTranscoding = (codecA !== codecB);
+      if (this.isTranscoding) {
+        this.logger.info(`Dial:_connectSingleDial - transcoding from ${codecA} (A leg) to ${codecB} (B leg)`);
+      }
+    }
     if (!this.bridged && !this.canReleaseMedia) {
       this.logger.debug('Dial:_connectSingleDial bridging endpoints');
       if (this.epOther) {
         this.epOther.api('uuid_break', this.epOther.uuid);
         this.epOther.bridge(sd.ep);
       }
+      else {
+        this.logger.error('Dial:_connectSingleDial - no other endpoint to bridge!');
+      }
       this.bridged = true;
     }
 
@@ -912,7 +939,6 @@ class TaskDial extends Task {
         this.logger.info({err}, 'Dial:_selectSingleDial - Error boosting audio signal');
       }
     }
-
     /* if we can release the media back to the SBC, do so now */
     if (this.canReleaseMedia || this.shouldExitMediaPathEntirely) {
       setTimeout(this._releaseMedia.bind(this, cs, sd, this.shouldExitMediaPathEntirely), 200);
@@ -986,7 +1012,7 @@ class TaskDial extends Task {
       this.epOther = null;
       this._mediaPath = releaseEntirely ? MediaPath.NoMedia : MediaPath.PartialMedia;
       this.logger.info(
-        `Dial:_releaseMedia - successfully released media from freewitch, media path is now ${this._mediaPath}`);
+        `Dial:_releaseMedia - successfully released media from freeswitch, media path is now ${this._mediaPath}`);
     } catch (err) {
       this.logger.info({err}, 'Dial:_releaseMedia error');
     }
@@ -995,7 +1021,7 @@ class TaskDial extends Task {
   async reAnchorMedia(cs, sd) {
     if (cs.ep && sd.ep) return;
 
-    this.logger.info('Dial:reAnchorMedia - re-anchoring media to freewitch');
+    this.logger.info('Dial:reAnchorMedia - re-anchoring media to freeswitch');
     await Promise.all([sd.reAnchorMedia(this._mediaPath), cs.reAnchorMedia(this._mediaPath)]);
     this.epOther = cs.ep;
 
@@ -1003,7 +1029,7 @@ class TaskDial extends Task {
 
     this._mediaPath = MediaPath.FullMedia;
     this.logger.info(
-      `Dial:_releaseMedia - successfully re-anchored media to freewitch, media path is now ${this._mediaPath}`);
+      `Dial:_releaseMedia - successfully re-anchored media to freeswitch, media path is now ${this._mediaPath}`);
   }
 
   // Handle RE-INVITE hold from caller leg.
@@ -1087,7 +1113,8 @@ class TaskDial extends Task {
               accountInfo: this.cs.accountInfo,
               tasks,
               rootSpan: this.cs.rootSpan,
-              req: this.cs.req
+              req: this.cs.req,
+              tmpFiles: this.cs.tmpFiles,
             });
             await this._onHoldSession.exec();
             this._onHoldSession = null;

lib/tasks/dub.js

@@ -83,7 +83,8 @@ class TaskDub extends TtsTask {
       action: 'playOnTrack',
       track: this.track,
       play: this.play,
-      loop: this.loop ? 'loop' : 'once',
+      // drachtio-fsmrf will convert loop from boolean to 'loop' or 'once'
+      loop: this.loop,
       gain: this.gain
     });
   }

lib/tasks/enqueue.js

@@ -370,7 +370,8 @@ class TaskEnqueue extends Task {
         accountInfo: cs.accountInfo,
         tasks: tasksToRun,
         rootSpan: cs.rootSpan,
-        req: cs.req
+        req: cs.req,
+        tmpFiles: cs.tmpFiles,
       });
       await this._playSession.exec();
       this._playSession = null;

lib/tasks/gather.js

@@ -5,13 +5,17 @@ const {
   AwsTranscriptionEvents,
   AzureTranscriptionEvents,
   DeepgramTranscriptionEvents,
+  GladiaTranscriptionEvents,
   SonioxTranscriptionEvents,
   CobaltTranscriptionEvents,
   IbmTranscriptionEvents,
   NvidiaTranscriptionEvents,
   JambonzTranscriptionEvents,
   AssemblyAiTranscriptionEvents,
+  HoundifyTranscriptionEvents,
+  DeepgramfluxTranscriptionEvents,
   VoxistTranscriptionEvents,
+  CartesiaTranscriptionEvents,
   OpenAITranscriptionEvents,
   VadDetection,
   VerbioTranscriptionEvents,
@@ -91,6 +95,8 @@ class TaskGather extends SttTask {
 
   get needsStt() { return this.input.includes('speech'); }
 
+  get isBackgroundGather() { return this.bugname_prefix === 'background_bargeIn_'; }
+
   get wantsSingleUtterance() {
     return this.data.recognizer?.singleUtterance === true;
   }
@@ -138,7 +144,11 @@ class TaskGather extends SttTask {
     try {
       await this.handling(cs, obj);
     } catch (error) {
-      if (error instanceof SpeechCredentialError) {
+      if (
+        // avoid bargein task with sticky will restart forever
+        // throw exception to stop the loop.
+        !this.sticky &&
+        error instanceof SpeechCredentialError) {
         this.logger.info('Gather failed due to SpeechCredentialError, finished!');
         this.notifyTaskDone();
         return;
@@ -221,7 +231,9 @@ class TaskGather extends SttTask {
 
     const startListening = async(cs, ep) => {
       this._startTimer();
-      if (this.isContinuousAsr && 0 === this.timeout) this._startAsrTimer();
+      if (this.isContinuousAsr && 0 === this.timeout && !this.isBackgroundGather) {
+        this._startAsrTimer();
+      }
       if (this.input.includes('speech') && !this.listenDuringPrompt) {
         try {
           await this._setSpeechHandlers(cs, ep);
@@ -246,7 +258,7 @@ class TaskGather extends SttTask {
             startDtmfListener();
           }
           this._stopVad();
-          if (!this.killed) {
+          if (!this.killed && !this.resolved) {
             startListening(cs, ep);
             if (this.input.includes('speech') && this.vendor === 'nuance' && this.listenDuringPrompt) {
               this.logger.debug('Gather:exec - starting transcription timers after say completes');
@@ -258,15 +270,22 @@ class TaskGather extends SttTask {
         };
         this.sayTask.span = span;
         this.sayTask.ctx = ctx;
-        this.sayTask.exec(cs, {ep})  // kicked off, _not_ waiting for it to complete
+        this.sayTask
+          .exec(cs, {ep})  // kicked off, _not_ waiting for it to complete
+          .then(() => {
+            if (this.sayTask.isStreamingTts) return;
+            this.logger.debug('Gather:exec - nested say task completed');
+            span.end();
+            process();
+            return;
+          })
           .catch((err) => {
             process();
           });
-        this.sayTask.on('playDone', (err) => {
-          span.end();
-          if (err) this.logger.error({err}, 'Gather:exec Error playing tts');
+        if (this.sayTask.isStreamingTts && !this.sayTask.closeOnStreamEmpty) {
+          // if streaming tts, we do not wait for it to complete if it is not closing the stream automatically
           process();
-        });
+        }
       }
       else if (this.playTask) {
         const {span, ctx} = this.startChildSpan(`nested:${this.playTask.summary}`);
@@ -277,7 +296,7 @@ class TaskGather extends SttTask {
             startDtmfListener();
           }
           this._stopVad();
-          if (!this.killed) {
+          if (!this.killed && !this.resolved) {
             startListening(cs, ep);
             if (this.input.includes('speech') && this.vendor === 'nuance' && this.listenDuringPrompt) {
               this.logger.debug('Gather:exec - starting transcription timers after play completes');
@@ -289,15 +308,17 @@ class TaskGather extends SttTask {
         };
         this.playTask.span = span;
         this.playTask.ctx = ctx;
-        this.playTask.exec(cs, {ep})  // kicked off, _not_ waiting for it to complete
+        this.playTask
+          .exec(cs, {ep})  // kicked off, _not_ waiting for it to complete
+          .then(() => {
+            this.logger.debug('Gather:exec - nested play task completed');
+            span.end();
+            process();
+            return;
+          })
           .catch((err) => {
             process();
           });
-        this.playTask.on('playDone', (err) => {
-          span.end();
-          if (err) this.logger.error({err}, 'Gather:exec Error playing url');
-          process();
-        });
       }
       else {
         if (this.killed) {
@@ -341,6 +362,7 @@ class TaskGather extends SttTask {
     this.sayTask?.span.end();
     this._stopVad();
     this._resolve('killed');
+    cs.stopSttLatencyVad();
   }
 
   updateTaskInProgress(opts) {
@@ -356,6 +378,9 @@ class TaskGather extends SttTask {
 
   _onDtmf(cs, ep, evt) {
     this.logger.debug(evt, 'TaskGather:_onDtmf');
+    if (!this._timeoutTimer && this.timeout > 0) {
+      this._startTimer();
+    }
     clearTimeout(this.interDigitTimer);
     let resolved = false;
     if (this.dtmfBargein) {
@@ -380,12 +405,8 @@ class TaskGather extends SttTask {
       if (this.digitBuffer.length === 0 && this.needsStt) {
         // DTMF is higher priority than STT.
         this.removeCustomEventListeners();
-        ep.stopTranscription({
-          vendor: this.vendor,
-          bugname: this.bugname,
-        })
-          .catch((err) => this.logger.error({err},
-            ` Received DTMF, Error stopping transcription for vendor ${this.vendor}`));
+        this._clearAsrTimer(); //clear ASR timer as we're now using dtmf
+        this._stopTranscribing(ep);
       }
       this.digitBuffer += evt.dtmf;
       const len = this.digitBuffer.length;
@@ -399,6 +420,7 @@ class TaskGather extends SttTask {
       const ms = this.interDigitTimeout * 1000;
       this.logger.debug(`starting interdigit timer of ${ms}`);
       this.interDigitTimer = setTimeout(() => this._resolve('dtmf-interdigit-timeout'), ms);
+      this._clearTimer(); //clear main timer as we're now using interdigit dtmf timer
     }
   }
 
@@ -456,6 +478,28 @@ class TaskGather extends SttTask {
         this.addCustomEventListener(ep, DeepgramTranscriptionEvents.Connect, this._onVendorConnect.bind(this, cs, ep));
         this.addCustomEventListener(ep, DeepgramTranscriptionEvents.ConnectFailure,
           this._onVendorConnectFailure.bind(this, cs, ep));
+        this.addCustomEventListener(ep, DeepgramTranscriptionEvents.Error, this._onVendorError.bind(this, cs, ep));
+        break;
+
+      case 'deepgramflux':
+        this.bugname = `${this.bugname_prefix}deepgramflux_transcribe`;
+        this.addCustomEventListener(
+          ep, DeepgramfluxTranscriptionEvents.Transcription, this._onTranscription.bind(this, cs, ep));
+        this.addCustomEventListener(
+          ep, DeepgramfluxTranscriptionEvents.Connect, this._onVendorConnect.bind(this, cs, ep));
+        this.addCustomEventListener(ep, DeepgramfluxTranscriptionEvents.ConnectFailure,
+          this._onVendorConnectFailure.bind(this, cs, ep));
+        this.addCustomEventListener(ep, DeepgramfluxTranscriptionEvents.Error, this._onVendorError.bind(this, cs, ep));
+        break;
+
+      case 'gladia':
+        this.bugname = `${this.bugname_prefix}gladia_transcribe`;
+        this.addCustomEventListener(
+          ep, GladiaTranscriptionEvents.Transcription, this._onTranscription.bind(this, cs, ep));
+        this.addCustomEventListener(ep, GladiaTranscriptionEvents.Connect, this._onVendorConnect.bind(this, cs, ep));
+        this.addCustomEventListener(ep, GladiaTranscriptionEvents.ConnectFailure,
+          this._onVendorConnectFailure.bind(this, cs, ep));
+        this.addCustomEventListener(ep, GladiaTranscriptionEvents.Error, this._onVendorError.bind(this, cs, ep));
         break;
 
       case 'soniox':
@@ -535,6 +579,18 @@ class TaskGather extends SttTask {
           this._onVendorConnectFailure.bind(this, cs, ep));
         break;
 
+      case 'houndify':
+        this.bugname = `${this.bugname_prefix}houndify_transcribe`;
+        this.addCustomEventListener(ep, HoundifyTranscriptionEvents.Transcription,
+          this._onTranscription.bind(this, cs, ep));
+        this.addCustomEventListener(ep, HoundifyTranscriptionEvents.Error,
+          this._onVendorError.bind(this, cs, ep));
+        this.addCustomEventListener(ep, HoundifyTranscriptionEvents.ConnectFailure,
+          this._onVendorConnectFailure.bind(this, cs, ep));
+        this.addCustomEventListener(ep, HoundifyTranscriptionEvents.Connect,
+          this._onVendorConnect.bind(this, cs, ep));
+        break;
+
       case 'voxist':
         this.bugname = `${this.bugname_prefix}voxist_transcribe`;
         this.addCustomEventListener(ep, VoxistTranscriptionEvents.Transcription,
@@ -546,6 +602,17 @@ class TaskGather extends SttTask {
           this._onVendorConnectFailure.bind(this, cs, ep));
         break;
 
+      case 'cartesia':
+        this.bugname = `${this.bugname_prefix}cartesia_transcribe`;
+        this.addCustomEventListener(ep, CartesiaTranscriptionEvents.Transcription,
+          this._onTranscription.bind(this, cs, ep));
+        this.addCustomEventListener(
+          ep, CartesiaTranscriptionEvents.Connect, this._onVendorConnect.bind(this, cs, ep));
+        this.addCustomEventListener(ep, CartesiaTranscriptionEvents.Error, this._onVendorError.bind(this, cs, ep));
+        this.addCustomEventListener(ep, CartesiaTranscriptionEvents.ConnectFailure,
+          this._onVendorConnectFailure.bind(this, cs, ep));
+        break;
+
       case 'speechmatics':
         this.bugname = `${this.bugname_prefix}speechmatics_transcribe`;
         this.addCustomEventListener(
@@ -661,14 +728,18 @@ class TaskGather extends SttTask {
         target_sid: this.cs.callSid
       });
     }).catch((err) => this.logger.info({err}, 'Error generating alert for tts failure'));
+
+    // Some vendor use single connection, that we cannot use onConnect event to track transcription start
+    this.cs.emit('transcribe-start');
   }
 
   _startTimer() {
     if (0 === this.timeout) return;
+    this.logger.debug(`Starting timoutTimer of ${this.timeout}ms`);
     this._clearTimer();
     this._timeoutTimer = setTimeout(() => {
+      // If continuousASR in use then extend by the asr window for more transcripts.
       if (this.isContinuousAsr) this._startAsrTimer();
-      if (this.interDigitTimer) return; // let the inter-digit timer complete
       else {
         this._resolve(this.digitBuffer.length >= this.minDigits ? 'dtmf-num-digits' : 'timeout');
       }
@@ -810,17 +881,15 @@ class TaskGather extends SttTask {
         this._fillerNoiseOn = false;  // in a race, if we just started audio it may sneak through here
         this.ep.api('uuid_break', this.ep.uuid)
           .catch((err) => this.logger.info(err, 'Error killing audio'));
-        cs.clearTtsStream();
+        if (cs.isTtsStreamEnabled) cs.clearTtsStream();
       }
       return;
     }
     if (this.sayTask && !this.sayTask.killed) {
-      this.sayTask.removeAllListeners('playDone');
       this.sayTask.kill(cs);
       this.sayTask = null;
     }
     if (this.playTask && !this.playTask.killed) {
-      this.playTask.removeAllListeners('playDone');
       this.playTask.kill(cs);
       this.playTask = null;
     }
@@ -828,6 +897,10 @@ class TaskGather extends SttTask {
   }
 
   _onTranscription(cs, ep, evt, fsEvent) {
+    // check if we are in graceful shutdown mode
+    if (ep.gracefulShutdownResolver) {
+      ep.gracefulShutdownResolver();
+    }
     // make sure this is not a transcript from answering machine detection
     const bugname = fsEvent.getHeader('media-bugname');
     const finished = fsEvent.getHeader('transcription-session-finished');
@@ -840,6 +913,10 @@ class TaskGather extends SttTask {
     if (finished === 'true') return;
 
     if (this.vendor === 'ibm' && evt?.state === 'listening') return;
+
+    // emit an event to the call session to track the time transcription is received
+    cs.emit('on-transcription');
+
     if (this.vendor === 'deepgram' && evt.type === 'UtteranceEnd') {
       /* we will only get this when we have set utterance_end_ms */
       if (this._bufferedTranscripts.length === 0) {
@@ -867,7 +944,7 @@ class TaskGather extends SttTask {
 
     evt = this.normalizeTranscription(evt, this.vendor, 1, this.language,
       this.shortUtterance, this.data.recognizer.punctuation);
-    //this.logger.debug({evt, bugname, finished, vendor: this.vendor}, 'Gather:_onTranscription normalized transcript');
+    this.logger.debug({evt, bugname, finished, vendor: this.vendor}, 'Gather:_onTranscription normalized transcript');
 
     if (evt.alternatives.length === 0) {
       this.logger.info({evt}, 'TaskGather:_onTranscription - got empty transcript, continue listening');
@@ -923,6 +1000,12 @@ class TaskGather extends SttTask {
         }
       }
 
+      // receive a final transcript, calculate the stt latency for this transcript
+      const sttLatency = this.cs.calculateSttLatency();
+      if (!emptyTranscript && sttLatency) {
+        this.stt_latency_ms += `${sttLatency.stt_latency_ms},`;
+      }
+
       if (this.isContinuousAsr) {
         /* append the transcript and start listening again for asrTimeout */
         const t = evt.alternatives[0].transcript;
@@ -1027,6 +1110,11 @@ class TaskGather extends SttTask {
         this.cs.requestor.request('verb:hook', this.partialResultHook,  Object.assign({speech: evt},
           this.cs.callInfo, httpHeaders));
       }
+      else if (this.vendor === 'deepgramflux' &&
+        ['EagerEndOfTurn', 'TurnResumed'].includes(evt.vendor.evt?.event)) {
+        this.logger.debug(`Gather:_onTranscription - deepgramflux event detected: ${evt.event}`);
+        this.performAction({speech: evt, reason: 'speechDetected'}, false);
+      }
       if (this.vendor === 'soniox') {
         if (evt.vendor.finalWords.length) {
           this.logger.debug({evt}, 'TaskGather:_onTranscription - buffering soniox transcript');
@@ -1073,12 +1161,8 @@ class TaskGather extends SttTask {
   }
 
   async _startFallback(cs, ep, evt) {
-    if (this.canFallback) {
-      ep.stopTranscription({
-        vendor: this.vendor,
-        bugname: this.bugname
-      })
-        .catch((err) => this.logger.error({err}, `Error stopping transcription for primary vendor ${this.vendor}`));
+    if (this.canFallback()) {
+      this._stopTranscribing(ep);
       try {
         this.logger.debug('gather:_startFallback');
         this.notifyError({ msg: 'ASR error',
@@ -1207,17 +1291,26 @@ class TaskGather extends SttTask {
     }
   }
 
+  async _stopTranscribing(ep) {
+    // Fix for https://github.com/jambonz/jambonz-feature-server/issues/1281
+    // We should immediately call stop transcription from gather
+    // so that next gather can start transcription immediately
+    ep.stopTranscription({
+      vendor: this.vendor,
+      bugname: this.bugname,
+      gracefulShutdown: false
+    })
+      .catch((err) => {
+        if (this.resolved) return;
+        this.logger.error({err}, 'Error stopping transcription');
+      });
+    this.cs.emit('transcribe-stop');
+  }
+
   async _resolve(reason, evt) {
     this.logger.info({evt}, `TaskGather:resolve with reason ${reason}`);
     if (this.needsStt && this.ep && this.ep.connected) {
-      this.ep.stopTranscription({
-        vendor: this.vendor,
-        bugname: this.bugname
-      })
-        .catch((err) => {
-          if (this.resolved) return;
-          this.logger.error({err}, 'Error stopping transcription');
-        });
+      this._stopTranscribing(this.ep);
     }
     if (this.resolved) {
       this.logger.debug('TaskGather:_resolve - already resolved');
@@ -1225,6 +1318,8 @@ class TaskGather extends SttTask {
     }
 
     this.resolved = true;
+    // gather is resolved, prevent any further transcription events while resolve in progress
+    this.removeCustomEventListeners();
     // If bargin is false and ws application return ack to verb:hook
     // the gather should not play any audio
     this._killAudio(this.cs);
@@ -1236,11 +1331,28 @@ class TaskGather extends SttTask {
     this._clearAsrTimer();
     this._clearFinalAsrTimer();
 
+    let sttLatencyMetrics = {};
+    if (this.needsStt) {
+      const sttLatency = this.cs.calculateSttLatency();
+      if (sttLatency) {
+        this.stt_latency_ms = this.stt_latency_ms.endsWith(',') ?
+          this.stt_latency_ms.slice(0, -1) : this.stt_latency_ms;
+        sttLatencyMetrics = {
+          'stt.latency_ms': this.stt_latency_ms,
+          'stt.talkspurts': JSON.stringify(sttLatency.talkspurts),
+          'stt.start_time': sttLatency.stt_start_time,
+          'stt.stop_time': sttLatency.stt_stop_time,
+          'stt.usage': sttLatency.stt_usage,
+        };
+      }
+    }
+
     this.span.setAttributes({
       channel: 1,
       'stt.label': this.label || 'None',
       'stt.resolve': reason,
-      'stt.result': JSON.stringify(evt)
+      'stt.result': JSON.stringify(evt),
+      ...sttLatencyMetrics
     });
 
     if (this.callSession && this.callSession.callGone) {
@@ -1268,6 +1380,9 @@ class TaskGather extends SttTask {
 
     let returnedVerbs = false;
     try {
+      const latencies = Object.fromEntries(
+        Object.entries(sttLatencyMetrics).map(([key, value]) => [key.replace('stt.', 'stt_'), value])
+      );
       if (reason.startsWith('dtmf')) {
         if (this.parentTask) this.parentTask.emit('dtmf', evt);
         else {
@@ -1281,7 +1396,7 @@ class TaskGather extends SttTask {
         else {
           this.emit('transcription', evt);
           this.logger.debug('TaskGather:_resolve - invoking performAction');
-          returnedVerbs = await this.performAction({speech: evt, reason: 'speechDetected'});
+          returnedVerbs = await this.performAction({speech: evt, reason: 'speechDetected', ...latencies});
           this.logger.debug({returnedVerbs}, 'TaskGather:_resolve - back from performAction');
         }
       }
@@ -1289,20 +1404,20 @@ class TaskGather extends SttTask {
         if (this.parentTask) this.parentTask.emit('timeout', evt);
         else {
           this.emit('timeout', evt);
-          returnedVerbs = await this.performAction({reason: 'timeout'});
+          returnedVerbs = await this.performAction({reason: 'timeout', ...latencies});
         }
       }
       else if (reason.startsWith('stt-error')) {
         if (this.parentTask) this.parentTask.emit('stt-error', evt);
         else {
           this.emit('stt-error', evt);
-          returnedVerbs = await this.performAction({reason: 'error', details: evt.error});
+          returnedVerbs = await this.performAction({reason: 'error', details: evt.error, ...latencies});
         }
       } else if (reason.startsWith('stt-low-confidence')) {
         if (this.parentTask) this.parentTask.emit('stt-low-confidence', evt);
         else {
           this.emit('stt-low-confidence', evt);
-          returnedVerbs = await this.performAction({speech:evt, reason: 'stt-low-confidence'});
+          returnedVerbs = await this.performAction({speech:evt, reason: 'stt-low-confidence', ...latencies});
         }
       }
     } catch (err) {  /*already logged error*/ }

lib/tasks/listen.js

@@ -1,10 +1,21 @@
 const Task = require('./task');
-const {TaskName, TaskPreconditions, ListenEvents, ListenStatus} = require('../utils/constants');
+const {TaskName, TaskPreconditions, ListenEvents, ListenStatus} = require('../utils/constants.json');
 const makeTask = require('./make_task');
 const moment = require('moment');
 const MAX_PLAY_AUDIO_QUEUE_SIZE = 10;
 const DTMF_SPAN_NAME = 'dtmf';
 
+function escapeString(str) {
+  return str
+    .replace(/\\/g, '\\\\')   // Escape backslashes
+    .replace(/"/g, '\\"')     // Escape double quotes
+    .replace(/[\b]/g, '\\b')  // Escape backspace (NOTE: [\b] not \b)
+    .replace(/\f/g, '\\f')    // Escape formfeed
+    .replace(/\n/g, '\\n')    // Escape newlines
+    .replace(/\r/g, '\\r')    // Escape carriage returns
+    .replace(/\t/g, '\\t');   // Escape tabs
+}
+
 class TaskListen extends Task {
   constructor(logger, opts, parentTask) {
     super(logger, opts);
@@ -16,10 +27,21 @@ class TaskListen extends Task {
     this.preconditions = TaskPreconditions.Endpoint;
 
     [
-      'action', 'auth', 'method', 'url', 'finishOnKey', 'maxLength', 'metadata', 'mixType', 'passDtmf', 'playBeep',
+      'action', 'auth', 'method', 'url', 'finishOnKey', 'maxLength', 'mixType', 'passDtmf', 'playBeep',
       'sampleRate', 'timeout', 'transcribe', 'wsAuth', 'disableBidirectionalAudio', 'channel'
     ].forEach((k) => this[k] = this.data[k]);
 
+    //Escape JSON special characters in metadata
+    if (this.data.metadata) {
+      this.metadata = {};
+      for (const key in this.data.metadata) {
+        if (this.data.metadata.hasOwnProperty(key)) {
+          const value = this.data.metadata[key];
+          this.metadata[key] = typeof value === 'string' ? escapeString(value) : value;
+        }
+      }
+    }
+
     this.mixType = this.mixType || 'mono';
     this.sampleRate = this.sampleRate || 8000;
     this.earlyMedia = this.data.earlyMedia === true;
@@ -72,7 +94,7 @@ class TaskListen extends Task {
     } catch (err) {
       this.logger.info(err, `TaskListen:exec - error ${this.url}`);
     }
-    if (this.transcribeTask) this.transcribeTask.kill();
+    if (this.transcribeTask) this.transcribeTask.kill(cs);
     this._removeListeners(ep);
   }
 
@@ -103,9 +125,12 @@ class TaskListen extends Task {
     this.notifyTaskDone();
   }
 
-  async updateListen(status) {
+  async updateListen(status, silence = false) {
     if (!this.killed && this.ep && this.ep.connected) {
-      const args = this._bugname ? [this._bugname] : [];
+      const args = [
+        ...(this._bugname ? [this._bugname] : []),
+        ...(status === ListenStatus.Pause ? ([silence]) : []),
+      ];
       this.logger.info(`TaskListen:updateListen status ${status}`);
       switch (status) {
         case ListenStatus.Pause:

lib/tasks/llm/index.js

@@ -6,7 +6,6 @@ const TaskLlmUltravox_S2S = require('./llms/ultravox_s2s');
 const TaskLlmElevenlabs_S2S = require('./llms/elevenlabs_s2s');
 const TaskLlmGoogle_S2S = require('./llms/google_s2s');
 const LlmMcpService = require('../../utils/llm-mcp');
-const TaskLlmAws_S2S = require('./llms/aws_s2s');
 
 class TaskLlm extends Task {
   constructor(logger, opts) {
@@ -86,10 +85,6 @@ class TaskLlm extends Task {
         llm = new TaskLlmGoogle_S2S(this.logger, this.data, this);
         break;
 
-      case 'aws':
-        llm = new TaskLlmAws_S2S(this.logger, this.data, this);
-        break;
-
       default:
         throw new Error(`Unsupported vendor ${this.vendor} for LLM`);
     }

lib/tasks/llm/llms/aws_s2s.js

@@ -1,437 +0,0 @@
-const Task = require('../../task');
-const crypto = require('crypto');
-const TaskName = 'Llm_Aws_s2s';
-const {LlmEvents_Aws} = require('../../../utils/constants');
-const ClientEvent = 'client.event';
-const SessionDelete = 'session.delete';
-
-const aws_server_events = [
-  'error',
-  'session.created',
-  'session.updated',
-];
-
-const expandWildcards = (events) => {
-  const expandedEvents = [];
-
-  events.forEach((evt) => {
-    if (evt.endsWith('.*')) {
-      const prefix = evt.slice(0, -2); // Remove the wildcard ".*"
-      const matchingEvents = aws_server_events.filter((e) => e.startsWith(prefix));
-      expandedEvents.push(...matchingEvents);
-    } else {
-      expandedEvents.push(evt);
-    }
-  });
-
-  return expandedEvents;
-};
-
-class TaskLlmAws_S2S extends Task {
-  constructor(logger, opts, parentTask) {
-    super(logger, opts, parentTask);
-    this.parent = parentTask;
-
-    this.vendor = this.parent.vendor;
-    this.model = this.parent.model;
-    this.auth = this.parent.auth;
-    this.connectionOptions = this.parent.connectOptions;
-
-    const {access_key_id, secret_access_key, aws_region} = this.auth || {};
-    if (!access_key_id) throw new Error('auth.access_key_id is required for Aws S2S');
-    if (!secret_access_key) throw new Error('auth.secret_access_key is required for Aws S2S');
-    if (!aws_region) throw new Error('auth.aws_region is required for Aws S2S');
-
-    this.access_key_id = access_key_id;
-    this.secret_access_key = secret_access_key;
-    this.aws_region = aws_region;
-    this.actionHook = this.data.actionHook;
-    this.eventHook = this.data.eventHook;
-    this.toolHook = this.data.toolHook;
-    const {inferenceConfiguration, toolConfiguration, voiceId, systemPrompt} = this.data.llmOptions;
-
-    this.inferenceConfiguration = inferenceConfiguration;
-    this.toolConfiguration = toolConfiguration;
-    this.voiceId = voiceId;
-    this.systemPrompt = systemPrompt;
-    this.promptName = crypto.randomUUID();
-
-    this.results = {
-      completionReason: 'normal conversation end'
-    };
-
-    /**
-     * only one of these will have items,
-     * if includeEvents, then these are the events to include
-     * if excludeEvents, then these are the events to exclude
-     */
-    this.includeEvents = [];
-    this.excludeEvents = [];
-
-    /* default to all events if user did not specify */
-    this._populateEvents(this.data.events || aws_server_events);
-
-    this.addCustomEventListener = parentTask.addCustomEventListener.bind(parentTask);
-    this.removeCustomEventListeners = parentTask.removeCustomEventListeners.bind(parentTask);
-  }
-
-  get name() { return TaskName; }
-
-  get host() {
-    const {host} = this.connectionOptions || {};
-    return host || (this.vendor === 'aws' ? 'api.aws.com' : void 0);
-  }
-
-  get path() {
-    const {path} = this.connectionOptions || {};
-    if (path) return path;
-
-    switch (this.vendor) {
-      case 'aws':
-        return `v1/realtime?model=${this.model}`;
-      case 'microsoft':
-        return `aws/realtime?api-version=2024-10-01-preview&deployment=${this.model}`;
-    }
-  }
-
-  async _api(ep, args) {
-    const res = await ep.api('uuid_aws_s2s', `^^|${args.join('|')}`);
-    if (!res.body?.startsWith('+OK')) {
-      throw new Error({args}, `Error calling uuid_aws_s2s: ${res.body}`);
-    }
-  }
-
-  async exec(cs, {ep}) {
-    await super.exec(cs);
-
-    await this._startListening(cs, ep);
-
-    await this.awaitTaskDone();
-
-    /* note: the parent llm verb started the span, which is why this is necessary */
-    await this.parent.performAction(this.results);
-
-    this._unregisterHandlers();
-  }
-
-  async kill(cs) {
-    super.kill(cs);
-
-    this._api(cs.ep, [cs.ep.uuid, SessionDelete])
-      .catch((err) => this.logger.info({err}, 'TaskLlmAws_S2S:kill - error deleting session'));
-
-    this.notifyTaskDone();
-  }
-
-  /**
-   * Send function call output to the Aws server in the form of conversation.item.create
-   * per https://platform.aws.com/docs/guides/realtime/function-calls
-   */
-  async processToolOutput(ep, tool_call_id, data) {
-    try {
-      this.logger.debug({tool_call_id, data}, 'TaskLlmAws_S2S:processToolOutput');
-
-      if (!data.type || data.type !== 'conversation.item.create') {
-        this.logger.info({data},
-          'TaskLlmAws_S2S:processToolOutput - invalid tool output, must be conversation.item.create');
-      }
-      else {
-        await this._api(ep, [ep.uuid, ClientEvent, JSON.stringify(data)]);
-
-        // spec also recommends to send immediate response.create
-        await this._api(ep, [ep.uuid, ClientEvent, JSON.stringify({type: 'response.create'})]);
-      }
-    } catch (err) {
-      this.logger.info({err}, 'TaskLlmAws_S2S:processToolOutput');
-    }
-  }
-
-  /**
-   * Send a session.update to the Aws server
-   * Note: creating and deleting conversation items also supported as well as interrupting the assistant
-   */
-  async processLlmUpdate(ep, data, _callSid) {
-    try {
-      this.logger.debug({data, _callSid}, 'TaskLlmAws_S2S:processLlmUpdate');
-
-      if (!data.type || ![
-        'session.update',
-        'conversation.item.create',
-        'conversation.item.delete',
-        'response.cancel'
-      ].includes(data.type)) {
-        this.logger.info({data}, 'TaskLlmAws_S2S:processLlmUpdate - invalid mid-call request');
-      }
-      else {
-        await this._api(ep, [ep.uuid, ClientEvent, JSON.stringify(data)]);
-      }
-    } catch (err) {
-      this.logger.info({err}, 'TaskLlmAws_S2S:processLlmUpdate');
-    }
-  }
-
-  async _startListening(cs, ep) {
-    this._registerHandlers(ep);
-
-    try {
-      const args = [ep.uuid, 'session.create', this.aws_region, this.model, this.access_key_id, this.secret_access_key];
-      await this._api(ep, args);
-    } catch (err) {
-      this.logger.error({err}, 'TaskLlmAws_S2S:_startListening');
-      this.notifyTaskDone();
-    }
-  }
-
-  async _sendClientEvent(ep, obj) {
-    let ok = true;
-    this.logger.debug({obj}, 'TaskLlmAws_S2S:_sendClientEvent');
-    try {
-      const args = [ep.uuid, ClientEvent, JSON.stringify(obj)];
-      await this._api(ep, args);
-    } catch (err) {
-      ok = false;
-      this.logger.error({err}, 'TaskLlmAws_S2S:_sendClientEvent - Error');
-    }
-    return ok;
-  }
-
-  async _sendInitialMessage(ep) {
-    const settionStart = {
-      event: {
-        sessionStart: {
-          inferenceConfiguration: this.inferenceConfiguration || {
-            maxTokens: 1024,
-            temperature: 0.7,
-            topP: 0.9,
-          },
-        }
-      }
-    };
-
-    if (!await this._sendClientEvent(ep, settionStart)) {
-      return this.notifyTaskDone();
-    }
-
-    const promptStart = {
-      event: {
-        promptStart: {
-          promptName: this.promptName,
-          textOutputConfiguration: {
-            mediaType: 'text/plain',
-          },
-          audioOutputConfiguration: {
-            voiceId: this.voiceId || 'tiffany',
-          },
-          toolUseOutputConfiguration: {
-            mediaType: 'application/json',
-          },
-          ...(this.toolConfiguration && {
-            toolConfiguration: this.toolConfiguration
-          })
-        }
-      }
-    };
-    if (!await this._sendClientEvent(ep, promptStart)) {
-      return this.notifyTaskDone();
-    }
-
-    if (this.systemPrompt) {
-      const contentName = crypto.randomUUID();
-      const systemPromptContentStart = {
-        event: {
-          contentStart: {
-            promptName: this.promptName,
-            contentName,
-            type: 'TEXT',
-            interactive: true,
-            role: 'SYSTEM',
-            textInputConfiguration: {
-              mediaType: 'text/plain',
-            }
-          }
-        }
-      };
-      if (!await this._sendClientEvent(ep, systemPromptContentStart)) {
-        return this.notifyTaskDone();
-      }
-
-      const systemPromptContent = {
-        event: {
-          textInput: {
-            promptName: this.promptName,
-            contentName,
-            content: this.systemPrompt
-          }
-        }
-      };
-
-      if (!await this._sendClientEvent(ep, systemPromptContent)) {
-        return this.notifyTaskDone();
-      }
-
-      const systemPromptContentEnd = {
-        event: {
-          contentEnd: {
-            promptName: this.promptName,
-            contentName,
-          }
-        }
-      };
-      if (!await this._sendClientEvent(ep, systemPromptContentEnd)) {
-        return this.notifyTaskDone();
-      }
-    }
-
-    const audioContentStart = {
-      event: {
-        contentStart: {
-          promptName: this.promptName,
-          contentName: crypto.randomUUID(),
-          type: 'AUDIO',
-          interactive: true,
-          role: 'USER',
-        }
-      }
-    };
-    if (!await this._sendClientEvent(ep, audioContentStart)) {
-      return this.notifyTaskDone();
-    }
-
-  }
-
-  _registerHandlers(ep) {
-    this.addCustomEventListener(ep, LlmEvents_Aws.Connect, this._onConnect.bind(this, ep));
-    this.addCustomEventListener(ep, LlmEvents_Aws.ConnectFailure, this._onConnectFailure.bind(this, ep));
-    this.addCustomEventListener(ep, LlmEvents_Aws.Disconnect, this._onDisconnect.bind(this, ep));
-    this.addCustomEventListener(ep, LlmEvents_Aws.ServerEvent, this._onServerEvent.bind(this, ep));
-  }
-
-  _unregisterHandlers() {
-    this.removeCustomEventListeners();
-  }
-
-  _onError(ep, evt) {
-    this.logger.info({evt}, 'TaskLlmAws_S2S:_onError');
-    this.notifyTaskDone();
-  }
-
-  _onConnect(ep) {
-    this.logger.debug('TaskLlmAws_S2S:_onConnect');
-    this._sendInitialMessage(ep);
-  }
-  _onConnectFailure(_ep, evt) {
-    this.logger.info(evt, 'TaskLlmAws_S2S:_onConnectFailure');
-    this.results = {completionReason: 'connection failure'};
-    this.notifyTaskDone();
-  }
-  _onDisconnect(_ep, evt) {
-    this.logger.info(evt, 'TaskLlmAws_S2S:_onConnectFailure');
-    this.results = {completionReason: 'disconnect from remote end'};
-    this.notifyTaskDone();
-  }
-  async _onServerEvent(ep, evt) {
-    let endConversation = false;
-    const type = evt.type;
-    this.logger.info({evt}, 'TaskLlmAws_S2S:_onServerEvent');
-
-    /* check for failures, such as rate limit exceeded, that should terminate the conversation */
-    if (type === 'response.done' && evt.response.status === 'failed') {
-      endConversation = true;
-      this.results = {
-        completionReason: 'server failure',
-        error: evt.response.status_details?.error
-      };
-    }
-
-    /* server errors of some sort */
-    else if (type === 'error') {
-      endConversation = true;
-      this.results = {
-        completionReason: 'server error',
-        error: evt.error
-      };
-    }
-
-    /* tool calls */
-    else if (type === 'response.output_item.done' && evt.item?.type === 'function_call') {
-      this.logger.debug({evt}, 'TaskLlmAws_S2S:_onServerEvent - function_call');
-      const {name, call_id} = evt.item;
-      const args = JSON.parse(evt.item.arguments);
-
-      const mcpTools = this.parent.isMcpEnabled ? await this.parent.mcpService.getAvailableMcpTools() : [];
-      if (mcpTools.some((tool) => tool.name === name)) {
-        this.logger.debug({call_id, name, args}, 'TaskLlmAws_S2S:_onServerEvent - calling mcp tool');
-        try {
-          const res = await this.parent.mcpService.callMcpTool(name, args);
-          this.logger.debug({res}, 'TaskLlmAws_S2S:_onServerEvent - function_call - mcp result');
-          this.processToolOutput(ep, call_id, {
-            type: 'conversation.item.create',
-            item: {
-              type: 'function_call_output',
-              call_id,
-              output: res.content[0]?.text || 'There is no output from the function call',
-            }
-          });
-          return;
-        } catch (err) {
-          this.logger.info({err, evt}, 'TaskLlmAws_S2S - error calling function');
-          this.results = {
-            completionReason: 'client error calling mcp function',
-            error: err
-          };
-          endConversation = true;
-        }
-      }
-      else if (!this.toolHook) {
-        this.logger.warn({evt}, 'TaskLlmAws_S2S:_onServerEvent - no toolHook defined!');
-      }
-      else {
-        try {
-          await this.parent.sendToolHook(call_id, {name, args});
-        } catch (err) {
-          this.logger.info({err, evt}, 'TaskLlmAws - error calling function');
-          this.results = {
-            completionReason: 'client error calling function',
-            error: err
-          };
-          endConversation = true;
-        }
-      }
-    }
-
-    /* check whether we should notify on this event */
-    if (this.includeEvents.length > 0 ? this.includeEvents.includes(type) : !this.excludeEvents.includes(type)) {
-      this.parent.sendEventHook(evt)
-        .catch((err) => this.logger.info({err}, 'TaskLlmAws_S2S:_onServerEvent - error sending event hook'));
-    }
-
-    if (endConversation) {
-      this.logger.info({results: this.results}, 'TaskLlmAws_S2S:_onServerEvent - ending conversation due to error');
-      this.notifyTaskDone();
-    }
-  }
-
-  _populateEvents(events) {
-    if (events.includes('all')) {
-      /* work by excluding specific events */
-      const exclude = events
-        .filter((evt) => evt.startsWith('-'))
-        .map((evt) => evt.slice(1));
-      if (exclude.length === 0) this.includeEvents = aws_server_events;
-      else this.excludeEvents = expandWildcards(exclude);
-    }
-    else {
-      /* work by including specific events */
-      const include = events
-        .filter((evt) => !evt.startsWith('-'));
-      this.includeEvents = expandWildcards(include);
-    }
-
-    this.logger.debug({
-      includeEvents: this.includeEvents,
-      excludeEvents: this.excludeEvents
-    }, 'TaskLlmAws_S2S:_populateEvents');
-  }
-}
-
-module.exports = TaskLlmAws_S2S;

lib/tasks/llm/llms/ultravox_s2s.js

@@ -32,12 +32,14 @@ class TaskLlmUltravox_S2S extends Task {
     this.auth = this.parent.auth;
     this.connectionOptions = this.parent.connectOptions;
 
-    const {apiKey} = this.auth || {};
+    const {apiKey, agent_id} = this.auth || {};
     if (!apiKey) throw new Error('auth.apiKey is required for Vendor: Ultravox');
     this.apiKey = apiKey;
+    this.agentId = agent_id;
     this.actionHook = this.data.actionHook;
     this.eventHook = this.data.eventHook;
     this.toolHook = this.data.toolHook;
+    this.llmOptions = this.data.llmOptions || {};
 
     this.results = {
       completionReason: 'normal conversation end'
@@ -105,24 +107,29 @@ class TaskLlmUltravox_S2S extends Task {
       }
       );
       // merge with any existing tools
-      this.data.llmOptions.selectedTools = [
+      this.llmOptions.selectedTools = [
         ...convertedTools,
-        ...(this.data.llmOptions.selectedTools || [])
+        ...(this.llmOptions.selectedTools || [])
       ];
     }
 
     const payload = {
-      ...this.data.llmOptions,
-      model: this.model,
+      ...this.llmOptions,
+      ...(!this.agentId && {
+        model: this.model,
+      }),
       medium: {
-        ...(this.data.llmOptions.medium || {}),
+        ...(this.llmOptions.medium || {}),
         serverWebSocket: {
           inputSampleRate: 8000,
           outputSampleRate: 8000,
         }
       }
     };
-    const {statusCode, body} = await request('https://api.ultravox.ai/api/calls', {
+    const baseUrl = 'https://api.ultravox.ai';
+    const url = this.agentId ?
+      `${baseUrl}/api/agents/${this.agentId}/calls` : `${baseUrl}/api/calls`;
+    const {statusCode, body} = await request(url, {
       method: 'POST',
       headers: {
         'Content-Type': 'application/json',
@@ -139,8 +146,9 @@ class TaskLlmUltravox_S2S extends Task {
     return data;
   }
 
-  _unregisterHandlers() {
+  _unregisterHandlers(ep) {
     this.removeCustomEventListeners();
+    ep.removeAllListeners('dtmf');
   }
 
   _registerHandlers(ep) {
@@ -148,6 +156,7 @@ class TaskLlmUltravox_S2S extends Task {
     this.addCustomEventListener(ep, LlmEvents_Ultravox.ConnectFailure, this._onConnectFailure.bind(this, ep));
     this.addCustomEventListener(ep, LlmEvents_Ultravox.Disconnect, this._onDisconnect.bind(this, ep));
     this.addCustomEventListener(ep, LlmEvents_Ultravox.ServerEvent, this._onServerEvent.bind(this, ep));
+    ep.on('dtmf', this._onDtmf.bind(this, ep));
   }
 
   async _startListening(cs, ep) {
@@ -182,7 +191,7 @@ class TaskLlmUltravox_S2S extends Task {
     /* note: the parent llm verb started the span, which is why this is necessary */
     await this.parent.performAction(this.results);
 
-    this._unregisterHandlers();
+    this._unregisterHandlers(ep);
   }
 
   async kill(cs) {
@@ -211,7 +220,7 @@ class TaskLlmUltravox_S2S extends Task {
   async _onServerEvent(_ep, evt) {
     let endConversation = false;
     const type = evt.type;
-    this.logger.debug({evt}, 'TaskLlmUltravox_S2S:_onServerEvent');
+    //this.logger.debug({evt}, 'TaskLlmUltravox_S2S:_onServerEvent');
 
     /* server errors of some sort */
     if (type === 'error') {
@@ -339,6 +348,18 @@ class TaskLlmUltravox_S2S extends Task {
       excludeEvents: this.excludeEvents
     }, 'TaskLlmUltravox_S2S:_populateEvents');
   }
+
+  _onDtmf(ep, evt) {
+    this.logger.info({evt}, 'TaskLlmUltravox_S2S:_onDtmf - DTMF received');
+    const {dtmf} = evt;
+    const data = {
+      type: 'user_text_message',
+      text: `DTMF received: ${dtmf}`,
+      urgency: 'immediate'
+    };
+    this._api(ep, [ep.uuid, ClientEvent, JSON.stringify(data)])
+      .catch((err) => this.logger.info({err, evt}, 'TaskLlmUltravox_S2S:_onDtmf - Error sending DTMF as text message'));
+  }
 }
 
 module.exports = TaskLlmUltravox_S2S;

lib/tasks/make_task.js

@@ -96,6 +96,9 @@ function makeTask(logger, obj, parent) {
     case TaskName.Tag:
       const TaskTag = require('./tag');
       return new TaskTag(logger, data, parent);
+    case TaskName.Alert:
+      const TaskAlert = require('./alert');
+      return new TaskAlert(logger, data, parent);
   }
 
   // should never reach

lib/tasks/play.js

@@ -6,7 +6,21 @@ class TaskPlay extends Task {
     super(logger, opts);
     this.preconditions = TaskPreconditions.Endpoint;
 
-    this.url = this.data.url;
+    //Cleanup URLs that contain a querystring with a . unless that querystring is the filename
+    // see https://github.com/jambonz/jambonz-feature-server/pull/1293
+    // and https://github.com/jambonz/jambonz-feature-server/issues/1394 for background
+    if (this.data.url.includes('?')) {
+      if (['.mp3', '.wav'].includes(this.data.url.slice(-4))) {
+        this.url = this.data.url;
+      }
+      else {
+        this.url =  this.data.url.split('?')[0] + '?' + this.data.url.split('?')[1].replaceAll('.', '%2E');
+      }
+    }
+    else {
+      this.url = this.data.url;
+    }
+
     this.seekOffset = this.data.seekOffset || -1;
     this.timeoutSecs = this.data.timeoutSecs || -1;
     this.loop = this.data.loop || 1;

lib/tasks/redirect.js

@@ -17,28 +17,33 @@ class TaskRedirect extends Task {
   async exec(cs) {
     await super.exec(cs);
 
-    if (cs.requestor instanceof WsRequestor && cs.application.requestor._isAbsoluteUrl(this.actionHook)) {
-      this.logger.info(`Task:performAction redirecting to ${this.actionHook}, requires new ws connection`);
-      try {
-        this.cs.requestor.close();
-        const requestor = new WsRequestor(this.logger, cs.accountSid, {url: this.actionHook}, this.webhook_secret) ;
-        this.cs.application.requestor = requestor;
-      } catch (err) {
-        this.logger.info(err, `Task:performAction error redirecting to ${this.actionHook}`);
-      }
-    } else if (cs.application.requestor._isAbsoluteUrl(this.actionHook)) {
-      const baseUrl =  this.cs.application.requestor.baseUrl;
-      const newUrl = URL.parse(this.actionHook);
-      const newBaseUrl = newUrl.protocol + '//' + newUrl.host;
-      if (baseUrl != newBaseUrl) {
+    const isAbsoluteUrl = cs.application?.requestor?._isAbsoluteUrl(this.actionHook);
+
+    if (isAbsoluteUrl) {
+      this.logger.info(`TaskRedirect redirecting to new absolute URL ${this.actionHook}, requires new requestor`);
+
+      if (cs.requestor instanceof WsRequestor) {
         try {
-          this.logger.info(`Task:redirect updating base url to ${newBaseUrl}`);
-          const newRequestor = new HttpRequestor(this.logger, cs.accountSid, {url: this.actionHook},
-            cs.accountInfo.account.webhook_secret);
-          this.cs.requestor.removeAllListeners();
-          this.cs.application.requestor = newRequestor;
+          const requestor = new WsRequestor(this.logger, cs.accountSid, {url: this.actionHook},
+            cs.accountInfo.account.webhook_secret) ;
+          cs.requestor.emit('handover', requestor);
         } catch (err) {
-          this.logger.info(err, `Task:redirect error updating base url to ${this.actionHook}`);
+          this.logger.info(err, `TaskRedirect error redirecting to ${this.actionHook}`);
+        }
+      }
+      else {
+        const baseUrl =  this.cs.application.requestor.baseUrl;
+        const newUrl = URL.parse(this.actionHook);
+        const newBaseUrl = newUrl.protocol + '//' + newUrl.host;
+        if (baseUrl != newBaseUrl) {
+          try {
+            this.logger.info(`Task:redirect updating base url to ${newBaseUrl}`);
+            const newRequestor = new HttpRequestor(this.logger, cs.accountSid, {url: this.actionHook},
+              cs.accountInfo.account.webhook_secret);
+            cs.requestor.emit('handover', newRequestor);
+          } catch (err) {
+            this.logger.info(err, `TaskRedirect error updating base url to ${this.actionHook}`);
+          }
         }
       }
     }

lib/tasks/rest_dial.js

@@ -19,6 +19,7 @@ class TaskRestDial extends Task {
     this.timeout = this.data.timeout || 60;
     this.sipRequestWithinDialogHook = this.data.sipRequestWithinDialogHook;
     this.referHook = this.data.referHook;
+    this.recentCallStatus = 0;
 
     this.on('connect', this._onConnect.bind(this));
     this.on('callStatus', this._onCallStatus.bind(this));
@@ -57,7 +58,11 @@ class TaskRestDial extends Task {
     this._clearCallTimer();
     if (this.canCancel) {
       this.canCancel = false;
-      cs?.req?.cancel();
+      try {
+        cs?.req?.cancel();
+      } catch (err) {
+        this.logger.error({err}, 'TaskRestDial: error cancelling call');
+      }
     }
     this.notifyTaskDone();
   }
@@ -118,7 +123,8 @@ class TaskRestDial extends Task {
   }
 
   _onCallStatus(status) {
-    this.logger.debug(`CallStatus: ${status}`);
+    this.logger.debug(`RestDial CallStatus: ${status}`);
+    this.recentCallStatus = status;
     if (status >= 200) {
       this.canCancel = false;
       this._clearCallTimer();
@@ -136,11 +142,16 @@ class TaskRestDial extends Task {
   }
 
   _onCallTimeout() {
-    this.logger.debug('TaskRestDial: timeout expired without answer, killing task');
+    this.logger.debug(`TaskRestDial: timeout expired without answer, last status ${this.recentCallStatus}`);
     this.timer = null;
-    if (this.canCancel) {
+    if (this.canCancel && this.recentCallStatus < 200) {
+      this.logger.debug('TaskRestDial: cancelling call attempt');
       this.canCancel = false;
-      this.cs?.req?.cancel();
+      try {
+        this.cs?.req?.cancel();
+      } catch (err) {
+        this.logger.error({err}, 'TaskRestDial: error cancelling call');
+      }
     }
   }
 

lib/tasks/say.js

@@ -1,14 +1,38 @@
 const assert = require('assert');
 const TtsTask = require('./tts-task');
 const {TaskName, TaskPreconditions} = require('../utils/constants');
+const {JAMBONES_SAY_CHUNK_SIZE} = require('../config');
 const pollySSMLSplit = require('polly-ssml-split');
-const { SpeechCredentialError } = require('../utils/error');
+const { SpeechCredentialError, NonFatalTaskError } = require('../utils/error');
 const { sleepFor } = require('../utils/helpers');
+const { NON_FANTAL_ERRORS } = require('../utils/constants.json');
+
+/**
+ * Discard unmatching responses:
+ * (1) I sent a playback id but get a response with a different playback id
+ * (2) I sent a playback id but get a response with no playback id
+ * (3) I did not send a playback id but get a response with a playback id
+ * (4) I sent a cache file but get a response with a different cache file
+ */
+
+const isMatchingEvent = (logger, filename, playbackId, evt) => {
+
+  if (!!playbackId && !!evt.variable_tts_playback_id && evt.variable_tts_playback_id === playbackId) {
+    //logger.debug({filename, playbackId, evt}, 'Say:isMatchingEvent - playbackId matched');
+    return true;
+  }
+  if (!!filename && !!evt.file && evt.file === filename) {
+    //logger.debug({filename, playbackId, evt}, 'Say:isMatchingEvent - filename matched');
+    return true;
+  }
+  logger.info({filename, playbackId, evt}, 'Say:isMatchingEvent - no match');
+  return false;
+};
 
 const breakLengthyTextIfNeeded = (logger, text) => {
   // As The text can be used for tts streaming, we need to break lengthy text into smaller chunks
 // HIGH_WATER_BUFFER_SIZE defined in tts-streaming-buffer.js
-  const chunkSize = 900;
+  const chunkSize = JAMBONES_SAY_CHUNK_SIZE;
   const isSSML = text.startsWith('<speak>');
   const options = {
     softLimit: 100,
@@ -98,13 +122,11 @@ class TaskSay extends TtsTask {
       }
       if (this.isStreamingTts) await this.handlingStreaming(cs, obj);
       else await this.handling(cs, obj);
-      this.emit('playDone');
     } catch (error) {
       if (error instanceof SpeechCredentialError) {
         // if say failed due to speech credentials, alarm is writtern and error notification is sent
         // finished this say to move to next task.
         this.logger.info({error}, 'Say failed due to SpeechCredentialError, finished!');
-        this.emit('playDone');
         return;
       }
       throw error;
@@ -125,9 +147,6 @@ class TaskSay extends TtsTask {
 
       await cs.startTtsStream();
 
-      cs.requestor?.request('tts:streaming-event', '/streaming-event', {event_type: 'stream_open'})
-        .catch((err) => this.logger.info({err}, 'TaskSay:handlingStreaming - Error sending'));
-
       if (this.text.length !== 0) {
         this.logger.info('TaskSay:handlingStreaming - sending text to TTS stream');
         for (const t of this.text) {
@@ -247,6 +266,7 @@ class TaskSay extends TtsTask {
         throw new SpeechCredentialError(error.message);
       }
     };
+
     let filepath;
     try {
       filepath = await this._synthesizeWithSpecificVendor(cs, ep, {vendor, language, voice, label});
@@ -258,77 +278,145 @@ class TaskSay extends TtsTask {
     while (!this.killed && (this.loop === 'forever' || this.loop--) && ep?.connected) {
       let segment = 0;
       while (!this.killed && segment < filepath.length) {
+        const filename = filepath[segment];
         if (cs.isInConference) {
           const {memberId, confName, confUuid} = cs;
-          await this.playToConfMember(ep, memberId, confName, confUuid, filepath[segment]);
+          await this.playToConfMember(ep, memberId, confName, confUuid, filename);
         }
         else {
-          const isStreaming = filepath[segment].startsWith('say:{');
+          const isStreaming = filename.startsWith('say:{');
           if (isStreaming) {
-            const arr = /^say:\{.*\}\s*(.*)$/.exec(filepath[segment]);
-            if (arr) this.logger.debug(`Say:exec sending streaming tts request: ${arr[1].substring(0, 64)}..`);
+            const arr = /^say:\{.*\}\s*(.*)$/.exec(filename);
+            if (arr) this.logger.debug(`Say:exec sending streaming tts request ${arr[1].substring(0, 64)}..`);
+            else this.logger.debug(`Say:exec sending ${filename.substring(0, 64)}`);
           }
-          else this.logger.debug(`Say:exec sending ${filepath[segment].substring(0, 64)}`);
-          ep.once('playback-start', (evt) => {
-            this.logger.debug({evt}, 'Say got playback-start');
-            if (this.otelSpan) {
-              this._addStreamingTtsAttributes(this.otelSpan, evt, vendor);
-              this.otelSpan.end();
-              this.otelSpan = null;
-              if (evt.variable_tts_cache_filename) {
-                cs.trackTmpFile(evt.variable_tts_cache_filename);
-              }
-            }
-          });
-          ep.once('playback-stop', (evt) => {
-            this.logger.debug({evt}, 'Say got playback-stop');
-            this.notifyStatus({event: 'stop-playback'});
-            this.notifiedPlayBackStop = true;
-            const tts_error = evt.variable_tts_error;
-            let response_code = 200;
-            // Check if any property ends with _response_code
-            for (const [key, value] of Object.entries(evt)) {
-              if (key.endsWith('_response_code')) {
-                response_code = parseInt(value, 10) || 200;
-                break;
-              }
-            }
 
-            if (tts_error) {
-              writeAlerts({
-                account_sid,
-                alert_type: AlertType.TTS_FAILURE,
-                vendor,
-                detail: evt.variable_tts_error,
-                target_sid
-              }).catch((err) => this.logger.info({err}, 'Error generating alert for no tts'));
-            }
-            if (!tts_error && response_code < 300 && evt.variable_tts_cache_filename && !this.killed) {
-              const text = parseTextFromSayString(this.text[segment]);
-              addFileToCache(evt.variable_tts_cache_filename, {
-                account_sid,
-                vendor,
-                language,
-                voice,
-                engine,
-                model: this.model || this.model_id,
-                text,
-                instructions: this.instructions
-              }).catch((err) => this.logger.info({err}, 'Error adding file to cache'));
-            }
+          const onPlaybackStop = (evt) => {
+            try {
+              const playbackId = this.getPlaybackId(segment);
+              const isMatch = isMatchingEvent(this.logger, filename, playbackId, evt);
+              if (!isMatch) {
+                this.logger.info({currentPlaybackId: playbackId, stopPlaybackId: evt.variable_tts_playback_id},
+                  'Say:exec discarding playback-stop for earlier play');
+                ep.once('playback-stop',  this._boundOnPlaybackStop);
 
-            if (this._playResolve) {
-              (tts_error || response_code >= 300) ? this._playReject(new Error(evt.variable_tts_error)) :
-                this._playResolve();
+                return;
+              }
+              this.logger.debug({evt},
+                `Say got playback-stop ${evt.variable_tts_playback_id ? evt.variable_tts_playback_id : ''}`);
+              this.notifyStatus({event: 'stop-playback'});
+              this.notifiedPlayBackStop = true;
+              const tts_error = evt.variable_tts_error;
+              // some tts vendor may not provide response code, so we default to 200
+              let response_code = 200;
+              // Check if any property ends with _response_code
+              for (const [key, value] of Object.entries(evt)) {
+                if (key.endsWith('_response_code')) {
+                  response_code = parseInt(value, 10);
+                  if (isNaN(response_code)) {
+                    this.logger.info(`Say:exec playback-stop - Invalid response code: ${value}`);
+                    response_code = 0;
+                  }
+                  break;
+                }
+              }
+
+              if (tts_error ||
+                // error response codes indicate failure
+                response_code <= 199 || response_code >= 300) {
+                writeAlerts({
+                  account_sid,
+                  alert_type: AlertType.TTS_FAILURE,
+                  vendor,
+                  detail: evt.variable_tts_error || `TTS playback failed with response code ${response_code}`,
+                  target_sid
+                }).catch((err) => this.logger.info({err}, 'Error generating alert for no tts'));
+              }
+              if (
+                !tts_error &&
+                //2xx response codes indicate success
+                199 < response_code && response_code < 300 &&
+                evt.variable_tts_cache_filename &&
+                !this.killed &&
+                // if tts cache is not disabled, add the file to cache
+                !this.disableTtsCache
+              ) {
+                const text = parseTextFromSayString(this.text[segment]);
+                this.logger.debug({text, cacheFile: evt.variable_tts_cache_filename}, 'Say:exec cache tts');
+                addFileToCache(evt.variable_tts_cache_filename, {
+                  account_sid,
+                  vendor,
+                  language,
+                  voice,
+                  engine,
+                  model: this.model || this.model_id,
+                  text,
+                  instructions: this.instructions
+                }).catch((err) => this.logger.info({err}, 'Error adding file to cache'));
+              }
+
+              if (this._playResolve) {
+                (tts_error ||
+                  // error response codes indicate failure
+                  response_code <= 199 || response_code >= 300
+                ) ?
+                  this._playReject(
+                    new Error(evt.variable_tts_error || `TTS playback failed with response code ${response_code}`)
+                  ) : this._playResolve();
+              }
+            } catch (err) {
+              this.logger.info({err}, 'Error handling playback-stop event');
             }
-          });
+          };
+          this._boundOnPlaybackStop = onPlaybackStop.bind(this);
+
+          const onPlaybackStart = (evt) => {
+            try {
+              const playbackId = this.getPlaybackId(segment);
+              const isMatch = isMatchingEvent(this.logger, filename, playbackId, evt);
+              if (!isMatch) {
+                this.logger.info({currentPlaybackId: playbackId, startPlaybackId: evt.variable_tts_playback_id},
+                  'Say:exec playback-start - unmatched playback_id');
+                ep.once('playback-start',  this._boundOnPlaybackStart);
+                return;
+              }
+              ep.once('playback-stop',  this._boundOnPlaybackStop);
+              this.logger.debug({evt},
+                `Say got playback-start ${evt.variable_tts_playback_id ? evt.variable_tts_playback_id : ''}`);
+              if (this.otelSpan) {
+                this._addStreamingTtsAttributes(this.otelSpan, evt, vendor);
+                this.otelSpan.end();
+                this.otelSpan = null;
+                if (evt.variable_tts_cache_filename) {
+                  cs.trackTmpFile(evt.variable_tts_cache_filename);
+                }
+              }
+            } catch (err) {
+              this.logger.info({err}, 'Error handling playback-start event');
+            }
+          };
+          this._boundOnPlaybackStart = onPlaybackStart.bind(this);
+
+          ep.once('playback-start',  this._boundOnPlaybackStart);
+
           // wait for playback-stop event received to confirm if the playback is successful
           this._playPromise = new Promise((resolve, reject) => {
             this._playResolve = resolve;
             this._playReject = reject;
           });
-          const r = await ep.play(filepath[segment]);
-          this.logger.debug({r}, 'Say:exec play result');
+          try {
+            const r = await ep.play(filename);
+            this.logger.debug({r}, 'Say:exec play result');
+            if (r.playbackSeconds == null && r.playbackMilliseconds == null && r.playbackLastOffsetPos == null) {
+              this._playReject(new Error('Playback failed to start'));
+            }
+          } catch (err) {
+            if (NON_FANTAL_ERRORS.includes(err.message)) {
+              throw new NonFatalTaskError(err.message);
+            }
+            throw err;
+          }
+
           try {
             // wait for playback-stop event received to confirm if the playback is successful
             await this._playPromise;
@@ -345,12 +433,12 @@ class TaskSay extends TtsTask {
             this._playResolve = null;
             this._playReject = null;
           }
-          if (filepath[segment].startsWith('say:{')) {
-            const arr = /^say:\{.*\}\s*(.*)$/.exec(filepath[segment]);
+          if (filename.startsWith('say:{')) {
+            const arr = /^say:\{.*\}\s*(.*)$/.exec(filename);
             if (arr) this.logger.debug(`Say:exec complete playing streaming tts request: ${arr[1].substring(0, 64)}..`);
           } else {
             // This log will print spech credentials in say command for tts stream mode
-            this.logger.debug(`Say:exec completed play file ${filepath[segment]}`);
+            this.logger.debug(`Say:exec completed play file ${filename}`);
           }
         }
         segment++;
@@ -366,8 +454,8 @@ class TaskSay extends TtsTask {
         const {memberId, confName} = cs;
         this.killPlayToConfMember(this.ep, memberId, confName);
       } else if (this.isStreamingTts) {
-        this.logger.debug('TaskSay:kill - clearing TTS stream for streaming audio');
-        cs.clearTtsStream();
+        this.logger.debug('TaskSay:kill - stopping TTS stream for streaming audio');
+        cs.stopTtsStream();
       } else {
         if (!this.notifiedPlayBackStop) {
           this.notifyStatus({event: 'stop-playback'});
@@ -397,6 +485,8 @@ class TaskSay extends TtsTask {
           .replace('playht_', 'playht.')
           .replace('cartesia_', 'cartesia.')
           .replace('rimelabs_', 'rimelabs.')
+          .replace('resemble_', 'resemble.')
+          .replace('inworld_', 'inworld.')
           .replace('verbio_', 'verbio.')
           .replace('elevenlabs_', 'elevenlabs.');
         if (spanMapping[newKey]) newKey = spanMapping[newKey];
@@ -461,6 +551,14 @@ const spanMapping = {
   'rimelabs.name_lookup_time_ms': 'name_lookup_ms',
   'rimelabs.connect_time_ms': 'connect_ms',
   'rimelabs.final_response_time_ms': 'final_response_ms',
+  // Resemble
+  'resemble.connect_time_ms': 'connect_ms',
+  'resemble.final_response_time_ms': 'final_response_ms',
+  // inworld
+  'inworld.name_lookup_time_ms': 'name_lookup_ms',
+  'inworld.connect_time_ms': 'connect_ms',
+  'inworld.final_response_time_ms': 'final_response_ms',
+  'inworld.x_envoy_upstream_service_time': 'upstream_service_time',
   // verbio
   'verbio.name_lookup_time_ms': 'name_lookup_ms',
   'verbio.connect_time_ms': 'connect_ms',

lib/tasks/sip_refer.js

@@ -111,7 +111,12 @@ class TaskSipRefer extends Task {
     /* get IP address of the SBC to use as hostname if needed */
     const {host} = parseUri(dlg.remote.uri);
 
-    if (!referTo.startsWith('<') && !referTo.startsWith('sip') && !referTo.startsWith('"')) {
+    if (
+      !referTo.startsWith('<') &&
+      !referTo.startsWith('sip:') &&
+      !referTo.startsWith('"') &&
+      !referTo.startsWith('tel:')
+    ) {
       /* they may have only provided a phone number/user */
       referTo = `sip:${referTo}@${host}`;
     }
@@ -124,7 +129,12 @@ class TaskSipRefer extends Task {
     if (!referredByDisplayName) {
       referredByDisplayName = cs.req?.callingName;
     }
-    if (!referredBy.startsWith('<') && !referredBy.startsWith('sip') && !referredBy.startsWith('"')) {
+    if (
+      !referredBy.startsWith('<') &&
+      !referredBy.startsWith('sip:') &&
+      !referredBy.startsWith('"') &&
+      !referredBy.startsWith('tel:')
+    ) {
       /* they may have only provided a phone number/user */
       referredBy = `${referredByDisplayName ? `"${referredByDisplayName}"` : ''}<sip:${referredBy}@${host}>`;
     }

lib/tasks/stt-task.js

@@ -4,6 +4,7 @@ const crypto = require('crypto');
 const { TaskPreconditions, CobaltTranscriptionEvents } = require('../utils/constants');
 const { SpeechCredentialError } = require('../utils/error');
 const {JAMBONES_AWS_TRANSCRIBE_USE_GRPC} = require('../config');
+const {TaskName} = require('../utils/constants.json');
 
 /**
  * "Please insert turns here: {{turns:4}}"
@@ -84,6 +85,9 @@ class SttTask extends Task {
     /*bug name prefix */
     this.bugname_prefix = '';
 
+    // stt latency calculator
+    this.stt_latency_ms = '';
+
   }
 
   async exec(cs, {ep, ep2}) {
@@ -91,6 +95,12 @@ class SttTask extends Task {
     this.ep = ep;
     this.ep2 = ep2;
 
+    // start vad from stt latency calculator
+    if (this.name !== TaskName.Gather ||
+      this.name === TaskName.Gather && this.needsStt) {
+      cs.startSttLatencyVad();
+    }
+
     // use session preferences if we don't have specific verb-level settings.
     if (cs.recognizer) {
       for (const k in cs.recognizer) {
@@ -161,7 +171,7 @@ class SttTask extends Task {
       try {
         this.sttCredentials = await this._initSpeechCredentials(this.cs, this.vendor, this.label);
       } catch (error) {
-        if (this.canFallback) {
+        if (this.canFallback()) {
           this.notifyError(
             {
               msg: 'ASR error', details:`Invalid vendor ${this.vendor}, Error: ${error}`,
@@ -193,6 +203,56 @@ class SttTask extends Task {
     if (cs.hasGlobalSttPunctuation && !this.data.recognizer.punctuation) {
       this.data.recognizer.punctuation = cs.globalSttPunctuation;
     }
+    if (this.vendor === 'gladia') {
+      const { api_key, region } = this.sttCredentials;
+      const {url} = await this.createGladiaLiveSession({
+        api_key, region,
+        model: this.data.recognizer.model || 'solaria-1',
+        options: this.data.recognizer.gladiaOptions || {}
+      });
+      const {host, pathname, search} = new URL(url);
+      this.sttCredentials.host = host;
+      this.sttCredentials.path = `${pathname}${search}`;
+    }
+  }
+
+
+  async createGladiaLiveSession({
+    api_key,
+    region = 'us-west',
+    model = 'solaria-1',
+    options = {},
+  }) {
+    const url = `https://api.gladia.io/v2/live?region=${region}`;
+    const response = await fetch(url, {
+      method: 'POST',
+      headers: {
+        'x-gladia-key': api_key,
+        'Content-Type': 'application/json'
+      },
+      body: JSON.stringify({
+        encoding: 'wav/pcm',
+        bit_depth: 16,
+        sample_rate: 8000,
+        channels: 1,
+        model,
+        ...options,
+        messages_config: {
+          receive_final_transcripts: true,
+          receive_speech_events: true,
+          receive_errors: true,
+        }
+      })
+    });
+    if (!response.ok) {
+      const error = await response.text();
+      this.logger.error({url, status: response.status, error}, 'Error creating Gladia live session');
+      throw new Error(`Error creating Gladia live session: ${response.status} ${error}`);
+    }
+
+    const data = await response.json();
+    this.logger.debug({url: data.url}, 'Gladia Call registered');
+    return data;
   }
 
   addCustomEventListener(ep, event, handler) {
@@ -200,8 +260,19 @@ class SttTask extends Task {
     ep.addCustomEventListener(event, handler);
   }
 
-  removeCustomEventListeners() {
-    this.eventHandlers.forEach((h) => h.ep.removeCustomEventListener(h.event, h.handler));
+  removeCustomEventListeners(ep) {
+    if (ep) {
+      // for specific endpoint
+      this.eventHandlers.filter((h) => h.ep === ep).forEach((h) => {
+        h.ep.removeCustomEventListener(h.event, h.handler);
+      });
+      this.eventHandlers = this.eventHandlers.filter((h) => h.ep !== ep);
+      return;
+    } else {
+      // for all endpoints
+      this.eventHandlers.forEach((h) => h.ep.removeCustomEventListener(h.event, h.handler));
+      this.eventHandlers = [];
+    }
   }
 
   async _initSpeechCredentials(cs, vendor, label) {
@@ -215,6 +286,7 @@ class SttTask extends Task {
         account_sid: cs.accountSid,
         alert_type: AlertType.STT_NOT_PROVISIONED,
         vendor,
+        label,
         target_sid: cs.callSid
       }).catch((err) => this.logger.info({err}, 'Error generating alert for no stt'));
       // the ASR might have fallback configuration, should not done task here.
@@ -269,11 +341,13 @@ class SttTask extends Task {
     return credentials;
   }
 
-  get canFallback() {
+  canFallback() {
     return this.fallbackVendor && this.isHandledByPrimaryProvider && !this.cs.hasFallbackAsr;
   }
 
-  async _initFallback() {
+  // ep is optional for gather or any verb that have single ep,
+  // but transcribe does need as it might has 2 eps
+  async _initFallback(ep) {
     assert(this.fallbackVendor, 'fallback failed without fallbackVendor configuration');
     this.logger.info(`Failed to use primary STT provider, fallback to ${this.fallbackVendor}`);
     this.isHandledByPrimaryProvider = false;
@@ -286,7 +360,7 @@ class SttTask extends Task {
     this.data.recognizer.label = this.label;
     this.sttCredentials = await this._initSpeechCredentials(this.cs, this.vendor, this.label);
     // cleanup previous listener from previous vendor
-    this.removeCustomEventListeners();
+    this.removeCustomEventListeners(ep);
   }
 
   async compileHintsForCobalt(ep, hostport, model, token, hints) {
@@ -400,7 +474,7 @@ class SttTask extends Task {
     dgOptions.utteranceEndMs = dgOptions.utteranceEndMs || asrTimeout;
   }
 
-  _onVendorConnect(_cs, _ep) {
+  _onVendorConnect(cs, _ep) {
     this.logger.debug(`TaskGather:_on${this.vendor}Connect`);
   }
 
@@ -413,6 +487,7 @@ class SttTask extends Task {
       message: 'STT failure reported by vendor',
       detail: evt.error,
       vendor: this.vendor,
+      label: this.label,
       target_sid: cs.callSid
     }).catch((err) => this.logger.info({err}, `Error generating alert for ${this.vendor} connection failure`));
   }
@@ -426,6 +501,7 @@ class SttTask extends Task {
       alert_type: AlertType.STT_FAILURE,
       message: `Failed connecting to ${this.vendor} speech recognizer: ${reason}`,
       vendor: this.vendor,
+      label: this.label,
       target_sid: cs.callSid
     }).catch((err) => this.logger.info({err}, `Error generating alert for ${this.vendor} connection failure`));
   }

lib/tasks/transcribe.js

@@ -6,6 +6,8 @@ const {
   AwsTranscriptionEvents,
   AzureTranscriptionEvents,
   DeepgramTranscriptionEvents,
+  GladiaTranscriptionEvents,
+  DeepgramfluxTranscriptionEvents,
   SonioxTranscriptionEvents,
   CobaltTranscriptionEvents,
   IbmTranscriptionEvents,
@@ -13,7 +15,9 @@ const {
   JambonzTranscriptionEvents,
   TranscribeStatus,
   AssemblyAiTranscriptionEvents,
+  HoundifyTranscriptionEvents,
   VoxistTranscriptionEvents,
+  CartesiaTranscriptionEvents,
   OpenAITranscriptionEvents,
   VerbioTranscriptionEvents,
   SpeechmaticsTranscriptionEvents
@@ -66,6 +70,9 @@ class TaskTranscribe extends SttTask {
     this._bufferedTranscripts = [ [], [] ];  // for channel 1 and 2
     this.bugname_prefix = 'transcribe_';
     this.paused = false;
+    // fallback flags
+    this.isHandledByPrimaryProviderForEp1 = true;
+    this.isHandledByPrimaryProviderForEp2 = true;
   }
 
   get name() { return TaskName.Transcribe; }
@@ -136,22 +143,30 @@ class TaskTranscribe extends SttTask {
       stopTranscription = true;
       this.ep.stopTranscription({
         vendor: this.vendor,
-        bugname: this.bugname
+        bugname: this.bugname,
+        gracefulShutdown: this.paused ? false : true
       })
         .catch((err) => this.logger.info(err, 'Error TaskTranscribe:kill'));
     }
     if (this.transcribing2 && this.ep2?.connected) {
       stopTranscription = true;
-      this.ep2.stopTranscription({vendor: this.vendor, bugname: this.bugname})
+      this.ep2.stopTranscription({
+        vendor: this.vendor,
+        bugname: this.bugname,
+        gracefulShutdown: this.paused ? false : true
+      })
         .catch((err) => this.logger.info(err, 'Error TaskTranscribe:kill'));
     }
 
+    this.cs.emit('transcribe-stop');
+
     return stopTranscription;
   }
 
   async kill(cs) {
     super.kill(cs);
     const stopTranscription = this._stopTranscription();
+    cs.stopSttLatencyVad();
     // hangup after 1 sec if we don't get a final transcription
     if (stopTranscription) this._timer = setTimeout(() => this.notifyTaskDone(), 1500);
     else this.notifyTaskDone();
@@ -227,10 +242,35 @@ class TaskTranscribe extends SttTask {
           this._onVendorConnect.bind(this, cs, ep));
         this.addCustomEventListener(ep, DeepgramTranscriptionEvents.ConnectFailure,
           this._onVendorConnectFailure.bind(this, cs, ep, channel));
+        this.addCustomEventListener(ep, DeepgramTranscriptionEvents.Error, this._onVendorError.bind(this, cs, ep));
+
 
         /* if app sets deepgramOptions.utteranceEndMs they essentially want continuous asr */
         //if (opts.DEEPGRAM_SPEECH_UTTERANCE_END_MS) this.isContinuousAsr = true;
 
+        break;
+      case 'deepgramflux':
+        this.bugname = `${this.bugname_prefix}deepgramflux_transcribe`;
+        this.addCustomEventListener(ep, DeepgramfluxTranscriptionEvents.Transcription,
+          this._onTranscription.bind(this, cs, ep, channel));
+        this.addCustomEventListener(ep, DeepgramfluxTranscriptionEvents.Connect,
+          this._onVendorConnect.bind(this, cs, ep));
+        this.addCustomEventListener(ep, DeepgramfluxTranscriptionEvents.ConnectFailure,
+          this._onVendorConnectFailure.bind(this, cs, ep, channel));
+        this.addCustomEventListener(ep, DeepgramfluxTranscriptionEvents.Error, this._onVendorError.bind(this, cs, ep));
+
+        break;
+
+      case 'gladia':
+        this.bugname = `${this.bugname_prefix}gladia_transcribe`;
+        this.addCustomEventListener(ep, GladiaTranscriptionEvents.Transcription,
+          this._onTranscription.bind(this, cs, ep, channel));
+        this.addCustomEventListener(ep, GladiaTranscriptionEvents.Connect,
+          this._onVendorConnect.bind(this, cs, ep));
+        this.addCustomEventListener(ep, GladiaTranscriptionEvents.ConnectFailure,
+          this._onVendorConnectFailure.bind(this, cs, ep, channel));
+        this.addCustomEventListener(ep, GladiaTranscriptionEvents.Error, this._onVendorError.bind(this, cs, ep));
+
         break;
       case 'soniox':
         this.bugname = `${this.bugname_prefix}soniox_transcribe`;
@@ -301,6 +341,18 @@ class TaskTranscribe extends SttTask {
           this._onVendorConnectFailure.bind(this, cs, ep, channel));
         break;
 
+      case 'houndify':
+        this.bugname = `${this.bugname_prefix}houndify_transcribe`;
+        this.addCustomEventListener(ep, HoundifyTranscriptionEvents.Transcription,
+          this._onTranscription.bind(this, cs, ep, channel));
+        this.addCustomEventListener(ep, HoundifyTranscriptionEvents.Error,
+          this._onVendorError.bind(this, cs, ep));
+        this.addCustomEventListener(ep, HoundifyTranscriptionEvents.ConnectFailure,
+          this._onVendorConnectFailure.bind(this, cs, ep, channel));
+        this.addCustomEventListener(ep, HoundifyTranscriptionEvents.Connect,
+          this._onVendorConnect.bind(this, cs, ep));
+        break;
+
       case 'voxist':
         this.bugname = `${this.bugname_prefix}voxist_transcribe`;
         this.addCustomEventListener(ep, VoxistTranscriptionEvents.Transcription,
@@ -312,6 +364,17 @@ class TaskTranscribe extends SttTask {
           this._onVendorConnectFailure.bind(this, cs, ep, channel));
         break;
 
+      case 'cartesia':
+        this.bugname = `${this.bugname_prefix}cartesia_transcribe`;
+        this.addCustomEventListener(ep, CartesiaTranscriptionEvents.Transcription,
+          this._onTranscription.bind(this, cs, ep, channel));
+        this.addCustomEventListener(ep,
+          CartesiaTranscriptionEvents.Connect, this._onVendorConnect.bind(this, cs, ep));
+        this.addCustomEventListener(ep, CartesiaTranscriptionEvents.Error, this._onVendorError.bind(this, cs, ep));
+        this.addCustomEventListener(ep, CartesiaTranscriptionEvents.ConnectFailure,
+          this._onVendorConnectFailure.bind(this, cs, ep, channel));
+        break;
+
       case 'speechmatics':
         this.bugname = `${this.bugname_prefix}speechmatics_transcribe`;
         this.addCustomEventListener(
@@ -406,9 +469,16 @@ class TaskTranscribe extends SttTask {
       bugname: this.bugname,
       hostport: this.hostport
     });
+
+    // Some vendor use single connection, that we cannot use onConnect event to track transcription start
+    this.cs.emit('transcribe-start');
   }
 
   async _onTranscription(cs, ep, channel, evt, fsEvent) {
+    // check if we are in graceful shutdown mode
+    if (ep.gracefulShutdownResolver) {
+      ep.gracefulShutdownResolver();
+    }
     // make sure this is not a transcript from answering machine detection
     const bugname = fsEvent.getHeader('media-bugname');
     const finished = fsEvent.getHeader('transcription-session-finished');
@@ -420,6 +490,9 @@ class TaskTranscribe extends SttTask {
 
     if (this.vendor === 'ibm' && evt?.state === 'listening') return;
 
+    // emit an event to the call session to track the time transcription is received
+    cs.emit('on-transcription');
+
     if (this.vendor === 'deepgram' && evt.type === 'UtteranceEnd') {
       /* we will only get this when we have set utterance_end_ms */
 
@@ -581,14 +654,28 @@ class TaskTranscribe extends SttTask {
   }
 
   async _resolve(channel, evt) {
+    let sttLatencyMetrics = {};
     if (evt.is_final) {
+      const sttLatency = this.cs.calculateSttLatency();
+      if (sttLatency) {
+        sttLatencyMetrics = {
+          'stt.latency_ms': `${sttLatency.stt_latency_ms}`,
+          'stt.talkspurts': JSON.stringify(sttLatency.talkspurts),
+          'stt.start_time': sttLatency.stt_start_time,
+          'stt.stop_time': sttLatency.stt_stop_time,
+          'stt.usage': sttLatency.stt_usage,
+        };
+      }
+      // time to reset the stt latency
+      this.cs.emit('transcribe-start');
       /* we've got a final transcript, so end the otel child span for this channel */
       if (this.childSpan[channel - 1] && this.childSpan[channel - 1].span) {
         this.childSpan[channel - 1].span.setAttributes({
           channel,
           'stt.label': this.label || 'None',
           'stt.resolve': 'transcript',
-          'stt.result': JSON.stringify(evt)
+          'stt.result': JSON.stringify(evt),
+          ...sttLatencyMetrics
         });
         this.childSpan[channel - 1].span.end();
       }
@@ -597,9 +684,13 @@ class TaskTranscribe extends SttTask {
     if (this.transcriptionHook) {
       const b3 = this.getTracingPropagation();
       const httpHeaders = b3 && {b3};
+      const latencies = Object.fromEntries(
+        Object.entries(sttLatencyMetrics).map(([key, value]) => [key.replace('stt.', 'stt_'), value])
+      );
       const payload = {
         ...this.cs.callInfo,
         ...httpHeaders,
+        ...latencies,
         ...(evt.alternatives && {speech: evt}),
         ...(evt.type && {speechEvent: evt})
       };
@@ -688,16 +779,17 @@ class TaskTranscribe extends SttTask {
   }
 
   async _startFallback(cs, _ep, evt) {
-    if (this.canFallback) {
+    if (this.canFallback(_ep)) {
       _ep.stopTranscription({
         vendor: this.vendor,
-        bugname: this.bugname
+        bugname: this.bugname,
+        gracefulShutdown: false
       })
         .catch((err) => this.logger.error({err}, `Error stopping transcription for primary vendor ${this.vendor}`));
       try {
         this.notifyError({ msg: 'ASR error',
           details:`STT Vendor ${this.vendor} error: ${evt.error || evt.reason}`, failover: 'in progress'});
-        await this._initFallback();
+        await this._initFallback(_ep);
         let channel = 1;
         if (this.ep !== _ep) {
           channel = 2;
@@ -806,6 +898,41 @@ class TaskTranscribe extends SttTask {
     if (this._asrTimer) clearTimeout(this._asrTimer);
     this._asrTimer = null;
   }
+
+  // We need to keep track the fallback is happened for each endpoint
+  // override the canFallback and _initFallback methods to make sure that
+  // we only fallback once per endpoint
+  // we want to keep track this on task level instead of endpoint level
+  // because the endpoint instance is used across multiple tasks.
+  canFallback(ep) {
+    let isHandledByPrimaryProvider = this.isHandledByPrimaryProvider;
+    if (ep === this.ep) {
+      isHandledByPrimaryProvider = this.isHandledByPrimaryProviderForEp1;
+    } else if (ep === this.ep2) {
+      isHandledByPrimaryProvider = this.isHandledByPrimaryProviderForEp2;
+    }
+
+    const isOneOfEndpointAlreadyFallenBack = !!this.ep && !!this.ep2 &&
+      this.isHandledByPrimaryProviderForEp1 !== this.isHandledByPrimaryProviderForEp2;
+
+    // fallback is configured
+    return this.fallbackVendor &&
+      // has this endpoint already fallen back
+      isHandledByPrimaryProvider &&
+      // in global level, is there any fallback is already happened
+      // one fallen endpoint will mark cs.hasFallbackAsr to true,
+      // so if one endpoint was fallen, the other endpoint would be able to fallback.
+      (isOneOfEndpointAlreadyFallenBack || !this.cs.hasFallbackAsr);
+  }
+
+  _initFallback(ep) {
+    if (ep === this.ep) {
+      this.isHandledByPrimaryProviderForEp1 = false;
+    } else if (ep === this.ep2) {
+      this.isHandledByPrimaryProviderForEp2 = false;
+    }
+    return super._initFallback(ep);
+  }
 }
 
 module.exports = TaskTranscribe;

lib/tasks/tts-task.js

@@ -3,6 +3,16 @@ const { TaskPreconditions } = require('../utils/constants');
 const { SpeechCredentialError } = require('../utils/error');
 const dbUtils = require('../utils/db-utils');
 
+const extractPlaybackId = (str) => {
+  // Match say:{...} and capture the content inside braces
+  const match = str.match(/say:\{([^}]*)\}/);
+  if (!match) return null;
+
+  // Look for playback_id=value within the captured content
+  const playbackMatch = match[1].match(/playback_id=([^,]*)/);
+  return playbackMatch ? playbackMatch[1] : null;
+};
+
 class TtsTask extends Task {
 
   constructor(logger, data, parentTask) {
@@ -22,10 +32,19 @@ class TtsTask extends Task {
     this.disableTtsCache = this.data.disableTtsCache;
     this.options = this.synthesizer.options || {};
     this.instructions = this.data.instructions;
+    this.playbackIds = [];
+  }
+
+  getPlaybackId(offset) {
+    return this.playbackIds[offset];
   }
 
   async exec(cs) {
     super.exec(cs);
+    // update disableTtsCache from call session if not set in task
+    if (this.data.disableTtsCache == null) {
+      this.disableTtsCache = cs.disableTtsCache;
+    }
     if (cs.synthesizer) {
       this.options = {...cs.synthesizer.options, ...this.options};
       this.data.synthesizer = this.data.synthesizer || {};
@@ -66,55 +85,67 @@ class TtsTask extends Task {
   }
 
   async setTtsStreamingChannelVars(vendor, language, voice, credentials, ep) {
-    const {api_key, model_id, custom_tts_streaming_url, auth_token} = credentials;
-    let obj;
+    const {api_key, model_id, api_uri, custom_tts_streaming_url, auth_token, options} = credentials;
+    // api_key, model_id, api_uri, custom_tts_streaming_url, and auth_token are encoded in the credentials
+    // allow them to be overriden via config, using options
+    // give preference to options passed in via config
+    const parsed_options = options ? JSON.parse(options) : {};
+    const local_options = {...parsed_options, ...this.options};
+    const local_voice_settings = {...(parsed_options.voice_settings || {}), ...(this.options.voice_settings || {})};
+    const local_api_key =  local_options.api_key ?? api_key;
+    const local_model_id = local_options.model_id ?? model_id;
+    const local_api_uri = local_options.api_uri ?? api_uri;
+    const local_custom_tts_streaming_url = local_options.custom_tts_streaming_url ?? custom_tts_streaming_url;
+    const local_auth_token = local_options.auth_token ?? auth_token;
 
-    this.logger.debug({credentials},
-      `setTtsStreamingChannelVars: vendor: ${vendor}, language: ${language}, voice: ${voice}`);
+
+    let obj;
 
     switch (vendor) {
       case 'deepgram':
         obj = {
-          DEEPGRAM_API_KEY: api_key,
+          DEEPGRAM_API_KEY: local_api_key,
           DEEPGRAM_TTS_STREAMING_MODEL: voice
         };
         break;
       case 'cartesia':
         obj = {
-          CARTESIA_API_KEY: api_key,
-          CARTESIA_TTS_STREAMING_MODEL_ID: model_id,
+          CARTESIA_API_KEY: local_api_key,
+          CARTESIA_TTS_STREAMING_MODEL_ID: local_model_id,
           CARTESIA_TTS_STREAMING_VOICE_ID: voice,
           CARTESIA_TTS_STREAMING_LANGUAGE: language || 'en',
         };
         break;
       case 'elevenlabs':
-        const {stability, similarity_boost, use_speaker_boost, style, speed} = this.options.voice_settings || {};
+        // eslint-disable-next-line max-len
+        const {stability, similarity_boost, use_speaker_boost, style, speed} = local_voice_settings || {};
         obj = {
-          ELEVENLABS_API_KEY: api_key,
-          ELEVENLABS_TTS_STREAMING_MODEL_ID: model_id,
+          ELEVENLABS_API_KEY: local_api_key,
+          ...(api_uri && {ELEVENLABS_API_URI: local_api_uri}),
+          ELEVENLABS_TTS_STREAMING_MODEL_ID: local_model_id,
           ELEVENLABS_TTS_STREAMING_VOICE_ID: voice,
           // 20/12/2024 - only eleven_turbo_v2_5 support multiple language
-          ...(['eleven_turbo_v2_5'].includes(model_id) && {ELEVENLABS_TTS_STREAMING_LANGUAGE: language}),
+          ...(['eleven_turbo_v2_5'].includes(local_model_id) && {ELEVENLABS_TTS_STREAMING_LANGUAGE: language}),
           ...(stability && {ELEVENLABS_TTS_STREAMING_VOICE_SETTINGS_STABILITY: stability}),
           ...(similarity_boost && {ELEVENLABS_TTS_STREAMING_VOICE_SETTINGS_SIMILARITY_BOOST: similarity_boost}),
           ...(use_speaker_boost && {ELEVENLABS_TTS_STREAMING_VOICE_SETTINGS_USE_SPEAKER_BOOST: use_speaker_boost}),
           ...(style && {ELEVENLABS_TTS_STREAMING_VOICE_SETTINGS_STYLE: style}),
           // speed has value 0.7 to 1.2, 1.0 is default, make sure we send the value event it's 0
           ...(speed !== null && speed !== undefined && {ELEVENLABS_TTS_STREAMING_VOICE_SETTINGS_SPEED: `${speed}`}),
-          ...(this.options.pronunciation_dictionary_locators &&
-            Array.isArray(this.options.pronunciation_dictionary_locators) && {
+          ...(local_options.pronunciation_dictionary_locators &&
+            Array.isArray(local_options.pronunciation_dictionary_locators) && {
             ELEVENLABS_TTS_STREAMING_PRONUNCIATION_DICTIONARY_LOCATORS:
-            JSON.stringify(this.options.pronunciation_dictionary_locators)
+            JSON.stringify(local_options.pronunciation_dictionary_locators)
           }),
         };
         break;
       case 'rimelabs':
         const {
           pauseBetweenBrackets, phonemizeBetweenBrackets, inlineSpeedAlpha, speedAlpha, reduceLatency
-        } = this.options;
+        } = local_options;
         obj = {
-          RIMELABS_API_KEY: api_key,
-          RIMELABS_TTS_STREAMING_MODEL_ID: model_id,
+          RIMELABS_API_KEY: local_api_key,
+          RIMELABS_TTS_STREAMING_MODEL_ID: local_model_id,
           RIMELABS_TTS_STREAMING_VOICE_ID: voice,
           RIMELABS_TTS_STREAMING_LANGUAGE: language || 'en',
           ...(pauseBetweenBrackets && {RIMELABS_TTS_STREAMING_PAUSE_BETWEEN_BRACKETS: pauseBetweenBrackets}),
@@ -129,8 +160,8 @@ class TtsTask extends Task {
         if (vendor.startsWith('custom:')) {
           const use_tls = custom_tts_streaming_url.startsWith('wss://');
           obj = {
-            CUSTOM_TTS_STREAMING_HOST: custom_tts_streaming_url.replace(/^(ws|wss):\/\//, ''),
-            CUSTOM_TTS_STREAMING_API_KEY: auth_token,
+            CUSTOM_TTS_STREAMING_HOST: local_custom_tts_streaming_url.replace(/^(ws|wss):\/\//, ''),
+            CUSTOM_TTS_STREAMING_API_KEY: local_auth_token,
             CUSTOM_TTS_STREAMING_VOICE_ID: voice,
             CUSTOM_TTS_STREAMING_LANGUAGE: language || 'en',
             CUSTOM_TTS_STREAMING_USE_TLS: use_tls
@@ -185,6 +216,9 @@ class TtsTask extends Task {
     } else if (vendor === 'rimelabs') {
       credentials = credentials || {};
       credentials.model_id = this.options.model_id || credentials.model_id;
+    } else if (vendor === 'inworld') {
+      credentials = credentials || {};
+      credentials.model_id = this.options.model_id || credentials.model_id;
     } else if (vendor === 'whisper') {
       credentials = credentials || {};
       credentials.model_id = this.options.model_id || credentials.model_id;
@@ -240,15 +274,16 @@ class TtsTask extends Task {
           account_sid,
           alert_type: AlertType.TTS_NOT_PROVISIONED,
           vendor,
+          label,
           target_sid: cs.callSid
         }).catch((err) => this.logger.info({err}, 'Error generating alert for no tts'));
         throw new SpeechCredentialError('no provisioned speech credentials for TTS');
       }
 
       /* produce an audio segment from the provided text */
-      const generateAudio = async(text) => {
-        if (this.killed) return;
-        if (text.startsWith('silence_stream://')) return text;
+      const generateAudio = async(text, index) => {
+        if (this.killed) return {index, filePath: null};
+        if (text.startsWith('silence_stream://')) return {index, filePath: text};
 
         /* otel: trace time for tts */
         if (!preCache && !this._disableTracing)  {
@@ -290,16 +325,35 @@ class TtsTask extends Task {
                 vendor,
                 language,
                 characters: text.length,
-                elapsedTime: rtt
+                elapsedTime: rtt,
+                servedFromCache,
+                'id': this.id
               });
             }
+            if (servedFromCache) {
+              this.notifyStatus({
+                event: 'synthesized-audio',
+                vendor,
+                language,
+                servedFromCache,
+                'id': this.id
+              });
+            }
+            return {index, filePath, playbackId: null};
           }
           else {
+            const playbackId = extractPlaybackId(filePath);
             this.logger.debug('Say: a streaming tts api will be used');
             const modifiedPath = filePath.replace('say:{', `say:{session-uuid=${ep.uuid},`);
-            return modifiedPath;
+            this.notifyStatus({
+              event: 'synthesized-audio',
+              vendor,
+              language,
+              servedFromCache,
+              'id': this.id
+            });
+            return {index, filePath: modifiedPath, playbackId};
           }
-          return filePath;
         } catch (err) {
           this.logger.info({err}, 'Error synthesizing tts');
           if (this.otelSpan) this.otelSpan.end();
@@ -307,6 +361,7 @@ class TtsTask extends Task {
             account_sid: cs.accountSid,
             alert_type: AlertType.TTS_FAILURE,
             vendor,
+            label,
             detail: err.message,
             target_sid: cs.callSid
           }).catch((err) => this.logger.info({err}, 'Error generating alert for tts failure'));
@@ -314,8 +369,20 @@ class TtsTask extends Task {
         }
       };
 
-      const arr = this.text.map((t) => (this._validateURL(t) ? t : generateAudio(t)));
-      return (await Promise.all(arr)).filter((fp) => fp && fp.length);
+      // process all text segments in parallel will cause ordering issue
+      // so we attach index to each promise result and sort them later
+
+      const arr = this.text.map((t, index) => (this._validateURL(t) ?
+        Promise.resolve({index, filePath: t, playbackId: null}) : generateAudio(t, index)));
+      const results = await Promise.all(arr);
+      const sorted = results.sort((a, b) => a.index - b.index);
+
+      return sorted
+        .filter((fp) => fp.filePath && fp.filePath.length)
+        .map((r) => {
+          this.playbackIds.push(r.playbackId);
+          return r.filePath;
+        });
     } catch (err) {
       this.logger.info(err, 'TaskSay:exec error');
       throw err;

lib/utils/amd-utils.js

@@ -281,13 +281,17 @@ module.exports = (logger) => {
 
     /* set stt options */
     logger.info(`starting amd for vendor ${vendor} and language ${language}`);
-    const sttOpts = amd.setChannelVarsForStt({name: TaskName.Gather}, sttCredentials, language, {
-      vendor,
-      hints,
-      enhancedModel: true,
-      altLanguages: opts.recognizer?.altLanguages || [],
-      initialSpeechTimeoutMs: opts.resolveTimeoutMs,
-    });
+    /* if opts contains recognizer object use that config for stt, otherwise use defaults */
+    const rOpts = opts.recognizer ?
+      opts.recognizer :
+      {
+        vendor,
+        hints,
+        enhancedModel: true,
+        altLanguages: opts.recognizer?.altLanguages || [],
+        initialSpeechTimeoutMs: opts.resolveTimeoutMs,
+      };
+    const sttOpts = amd.setChannelVarsForStt({name: TaskName.Gather}, sttCredentials, language, rOpts);
 
     await ep.set(sttOpts).catch((err) => logger.info(err, 'Error setting channel variables'));
 
@@ -401,24 +405,30 @@ module.exports = (logger) => {
     if (ep.amd) {
       vendor = ep.amd.vendor;
       ep.amd.stopAllTimers();
-
-      ep.removeListener(GoogleTranscriptionEvents.Transcription, ep.amd.transcriptionHandler);
-      ep.removeListener(GoogleTranscriptionEvents.EndOfUtterance, ep.amd.EndOfUtteranceHandler);
-      ep.removeListener(AwsTranscriptionEvents.Transcription, ep.amd.transcriptionHandler);
-      ep.removeListener(AzureTranscriptionEvents.Transcription, ep.amd.transcriptionHandler);
-      ep.removeListener(AzureTranscriptionEvents.NoSpeechDetected, ep.amd.noSpeechHandler);
-      ep.removeListener(NuanceTranscriptionEvents.Transcription, ep.amd.transcriptionHandler);
-      ep.removeListener(DeepgramTranscriptionEvents.Transcription, ep.amd.transcriptionHandler);
-      ep.removeListener(SonioxTranscriptionEvents.Transcription, ep.amd.transcriptionHandler);
-      ep.removeListener(IbmTranscriptionEvents.Transcription, ep.amd.transcriptionHandler);
-      ep.removeListener(NvidiaTranscriptionEvents.Transcription, ep.amd.transcriptionHandler);
-      ep.removeListener(JambonzTranscriptionEvents.Transcription, ep.amd.transcriptionHandler);
-
+      try {
+        ep.removeListener(GoogleTranscriptionEvents.Transcription, ep.amd.transcriptionHandler);
+        ep.removeListener(GoogleTranscriptionEvents.EndOfUtterance, ep.amd.EndOfUtteranceHandler);
+        ep.removeListener(AwsTranscriptionEvents.Transcription, ep.amd.transcriptionHandler);
+        ep.removeListener(AzureTranscriptionEvents.Transcription, ep.amd.transcriptionHandler);
+        ep.removeListener(AzureTranscriptionEvents.NoSpeechDetected, ep.amd.noSpeechHandler);
+        ep.removeListener(NuanceTranscriptionEvents.Transcription, ep.amd.transcriptionHandler);
+        ep.removeListener(DeepgramTranscriptionEvents.Transcription, ep.amd.transcriptionHandler);
+        ep.removeListener(SonioxTranscriptionEvents.Transcription, ep.amd.transcriptionHandler);
+        ep.removeListener(IbmTranscriptionEvents.Transcription, ep.amd.transcriptionHandler);
+        ep.removeListener(NvidiaTranscriptionEvents.Transcription, ep.amd.transcriptionHandler);
+        ep.removeListener(JambonzTranscriptionEvents.Transcription, ep.amd.transcriptionHandler);
+      } catch (error) {
+        logger.error('Unable to Remove AMD Listener', error);
+      }
       ep.amd = null;
     }
 
     if (ep.connected) {
-      ep.stopTranscription({vendor, bugname})
+      ep.stopTranscription({
+        vendor,
+        bugname,
+        gracefulShutdown: false
+      })
         .catch((err) => logger.info(err, 'stopAmd: Error stopping transcription'));
       task.emit('amd', {type: AmdEvents.Stopped});
       ep.execute('avmd_stop').catch((err) => this.logger.info(err, 'Error stopping avmd'));

lib/utils/background-task-manager.js

@@ -65,7 +65,7 @@ class BackgroundTaskManager extends Emitter {
       this.logger.info(`stopping background task: ${type}`);
       task.removeAllListeners();
       task.span.end();
-      task.kill();
+      task.kill(this.cs);
       // Remove task from managed List
       this.tasks.delete(type);
     }

lib/utils/constants.json

@@ -1,5 +1,6 @@
 {
   "TaskName": {
+    "Alert": "alert",
     "Answer": "answer",
     "Conference": "conference",
     "Config": "config",
@@ -93,7 +94,20 @@
   "DeepgramTranscriptionEvents": {
     "Transcription": "deepgram_transcribe::transcription",
     "ConnectFailure": "deepgram_transcribe::connect_failed",
-    "Connect": "deepgram_transcribe::connect"
+    "Connect": "deepgram_transcribe::connect",
+    "Error": "deepgram_transcribe::error"
+  },
+  "DeepgramfluxTranscriptionEvents": {
+    "Transcription": "deepgramflux_transcribe::transcription",
+    "ConnectFailure": "deepgramflux_transcribe::connect_failed",
+    "Connect": "deepgramflux_transcribe::connect",
+    "Error": "deepgramflux_transcribe::error"
+  },
+  "GladiaTranscriptionEvents": {
+    "Transcription": "gladia_transcribe::transcription",
+    "ConnectFailure": "gladia_transcribe::connect_failed",
+    "Connect": "gladia_transcribe::connect",
+    "Error": "gladia_transcribe::error"
   },
   "SonioxTranscriptionEvents": {
     "Transcription": "soniox_transcribe::transcription",
@@ -161,15 +175,30 @@
     "ConnectFailure": "assemblyai_transcribe::connect_failed",
     "Connect": "assemblyai_transcribe::connect"
   },
+  "HoundifyTranscriptionEvents": {
+    "Transcription": "houndify_transcribe::transcription",
+    "Error": "houndify_transcribe::error",
+    "ConnectFailure": "houndify_transcribe::connect_failed",
+    "Connect": "houndify_transcribe::connect"
+  },
   "VoxistTranscriptionEvents": {
     "Transcription": "voxist_transcribe::transcription",
     "Error": "voxist_transcribe::error",
     "ConnectFailure": "voxist_transcribe::connect_failed",
     "Connect": "voxist_transcribe::connect"
   },
+  "CartesiaTranscriptionEvents": {
+    "Transcription": "cartesia_transcribe::transcription",
+    "Error": "cartesia_transcribe::error",
+    "ConnectFailure": "cartesia_transcribe::connect_failed",
+    "Connect": "cartesia_transcribe::connect"
+  },
   "VadDetection": {
     "Detection": "vad_detect:detection"
   },
+  "SileroVadDetection": {
+    "Detection": "vad_silero:detect"
+  },
   "ListenEvents": {
     "Connect": "mod_audio_fork::connect",
     "ConnectFailure": "mod_audio_fork::connect_failed",
@@ -194,13 +223,6 @@
     "Disconnect": "openai_s2s::disconnect",
     "ServerEvent": "openai_s2s::server_event"
   },
-  "LlmEvents_Aws": {
-    "Error": "error",
-    "Connect": "aws_s2s::connect",
-    "ConnectFailure": "aws_s2s::connect_failed",
-    "Disconnect": "aws_s2s::disconnect",
-    "ServerEvent": "aws_s2s::server_event"
-  },
   "LlmEvents_Google": {
     "Error": "error",
     "Connect": "google_s2s::connect",
@@ -244,6 +266,7 @@
   "KillReason": {
     "Hangup": "hangup",
     "Replaced": "replaced",
+    "ReferComplete": "refer-complete",
     "MediaTimeout": "media_timeout"
   },
   "HookMsgTypes": [
@@ -312,7 +335,8 @@
     "Empty": "tts_streaming::empty",
     "Pause": "tts_streaming::pause",
     "Resume": "tts_streaming::resume",
-    "ConnectFailure": "tts_streaming::connect_failed"
+    "ConnectFailure": "tts_streaming::connect_failed",
+    "Connected": "tts_streaming::connected"
   },
   "TtsStreamingConnectionStatus": {
     "NotConnected": "not_connected",
@@ -332,5 +356,8 @@
   "WS_CLOSE_CODES": {
     "NormalClosure": 1000,
     "GoingAway": 1001
-  }
+  },
+  "NON_FANTAL_ERRORS": [
+    "File Not Found"
+  ]
 }

lib/utils/db-utils.js

@@ -81,6 +81,15 @@ const speechMapper = (cred) => {
       obj.deepgram_tts_uri = o.deepgram_tts_uri;
       obj.deepgram_stt_use_tls = o.deepgram_stt_use_tls;
     }
+    else if ('gladia' === obj.vendor) {
+      const o = JSON.parse(decrypt(credential));
+      obj.api_key = o.api_key;
+      obj.region = o.region;
+    }
+    else if ('deepgramflux' === obj.vendor) {
+      const o = JSON.parse(decrypt(credential));
+      obj.api_key = o.api_key;
+    }
     else if ('soniox' === obj.vendor) {
       const o = JSON.parse(decrypt(credential));
       obj.api_key = o.api_key;
@@ -97,6 +106,7 @@ const speechMapper = (cred) => {
       const o = JSON.parse(decrypt(credential));
       obj.api_key = o.api_key;
       obj.model_id = o.model_id;
+      obj.api_uri = o.api_uri;
       obj.options = o.options;
     }
     else if ('playht' === obj.vendor) {
@@ -110,6 +120,7 @@ const speechMapper = (cred) => {
       const o = JSON.parse(decrypt(credential));
       obj.api_key = o.api_key;
       obj.model_id = o.model_id;
+      obj.stt_model_id = o.stt_model_id;
       obj.embedding = o.embedding;
       obj.options = o.options;
     }
@@ -119,9 +130,29 @@ const speechMapper = (cred) => {
       obj.model_id = o.model_id;
       obj.options = o.options;
     }
+    else if ('resemble' === obj.vendor) {
+      const o = JSON.parse(decrypt(credential));
+      obj.api_key = o.api_key;
+      obj.resemble_tts_use_tls = o.resemble_tts_use_tls;
+      obj.resemble_tts_uri = o.resemble_tts_uri;
+    }
+    else if ('inworld' === obj.vendor) {
+      const o = JSON.parse(decrypt(credential));
+      obj.api_key = o.api_key;
+      obj.model_id = o.model_id;
+      obj.options = o.options;
+    }
     else if ('assemblyai' === obj.vendor) {
       const o = JSON.parse(decrypt(credential));
       obj.api_key = o.api_key;
+      obj.service_version = o.service_version;
+    }
+    else if ('houndify' === obj.vendor) {
+      const o = JSON.parse(decrypt(credential));
+      obj.client_id = o.client_id;
+      obj.client_key = o.client_key;
+      obj.user_id = o.user_id;
+      obj.houndify_server_uri = o.houndify_server_uri;
     }
     else if ('voxist' === obj.vendor) {
       const o = JSON.parse(decrypt(credential));

lib/utils/http-requestor.js

@@ -41,7 +41,7 @@ class HttpRequestor extends BaseRequestor {
   constructor(logger, account_sid, hook, secret) {
     super(logger, account_sid, hook, secret);
 
-    this.method = hook.method || 'POST';
+    this.method = hook.method?.toUpperCase() || 'POST';
     this.authHeader = basicAuth(hook.username, hook.password);
     this.backoffMs = 500;
 
@@ -111,7 +111,7 @@ class HttpRequestor extends BaseRequestor {
 
     const payload = params ? snakeCaseKeys(params, ['customerData', 'sip', 'env_vars', 'args']) : null;
     const url = hook.url || hook;
-    const method = hook.method || 'POST';
+    const method = hook.method?.toUpperCase() || 'POST';
     let buf = '';
     httpHeaders = {
       ...httpHeaders,
@@ -119,7 +119,7 @@ class HttpRequestor extends BaseRequestor {
     };
 
     assert.ok(url, 'HttpRequestor:request url was not provided');
-    assert.ok, (['GET', 'POST'].includes(method), `HttpRequestor:request method must be 'GET' or 'POST' not ${method}`);
+    assert.ok(['GET', 'POST'].includes(method), `HttpRequestor:request method must be 'GET' or 'POST' not ${method}`);
     const startAt = process.hrtime();
 
     /* if we have an absolute url, and it is ws then do a websocket connection */
@@ -191,7 +191,7 @@ class HttpRequestor extends BaseRequestor {
         method,
         headers: hdrs,
         ...('POST' === method && {body: JSON.stringify(payload)}),
-        timeout: HTTP_TIMEOUT,
+        headersTimeout: HTTP_TIMEOUT,
         followRedirects: false
       };
 

lib/utils/install-srf-locals.js

@@ -173,7 +173,8 @@ function installSrfLocals(srf, logger, {
     lookupAccountCapacitiesBySid,
     lookupSmppGateways,
     lookupClientByAccountAndUsername,
-    lookupSystemInformation
+    lookupSystemInformation,
+    lookupLcrByAccount
   } = require('@jambonz/db-helpers')({
     host: JAMBONES_MYSQL_HOST,
     user: JAMBONES_MYSQL_USER,
@@ -279,7 +280,8 @@ function installSrfLocals(srf, logger, {
       retrieveByPatternSortedSet,
       sortedSetLength,
       sortedSetPositionByPattern,
-      getVerbioAccessToken
+      getVerbioAccessToken,
+      lookupLcrByAccount
     },
     parentLogger: logger,
     getSBC,

lib/utils/media-endpoint.js

@@ -0,0 +1,115 @@
+const {
+  JAMBONES_USE_FREESWITCH_TIMER_FD,
+  JAMBONES_MEDIA_TIMEOUT_MS,
+  JAMBONES_MEDIA_HOLD_TIMEOUT_MS,
+  JAMBONES_TRANSCRIBE_EP_DESTROY_DELAY_MS,
+} = require('../config');
+const { sleepFor } = require('./helpers');
+
+const createMediaEndpoint = async(srf, logger, {
+  activeMs,
+  drachtioFsmrfOptions = {},
+  onHoldMusic,
+  inbandDtmfEnabled,
+  mediaTimeoutHandler,
+} = {}) => {
+  const { getFreeswitch } = srf.locals;
+  const ms = activeMs || getFreeswitch();
+  if (!ms)
+    throw new Error('no available Freeswitch for creating media endpoint');
+
+  const ep = await ms.createEndpoint(drachtioFsmrfOptions);
+
+  // Configure the endpoint
+  const opts = {
+    ...(onHoldMusic && {holdMusic: `shout://${onHoldMusic.replace(/^https?:\/\//, '')}`}),
+    ...(JAMBONES_USE_FREESWITCH_TIMER_FD && {timer_name: 'timerfd'}),
+    ...(JAMBONES_MEDIA_TIMEOUT_MS && {media_timeout: JAMBONES_MEDIA_TIMEOUT_MS}),
+    ...(JAMBONES_MEDIA_HOLD_TIMEOUT_MS && {media_hold_timeout: JAMBONES_MEDIA_HOLD_TIMEOUT_MS})
+  };
+  if (Object.keys(opts).length > 0) {
+    ep.set(opts);
+  }
+  // inbandDtmfEnabled
+  if (inbandDtmfEnabled) {
+    // https://developer.signalwire.com/freeswitch/FreeSWITCH-Explained/Modules/mod-dptools/6587132/#0-about
+    ep.execute('start_dtmf').catch((err) => {
+      logger.error('Error starting inband DTMF', { error: err });
+    });
+    ep.inbandDtmfEnabled = true;
+  }
+  // Handle Media Timeout
+  if (mediaTimeoutHandler) {
+    ep.once('destroy', (evt) => {
+      mediaTimeoutHandler(evt, ep);
+    });
+  }
+  // Handle graceful shutdown for endpoint if required
+  if (JAMBONES_TRANSCRIBE_EP_DESTROY_DELAY_MS > 0) {
+    const getEpGracefulShutdownPromise = () => {
+      if (!ep.gracefulShutdownPromise) {
+        ep.gracefulShutdownPromise = new Promise((resolve) => {
+          // this resolver will be called when stt task received transcription.
+          ep.gracefulShutdownResolver = () => {
+            resolve();
+            ep.gracefulShutdownPromise = null;
+          };
+        });
+      }
+      return ep.gracefulShutdownPromise;
+    };
+
+    const gracefulShutdownHandler = async() => {
+      // resolve when one of the following happens:
+      // 1. stt task received transcription
+      // 2. JAMBONES_TRANSCRIBE_EP_DESTROY_DELAY_MS passed
+      await Promise.race([
+        getEpGracefulShutdownPromise(),
+        sleepFor(JAMBONES_TRANSCRIBE_EP_DESTROY_DELAY_MS)
+      ]);
+    };
+
+    const origStartTranscription = ep.startTranscription.bind(ep);
+    ep.startTranscription = async(...args) => {
+      try {
+        const result = await origStartTranscription(...args);
+        ep.isTranscribeActive = true;
+        return result;
+      } catch (err) {
+        ep.isTranscribeActive = false;
+        throw err;
+      }
+    };
+
+    const origStopTranscription = ep.stopTranscription.bind(ep);
+    ep.stopTranscription = async(opts = {}, ...args) => {
+      const { gracefulShutdown = true, ...others } = opts;
+      if (ep.isTranscribeActive && gracefulShutdown) {
+        // only wait for graceful shutdown if transcription is active
+        await gracefulShutdownHandler();
+      }
+      try {
+        const result = await origStopTranscription({...others}, ...args);
+        ep.isTranscribeActive = false;
+        return result;
+      } catch (err) {
+        ep.isTranscribeActive = false;
+        throw err;
+      }
+    };
+
+    const origDestroy = ep.destroy.bind(ep);
+    ep.destroy = async() => {
+      if (ep.isTranscribeActive) {
+        await gracefulShutdownHandler();
+      }
+      return await origDestroy();
+    };
+  }
+
+  return ep;
+};
+
+module.exports = {
+  createMediaEndpoint,
+};

lib/utils/place-outdial.js

@@ -16,17 +16,11 @@ const crypto = require('crypto');
 const HttpRequestor = require('./http-requestor');
 const WsRequestor = require('./ws-requestor');
 const {makeOpusFirst, removeVideoSdp} = require('./sdp-utils');
-const {
-  JAMBONES_USE_FREESWITCH_TIMER_FD,
-  JAMBONES_MEDIA_TIMEOUT_MS,
-  JAMBONES_MEDIA_HOLD_TIMEOUT_MS,
-  JAMBONES_TRANSCRIBE_EP_DESTROY_DELAY_MS
-} = require('../config');
-const { sleepFor } = require('./helpers');
+const { createMediaEndpoint } = require('./media-endpoint');
 
 class SingleDialer extends Emitter {
   constructor({logger, sbcAddress, target, opts, application, callInfo, accountInfo, rootSpan, startSpan, dialTask,
-    onHoldMusic}) {
+    onHoldMusic, tmpFiles}) {
     super();
     assert(target.type);
 
@@ -50,6 +44,7 @@ class SingleDialer extends Emitter {
     this.callSid = crypto.randomUUID();
     this.dialTask = dialTask;
     this.onHoldMusic = onHoldMusic;
+    this.tmpFiles = tmpFiles;
 
     this.on('callStatusChange', this._notifyCallStatusChange.bind(this));
   }
@@ -98,6 +93,7 @@ class SingleDialer extends Emitter {
       };
     }
     this.ms = ms;
+    this.srf = srf;
     let uri, to, inviteSpan;
     try {
       switch (this.target.type) {
@@ -139,8 +135,7 @@ class SingleDialer extends Emitter {
       this.updateCallStatus = srf.locals.dbHelpers.updateCallStatus;
       this.serviceUrl = srf.locals.serviceUrl;
 
-      this.ep = await ms.createEndpoint();
-      this._configMsEndpoint();
+      this.ep = await this._createMediaEndpoint();
       this.logger.debug(`SingleDialer:exec - created endpoint ${this.ep.uuid}`);
 
       /**
@@ -334,7 +329,13 @@ class SingleDialer extends Emitter {
    */
   async kill(Reason) {
     this.killed = true;
-    if (this.inviteInProgress) await this.inviteInProgress.cancel();
+    if (this.inviteInProgress) {
+      try {
+        await this.inviteInProgress.cancel();
+      } catch (err) {
+        this.logger.error({err}, 'SingleDialer:kill error cancelling invite');
+      }
+    }
     else if (this.dlg && this.dlg.connected) {
       const duration = moment().diff(this.dlg.connectTime, 'seconds');
       this.logger.debug('SingleDialer:kill hanging up called party');
@@ -352,43 +353,19 @@ class SingleDialer extends Emitter {
     }
   }
 
-  _configMsEndpoint() {
-    const opts = {
-      ...(this.onHoldMusic && {holdMusic: `shout://${this.onHoldMusic.replace(/^https?:\/\//, '')}`}),
-      ...(JAMBONES_USE_FREESWITCH_TIMER_FD && {timer_name: 'timerfd'}),
-      ...(JAMBONES_MEDIA_TIMEOUT_MS && {media_timeout: JAMBONES_MEDIA_TIMEOUT_MS}),
-      ...(JAMBONES_MEDIA_HOLD_TIMEOUT_MS && {media_hold_timeout: JAMBONES_MEDIA_HOLD_TIMEOUT_MS})
-    };
-    if (Object.keys(opts).length > 0) {
-      this.ep.set(opts);
-    }
-    if (this.dialTask?.inbandDtmfEnabled && !this.ep.inbandDtmfEnabled) {
-      // https://developer.signalwire.com/freeswitch/FreeSWITCH-Explained/Modules/mod-dptools/6587132/#0-about
-      try {
-        this.ep.execute('start_dtmf');
-        this.ep.inbandDtmfEnabled = true;
-      } catch (err) {
-        this.logger.info(err, 'place-outdial:_configMsEndpoint - error enable inband DTMF');
-      }
-    }
+  async _handleMediaTimeout(evt, ep) {
+    this.logger.info({evt}, 'SingleDialer:_handleMediaTimeout - media timeout event received');
+    this.dialTask.kill(this.dialTask.cs, 'media-timeout');
+  }
 
-    const origDestroy = this.ep.destroy.bind(this.ep);
-    this.ep.destroy = async() => {
-      try {
-        if (this.dialTask.transcribeTask && JAMBONES_TRANSCRIBE_EP_DESTROY_DELAY_MS) {
-          // transcribe task is being used, wait for some time before destroy
-          // if final transcription is received but endpoint is already closed,
-          // freeswitch module will not be able to send the transcription
-
-          this.logger.info('SingleDialer:_configMsEndpoint -' +
-            ' Dial with transcribe task, wait for some time before destroy');
-          await sleepFor(JAMBONES_TRANSCRIBE_EP_DESTROY_DELAY_MS);
-        }
-        await origDestroy();
-      } catch (err) {
-        this.logger.error(err, 'SingleDialer:_configMsEndpoint - error destroying endpoint');
-      }
-    };
+  async _createMediaEndpoint(drachtioFsmrfOptions = {}) {
+    return await createMediaEndpoint(this.srf, this.logger, {
+      acactiveMs: this.ms,
+      drachtioFsmrfOptions,
+      onHoldMusic: this.onHoldMusic,
+      inbandDtmfEnabled: this.dialTask?.inbandDtmfEnabled,
+      mediaTimeoutHandler: this._handleMediaTimeout.bind(this),
+    });
   }
 
   /**
@@ -431,7 +408,8 @@ class SingleDialer extends Emitter {
         accountInfo: this.accountInfo,
         tasks,
         rootSpan: this.rootSpan,
-        req: this.req
+        req: this.req,
+        tmpFiles: this.tmpFiles,
       });
       await cs.exec();
 
@@ -528,8 +506,7 @@ class SingleDialer extends Emitter {
     assert(this.dlg && this.dlg.connected && !this.ep);
 
     this.logger.debug('SingleDialer:reAnchorMedia: re-anchoring media after partial media');
-    this.ep = await this.ms.createEndpoint({remoteSdp: this.dlg.remote.sdp});
-    this._configMsEndpoint();
+    this.ep = await this._createMediaEndpoint({remoteSdp: this.dlg.remote.sdp});
     await this.dlg.modify(this.ep.local.sdp, {
       headers: {
         'X-Reason': 'anchor-media'
@@ -566,12 +543,12 @@ class SingleDialer extends Emitter {
 
 function placeOutdial({
   logger, srf, ms, sbcAddress, target, opts, application, callInfo, accountInfo, rootSpan, startSpan, dialTask,
-  onHoldMusic
+  onHoldMusic, tmpFiles
 }) {
   const myOpts = deepcopy(opts);
   const sd = new SingleDialer({
     logger, sbcAddress, target, opts: myOpts, application, callInfo,
-    accountInfo, rootSpan, startSpan, dialTask, onHoldMusic
+    accountInfo, rootSpan, startSpan, dialTask, onHoldMusic, tmpFiles
   });
   sd.exec(srf, ms, myOpts);
   return sd;

lib/utils/process-monitor.js

@@ -0,0 +1,91 @@
+// lib/utils/process-monitor.js
+const fs = require('fs');
+const path = require('path');
+
+class ProcessMonitor {
+  constructor(logger) {
+    this.logger = logger;
+    this.packageInfo = this.getPackageInfo();
+    this.processName = this.packageInfo.name || 'unknown-app';
+  }
+
+  getPackageInfo() {
+    try {
+      const packagePath = path.join(process.cwd(), 'package.json');
+      return JSON.parse(fs.readFileSync(packagePath, 'utf8'));
+    } catch (e) {
+      return { name: 'unknown', version: 'unknown' };
+    }
+  }
+
+  logStartup(additionalInfo = {}) {
+    const startupInfo = {
+      msg: `${this.processName} started`,
+      app_name: this.processName,
+      app_version: this.packageInfo.version,
+      pid: process.pid,
+      ppid: process.ppid,
+      pm2_instance_id: process.env.NODE_APP_INSTANCE || 'not_pm2',
+      pm2_id: process.env.pm_id,
+      is_pm2: !!process.env.PM2,
+      node_version: process.version,
+      uptime: process.uptime(),
+      timestamp: new Date().toISOString(),
+      ...additionalInfo
+    };
+
+    this.logger.info(startupInfo);
+    return startupInfo;
+  }
+
+  setupSignalHandlers() {
+    // Log when we receive signals that would cause restart
+    process.on('SIGINT', () => {
+      this.logger.info({
+        msg: 'SIGINT received',
+        app_name: this.processName,
+        pid: process.pid,
+        ppid: process.ppid,
+        uptime: process.uptime(),
+        timestamp: new Date().toISOString()
+      });
+      process.exit(0);
+    });
+
+    process.on('SIGTERM', () => {
+      this.logger.info({
+        msg: 'SIGTERM received',
+        app_name: this.processName,
+        pid: process.pid,
+        ppid: process.ppid,
+        uptime: process.uptime(),
+        timestamp: new Date().toISOString()
+      });
+      process.exit(0);
+    });
+
+    process.on('uncaughtException', (error) => {
+      this.logger.error({
+        msg: 'Uncaught exception - process will restart',
+        app_name: this.processName,
+        error: error.message,
+        stack: error.stack,
+        pid: process.pid,
+        timestamp: new Date().toISOString()
+      });
+      process.exit(1);
+    });
+
+    process.on('unhandledRejection', (reason, promise) => {
+      this.logger.error({
+        msg: 'Unhandled rejection',
+        app_name: this.processName,
+        reason,
+        pid: process.pid,
+        timestamp: new Date().toISOString()
+      });
+    });
+  }
+}
+
+module.exports = ProcessMonitor;

lib/utils/sbc-pinger.js

@@ -100,6 +100,39 @@ module.exports = (logger) => {
   else if (K8S) {
     lifecycleEmitter.scaleIn = () => process.exit(0);
   }
+  else {
+    process.on('SIGUSR1', () => {
+      logger.info('received SIGUSR1: begin drying up calls for scale-in');
+      dryUpCalls = true;
+
+      const {srf} = require('../..');
+      const {writeSystemAlerts} = srf.locals;
+      if (writeSystemAlerts) {
+        const {SystemState, FEATURE_SERVER} = require('./constants');
+        writeSystemAlerts({
+          system_component: FEATURE_SERVER,
+          state : SystemState.GracefulShutdownInProgress,
+          fields : {
+            detail: `feature-server with process_id ${process.pid} shutdown in progress`,
+            host: srf.locals?.ipv4
+          }
+        });
+      }
+      pingProxies(srf);
+
+      // if we have zero calls, we can complete the scale-in right
+      setTimeout(() => {
+        const calls = srf.locals.sessionTracker.count;
+        if (calls === 0) {
+          logger.info('scale-in can complete immediately as we have no calls in progress');
+          process.exit(0);
+        }
+        else {
+          logger.info(`${calls} calls in progress; scale-in will complete when they are done`);
+        }
+      }, 5000);
+    });
+  }
 
 
   async function pingProxies(srf) {

lib/utils/sdp-utils.js

@@ -55,11 +55,28 @@ const extractSdpMedia = (sdp) => {
   }
 };
 
+const getLeadingCodec = (sdp) => {
+  if (!sdp) {
+    return null;
+  }
+
+  const parsed = sdpTransform.parse(sdp);
+  const audio = parsed.media?.find((m) => m.type === 'audio');
+
+  if (!audio) {
+    return null;
+  }
+
+  return audio.rtp?.[0]?.codec || null;
+};
+
+
 module.exports = {
   isOnhold,
   mergeSdpMedia,
   extractSdpMedia,
   isOpusFirst,
   makeOpusFirst,
-  removeVideoSdp
+  removeVideoSdp,
+  getLeadingCodec
 };

lib/utils/stt-latency-calculator.js

@@ -0,0 +1,196 @@
+const { assert } = require('console');
+const Emitter = require('events');
+const {
+  VadDetection,
+  SileroVadDetection
+} = require('../utils/constants.json');
+
+class SttLatencyCalculator extends Emitter {
+  constructor({ logger, cs}) {
+    super();
+    this.logger = logger;
+    this.cs = cs;
+    this.isRunning = false;
+    this.isInTalkSpurt = false;
+    this.start_talking_time = 0;
+    this.talkspurts = [];
+    this.vendor = this.cs.vad?.vendor || 'silero';
+    this.stt_start_time = 0;
+    this.stt_stop_time = 0;
+    this.stt_on_transcription_time = 0;
+  }
+
+  set sttStartTime(time) {
+    this.stt_start_time = time;
+  }
+
+  get sttStartTime() {
+    return this.stt_start_time || 0;
+  }
+
+  set sttStopTime(time) {
+    this.stt_stop_time = time;
+  }
+
+  get sttStopTime() {
+    return this.stt_stop_time || 0;
+  }
+
+  set sttOnTranscriptionTime(time) {
+    this.stt_on_transcription_time = time;
+  }
+
+  get sttOnTranscriptionTime() {
+    return this.stt_on_transcription_time || 0;
+  }
+
+  _onVadDetected(_ep, _evt, fsEvent) {
+    if (fsEvent.getHeader('detected-event') === 'stop_talking') {
+      if (this.isInTalkSpurt) {
+        this.talkspurts.push({
+          start: this.start_talking_time,
+          stop: Date.now()
+        });
+      }
+
+      this.start_talking_time = 0;
+      this.isInTalkSpurt = false;
+    } else if (fsEvent.getHeader('detected-event') === 'start_talking') {
+      this.start_talking_time = Date.now();
+      this.isInTalkSpurt = true;
+    }
+  }
+
+  _startVad() {
+    assert(!this.isRunning, 'Latency calculator is already running');
+    assert(this.cs.ep, 'Callsession has no endpoint to start the latency calculator');
+    const ep = this.cs.ep;
+    if (!ep.sttLatencyVadHandler) {
+      ep.sttLatencyVadHandler = this._onVadDetected.bind(this, ep);
+      if (this.vendor === 'silero') {
+        ep.addCustomEventListener(SileroVadDetection.Detection, ep.sttLatencyVadHandler);
+      } else {
+        ep.addCustomEventListener(VadDetection.Detection, ep.sttLatencyVadHandler);
+      }
+    }
+    this.stop_talking_time = 0;
+    this.start_talking_time = 0;
+    this.vad = {
+      ...(this.cs.vad || {}),
+      strategy: 'continuous',
+      bugname: 'stt-latency-calculator-vad',
+      vendor: this.vendor
+    };
+
+    ep.startVadDetection(this.vad);
+    this.isRunning = true;
+  }
+
+  _stopVad() {
+    if (this.isRunning) {
+      this.logger.warn('Latency calculator is still running, stopping VAD detection');
+      const ep = this.cs.ep;
+      ep.stopVadDetection(this.vad);
+      if (ep.sttLatencyVadHandler) {
+        if (this.vendor === 'silero') {
+          this.ep?.removeCustomEventListener(SileroVadDetection.Detection, ep.sttLatencyVadHandler);
+        } else {
+          this.ep?.removeCustomEventListener(VadDetection.Detection, ep.sttLatencyVadHandler);
+        }
+        ep.sttLatencyVadHandler = null;
+      }
+      this.isRunning = false;
+      this.logger.info('STT Latency Calculator stopped');
+    }
+  }
+
+  start() {
+    if (this.isRunning) {
+      this.logger.warn('Latency calculator is already running');
+      return;
+    }
+    if (!this.cs.ep) {
+      this.logger.error('Callsession has no endpoint to start the latency calculator');
+      return;
+    }
+    this._startVad();
+    this.logger.debug('STT Latency Calculator started');
+  }
+
+  stop() {
+    this._stopVad();
+  }
+
+  toUnixTimestamp(date) {
+    return Math.floor(date / 1000);
+  }
+
+  calculateLatency() {
+    if (!this.isRunning) {
+      return null;
+    }
+
+    const stt_stop_time = this.stt_stop_time || Date.now();
+    if (this.isInTalkSpurt) {
+      this.talkspurts.push({
+        start: this.start_talking_time,
+        stop: stt_stop_time
+      });
+      this.isInTalkSpurt = false;
+      this.start_talking_time = 0;
+    }
+    const stt_on_transcription_time = this.stt_on_transcription_time || stt_stop_time;
+    const start_talking_time = this.talkspurts[0]?.start;
+    let lastIdx = this.talkspurts.length - 1;
+    lastIdx = lastIdx < 0 ? 0 : lastIdx;
+    const stop_talking_time = this.talkspurts[lastIdx]?.stop || stt_stop_time;
+
+    return {
+      stt_start_time: this.toUnixTimestamp(this.stt_start_time),
+      stt_stop_time: this.toUnixTimestamp(stt_stop_time),
+      start_talking_time: this.toUnixTimestamp(start_talking_time),
+      stop_talking_time: this.toUnixTimestamp(stop_talking_time),
+      stt_latency: parseFloat((Math.abs(stt_on_transcription_time - stop_talking_time)) / 1000).toFixed(2),
+      stt_latency_ms: Math.abs(stt_on_transcription_time - stop_talking_time),
+      stt_usage: parseFloat((stt_stop_time - this.stt_start_time) / 1000).toFixed(2),
+      talkspurts: this.talkspurts.map((ts) =>
+        ([this.toUnixTimestamp(ts.start || 0), this.toUnixTimestamp(ts.stop || 0)]))
+    };
+  }
+
+  resetTime() {
+    if (!this.isRunning) {
+      return;
+    }
+    this.stt_start_time = Date.now();
+    this.stt_stop_time = 0;
+    this.stt_on_transcription_time = 0;
+    this.clearTalkspurts();
+    this.logger.info('STT Latency Calculator reset');
+  }
+
+  onTranscriptionReceived() {
+    if (!this.isRunning) {
+      return;
+    }
+    this.stt_on_transcription_time = Date.now();
+    this.logger.debug(`CallSession:on-transcription set to ${this.stt_on_transcription_time}`);
+  }
+
+  onTranscribeStop() {
+    if (!this.isRunning) {
+      return;
+    }
+    this.stt_stop_time = Date.now();
+    this.logger.debug(`CallSession:transcribe-stop set to ${this.stt_stop_time}`);
+  }
+
+  clearTalkspurts() {
+    this.talkspurts = [];
+    if (!this.isInTalkSpurt) {
+      this.start_talking_time = 0;
+    }
+  }
+}
+
+module.exports = SttLatencyCalculator;

lib/utils/transcription-utils.js

@@ -110,6 +110,10 @@ const stickyVars = {
   voxist: [
     'VOXIST_API_KEY',
   ],
+  cartesia: [
+    'CARTESIA_API_KEY',
+    'CARTESIA_MODEL_ID'
+  ],
   speechmatics: [
     'SPEECHMATICS_API_KEY',
     'SPEECHMATICS_HOST',
@@ -127,6 +131,43 @@ const stickyVars = {
     'OPENAI_TURN_DETECTION_PREFIX_PADDING_MS',
     'OPENAI_TURN_DETECTION_SILENCE_DURATION_MS',
   ],
+  houndify: [
+    'HOUNDIFY_CLIENT_ID',
+    'HOUNDIFY_CLIENT_KEY',
+    'HOUNDIFY_USER_ID',
+    'HOUNDIFY_MAX_SILENCE_SECONDS',
+    'HOUNDIFY_MAX_SILENCE_AFTER_FULL_QUERY_SECONDS',
+    'HOUNDIFY_MAX_SILENCE_AFTER_PARTIAL_QUERY_SECONDS',
+    'HOUNDIFY_VAD_SENSITIVITY',
+    'HOUNDIFY_VAD_TIMEOUT',
+    'HOUNDIFY_VAD_MODE',
+    'HOUNDIFY_VAD_VOICE_MS',
+    'HOUNDIFY_VAD_SILENCE_MS',
+    'HOUNDIFY_VAD_DEBUG',
+    'HOUNDIFY_AUDIO_FORMAT',
+    'HOUNDIFY_ENABLE_NOISE_REDUCTION',
+    'HOUNDIFY_AUDIO_ENDPOINT',
+    'HOUNDIFY_ENABLE_PROFANITY_FILTER',
+    'HOUNDIFY_ENABLE_PUNCTUATION',
+    'HOUNDIFY_ENABLE_CAPITALIZATION',
+    'HOUNDIFY_CONFIDENCE_THRESHOLD',
+    'HOUNDIFY_ENABLE_DISFLUENCY_FILTER',
+    'HOUNDIFY_MAX_RESULTS',
+    'HOUNDIFY_ENABLE_WORD_TIMESTAMPS',
+    'HOUNDIFY_MAX_ALTERNATIVES',
+    'HOUNDIFY_PARTIAL_TRANSCRIPT_INTERVAL',
+    'HOUNDIFY_SESSION_TIMEOUT',
+    'HOUNDIFY_CONNECTION_TIMEOUT',
+    'HOUNDIFY_LATITUDE',
+    'HOUNDIFY_LONGITUDE',
+    'HOUNDIFY_CITY',
+    'HOUNDIFY_STATE',
+    'HOUNDIFY_COUNTRY',
+    'HOUNDIFY_TIMEZONE',
+    'HOUNDIFY_DOMAIN',
+    'HOUNDIFY_CUSTOM_VOCABULARY',
+    'HOUNDIFY_LANGUAGE_MODEL'
+  ],
 };
 
 /**
@@ -335,6 +376,61 @@ const normalizeDeepgram = (evt, channel, language, shortUtterance) => {
   };
 };
 
+const normalizeGladia = (evt, channel, language, shortUtterance) => {
+  const copy = JSON.parse(JSON.stringify(evt));
+
+  // Handle Gladia transcript format
+  if (evt.type === 'transcript' && evt.data && evt.data.utterance) {
+    const utterance = evt.data.utterance;
+    const alternatives = [{
+      confidence: utterance.confidence || 0,
+      transcript: utterance.text || '',
+    }];
+
+    return {
+      language_code: utterance.language || language,
+      channel_tag: channel,
+      is_final: evt.data.is_final || false,
+      alternatives,
+      vendor: {
+        name: 'gladia',
+        evt: copy
+      }
+    };
+  }
+};
+
+const normalizeDeepgramFlux = (evt, channel, language) => {
+  const copy = JSON.parse(JSON.stringify(evt));
+
+  let turnTakingEvent;
+  if (['StartOfTurn', 'EagerEndOfTurn', 'TurnResumed', 'EndOfTurn'].includes(evt.event)) {
+    turnTakingEvent = evt.event;
+  }
+
+  /* calculate total confidence based on word-level confidence */
+  const realWords = (evt.words || [])
+    .filter((w) => ![',.!?;'].includes(w.word));
+  const confidence = realWords.length > 0 ? realWords.reduce((acc, w) => acc + w.confidence, 0) / realWords.length : 0;
+  return {
+    language_code: language,
+    channel_tag: channel,
+    is_final: evt.event === 'EndOfTurn',
+    alternatives: [
+      {
+        confidence,
+        end_of_turn_confidence: evt.end_of_turn_confidence,
+        transcript: evt.transcript,
+        ...(turnTakingEvent && {turn_taking_event: turnTakingEvent})
+      }
+    ],
+    vendor: {
+      name: 'deepgramflux',
+      evt: copy
+    }
+  };
+};
+
 const normalizeNvidia = (evt, channel, language) => {
   const copy = JSON.parse(JSON.stringify(evt));
   const alternatives = (evt.alternatives || [])
@@ -519,16 +615,27 @@ const normalizeAws = (evt, channel, language) => {
 
 const normalizeAssemblyAi = (evt, channel, language) => {
   const copy = JSON.parse(JSON.stringify(evt));
+  const alternatives = [];
+  let is_final = false;
+  if (evt.type && evt.type === 'Turn') {
+    // v3 is here
+    alternatives.push({
+      confidence: evt.end_of_turn_confidence,
+      transcript: evt.transcript,
+    });
+    is_final = evt.end_of_turn;
+  } else {
+    alternatives.push({
+      confidence: evt.confidence,
+      transcript: evt.text,
+    });
+    is_final = evt.message_type === 'FinalTranscript';
+  }
   return {
     language_code: language,
     channel_tag: channel,
-    is_final: evt.message_type === 'FinalTranscript',
-    alternatives: [
-      {
-        confidence: evt.confidence,
-        transcript: evt.text,
-      }
-    ],
+    is_final,
+    alternatives,
     vendor: {
       name: 'assemblyai',
       evt: copy
@@ -536,6 +643,30 @@ const normalizeAssemblyAi = (evt, channel, language) => {
   };
 };
 
+const normalizeHoundify = (evt, channel, language) => {
+  const copy = JSON.parse(JSON.stringify(evt));
+  const alternatives = [];
+  const is_final = evt.ResultsAreFinal && evt.ResultsAreFinal[0] === true;
+  if (evt.Disambiguation && evt.Disambiguation.ChoiceData && evt.Disambiguation.ChoiceData.length > 0) {
+    // Handle Houndify Voice Search Result format
+    const choiceData = evt.Disambiguation.ChoiceData[0];
+    alternatives.push({
+      confidence: choiceData.ConfidenceScore || choiceData.ASRConfidence || 0.0,
+      transcript: choiceData.FormattedTranscription || choiceData.Transcription || '',
+    });
+  }
+  return {
+    language_code: language,
+    channel_tag: channel,
+    is_final,
+    alternatives,
+    vendor: {
+      name: 'houndify',
+      evt: copy
+    }
+  };
+};
+
 const normalizeVoxist = (evt, channel, language) => {
   const copy = JSON.parse(JSON.stringify(evt));
   return {
@@ -555,6 +686,25 @@ const normalizeVoxist = (evt, channel, language) => {
   };
 };
 
+const normalizeCartesia = (evt, channel, language) => {
+  const copy = JSON.parse(JSON.stringify(evt));
+  return {
+    language_code: language,
+    channel_tag: channel,
+    is_final: evt.is_final,
+    alternatives: [
+      {
+        confidence: 1.00,
+        transcript: evt.text,
+      }
+    ],
+    vendor: {
+      name: 'cartesia',
+      evt: copy
+    }
+  };
+};
+
 const normalizeSpeechmatics = (evt, channel, language) => {
   const copy = JSON.parse(JSON.stringify(evt));
   const is_final = evt.message === 'AddTranscript';
@@ -616,6 +766,10 @@ module.exports = (logger) => {
     switch (vendor) {
       case 'deepgram':
         return normalizeDeepgram(evt, channel, language, shortUtterance);
+      case 'gladia':
+        return normalizeGladia(evt, channel, language, shortUtterance);
+      case 'deepgramflux':
+        return normalizeDeepgramFlux(evt, channel, language, shortUtterance);
       case 'microsoft':
         return normalizeMicrosoft(evt, channel, language, punctuation);
       case 'google':
@@ -634,8 +788,12 @@ module.exports = (logger) => {
         return normalizeCobalt(evt, channel, language);
       case 'assemblyai':
         return normalizeAssemblyAi(evt, channel, language, shortUtterance);
+      case 'houndify':
+        return normalizeHoundify(evt, channel, language, shortUtterance);
       case 'voxist':
         return normalizeVoxist(evt, channel, language);
+      case 'cartesia':
+        return normalizeCartesia(evt, channel, language);
       case 'verbio':
         return normalizeVerbio(evt, channel, language);
       case 'speechmatics':
@@ -728,12 +886,15 @@ module.exports = (logger) => {
           AWS_ACCESS_KEY_ID: sttCredentials.accessKeyId,
           AWS_SECRET_ACCESS_KEY: sttCredentials.secretAccessKey,
           AWS_REGION: sttCredentials.region,
-          AWS_SECURITY_TOKEN: sttCredentials.securityToken
+          AWS_SECURITY_TOKEN: sttCredentials.securityToken,
+          AWS_SESSION_TOKEN: sttCredentials.sessionToken ? sttCredentials.sessionToken : sttCredentials.securityToken
         }),
         ...(awsOptions.accessKey && {AWS_ACCESS_KEY_ID: awsOptions.accessKey}),
         ...(awsOptions.secretKey && {AWS_SECRET_ACCESS_KEY: awsOptions.secretKey}),
         ...(awsOptions.region && {AWS_REGION: awsOptions.region}),
         ...(awsOptions.securityToken && {AWS_SECURITY_TOKEN: awsOptions.securityToken}),
+        ...(awsOptions.sessionToken && {AWS_SESSION_TOKEN: awsOptions.sessionToken ?
+          awsOptions.sessionToken : awsOptions.securityToken}),
         ...(awsOptions.languageModelName && {AWS_LANGUAGE_MODEL_NAME: awsOptions.languageModelName}),
         ...(awsOptions.piiEntityTypes?.length && {AWS_PII_ENTITY_TYPES: awsOptions.piiEntityTypes.join(',')}),
         ...(awsOptions.piiIdentifyEntities && {AWS_PII_IDENTIFY_ENTITIES: true}),
@@ -759,7 +920,7 @@ module.exports = (logger) => {
         ...(rOpts.initialSpeechTimeoutMs > 0 &&
           {AZURE_INITIAL_SPEECH_TIMEOUT_MS: rOpts.initialSpeechTimeoutMs}),
         ...(rOpts.requestSnr && {AZURE_REQUEST_SNR: 1}),
-        ...(rOpts.audioLogging && {AZURE_AUDIO_LOGGING: 1}),
+        ...(azureOptions.audioLogging && {AZURE_AUDIO_LOGGING: 1}),
         ...{AZURE_USE_OUTPUT_FORMAT_DETAILED: 1},
         ...(azureOptions.speechSegmentationSilenceTimeoutMs &&
           {AZURE_SPEECH_SEGMENTATION_SILENCE_TIMEOUT_MS: azureOptions.speechSegmentationSilenceTimeoutMs}),
@@ -834,6 +995,14 @@ module.exports = (logger) => {
       const deepgramUri = deepgramOptions.deepgramSttUri || sttCredentials.deepgram_stt_uri;
       const useTls = deepgramOptions.deepgramSttUseTls || sttCredentials.deepgram_stt_use_tls;
 
+      // DH (2025-08-11) entity_prompt is currently limited to 100 words
+      const entityPrompt = deepgramOptions.entityPrompt ?
+        deepgramOptions.entityPrompt
+          .split(/\s+/)
+          .slice(0, 100)
+          .join(' ')
+        : undefined;
+
       /* default to a sensible model if not supplied */
       if (!model) {
         model = selectDefaultDeepgramModel(task, language);
@@ -892,7 +1061,35 @@ module.exports = (logger) => {
         ...(deepgramOptions.fillerWords) &&
           {DEEPGRAM_SPEECH_FILLER_WORDS: deepgramOptions.fillerWords},
         ...((Array.isArray(deepgramOptions.keyterms) && deepgramOptions.keyterms.length > 0) &&
-          {DEEPGRAM_SPEECH_KEYTERMS: deepgramOptions.keyterms.join(',')})
+          {DEEPGRAM_SPEECH_KEYTERMS: deepgramOptions.keyterms.join(',')}),
+        ...(deepgramOptions.mipOptOut && {DEEPGRAM_SPEECH_MIP_OPT_OUT: deepgramOptions.mipOptOut}),
+        ...(entityPrompt && {DEEPGRAM_SPEECH_ENTITY_PROMPT: entityPrompt}),
+      };
+    }
+    else if ('deepgramflux' === vendor) {
+      const {
+        eotThreshold,
+        eotTimeoutMs,
+        mipOptOut,
+        model,
+        eagerEotThreshold,
+        keyterms
+      } = rOpts.deepgramOptions || {};
+      opts = {
+        DEEPGRAMFLUX_API_KEY: sttCredentials.api_key,
+        DEEPGRAMFLUX_SPEECH_MODEL: model || 'flux-general-en',
+        ...(eotThreshold && {DEEPGRAMFLUX_SPEECH_EOT_THRESHOLD: eotThreshold}),
+        ...(eotTimeoutMs && {DEEPGRAMFLUX_SPEECH_EOT_TIMEOUT_MS: eotTimeoutMs}),
+        ...(mipOptOut && {DEEPGRAMFLUX_SPEECH_MIP_OPT_OUT: mipOptOut}),
+        ...(eagerEotThreshold && {DEEPGRAMFLUX_SPEECH_EAGER_EOT_THRESHOLD: eagerEotThreshold}),
+        ...(keyterms && keyterms.length > 0 && {DEEPGRAMFLUX_SPEECH_KEYTERMS: keyterms.join(',')}),
+      };
+    }
+    else if ('gladia' === vendor) {
+      const {host, path} = sttCredentials;
+      opts = {
+        GLADIA_SPEECH_HOST: host,
+        GLADIA_SPEECH_PATH: path,
       };
     }
     else if ('soniox' === vendor) {
@@ -993,14 +1190,89 @@ module.exports = (logger) => {
       };
     }
     else if ('assemblyai' === vendor) {
+      const serviceVersion = rOpts.assemblyAiOptions?.serviceVersion || sttCredentials.service_version || 'v2';
+      const {
+        formatTurns,
+        endOfTurnConfidenceThreshold,
+        minEndOfTurnSilenceWhenConfident,
+        maxTurnSilence
+      } = rOpts.assemblyAiOptions || {};
       opts = {
         ...opts,
+        ASSEMBLYAI_API_VERSION: serviceVersion,
+        ...(serviceVersion === 'v3' && {
+          ...(formatTurns && {
+            ASSEMBLYAI_FORMAT_TURNS: formatTurns
+          }),
+          ...(endOfTurnConfidenceThreshold && {
+            ASSEMBLYAI_END_OF_TURN_CONFIDENCE_THRESHOLD: endOfTurnConfidenceThreshold
+          }),
+          ASSEMBLYAI_MIN_END_OF_TURN_SILENCE_WHEN_CONFIDENT: minEndOfTurnSilenceWhenConfident || 500,
+          ...(maxTurnSilence && {
+            ASSEMBLYAI_MAX_TURN_SILENCE: maxTurnSilence
+          }),
+        }),
         ...(sttCredentials.api_key) &&
           {ASSEMBLYAI_API_KEY: sttCredentials.api_key},
         ...(rOpts.hints?.length > 0 &&
           {ASSEMBLYAI_WORD_BOOST: JSON.stringify(rOpts.hints)})
       };
     }
+    else if ('houndify' === vendor) {
+      const {
+        latitude, longitude, city, state, country, timeZone, domain, audioEndpoint,
+        maxSilenceSeconds, maxSilenceAfterFullQuerySeconds, maxSilenceAfterPartialQuerySeconds,
+        vadSensitivity, vadTimeout, vadMode, vadVoiceMs, vadSilenceMs, vadDebug,
+        audioFormat, enableNoiseReduction, enableProfanityFilter, enablePunctuation,
+        enableCapitalization, confidenceThreshold, enableDisfluencyFilter,
+        maxResults, enableWordTimestamps, maxAlternatives, partialTranscriptInterval,
+        sessionTimeout, connectionTimeout, customVocabulary, languageModel,
+        requestInfo, sampleRate
+      } = rOpts.houndifyOptions || {};
+      const audioEndpointUri = audioEndpoint || sttCredentials.houndify_server_uri;
+
+      opts = {
+        ...opts,
+        HOUNDIFY_CLIENT_ID: sttCredentials.client_id,
+        HOUNDIFY_CLIENT_KEY: sttCredentials.client_key,
+        HOUNDIFY_USER_ID: sttCredentials.user_id,
+        HOUNDIFY_MAX_SILENCE_SECONDS: maxSilenceSeconds || 5,
+        HOUNDIFY_MAX_SILENCE_AFTER_FULL_QUERY_SECONDS: maxSilenceAfterFullQuerySeconds || 1,
+        HOUNDIFY_MAX_SILENCE_AFTER_PARTIAL_QUERY_SECONDS: maxSilenceAfterPartialQuerySeconds || 1.5,
+        ...(vadSensitivity && {HOUNDIFY_VAD_SENSITIVITY: vadSensitivity}),
+        ...(vadTimeout && {HOUNDIFY_VAD_TIMEOUT: vadTimeout}),
+        ...(vadMode && {HOUNDIFY_VAD_MODE: vadMode}),
+        ...(vadVoiceMs && {HOUNDIFY_VAD_VOICE_MS: vadVoiceMs}),
+        ...(vadSilenceMs && {HOUNDIFY_VAD_SILENCE_MS: vadSilenceMs}),
+        ...(vadDebug && {HOUNDIFY_VAD_DEBUG: vadDebug}),
+        ...(audioFormat && {HOUNDIFY_AUDIO_FORMAT: audioFormat}),
+        ...(enableNoiseReduction && {HOUNDIFY_ENABLE_NOISE_REDUCTION: enableNoiseReduction}),
+        ...(enableProfanityFilter && {HOUNDIFY_ENABLE_PROFANITY_FILTER: enableProfanityFilter}),
+        ...(enablePunctuation && {HOUNDIFY_ENABLE_PUNCTUATION: enablePunctuation}),
+        ...(enableCapitalization && {HOUNDIFY_ENABLE_CAPITALIZATION: enableCapitalization}),
+        ...(confidenceThreshold && {HOUNDIFY_CONFIDENCE_THRESHOLD: confidenceThreshold}),
+        ...(enableDisfluencyFilter && {HOUNDIFY_ENABLE_DISFLUENCY_FILTER: enableDisfluencyFilter}),
+        ...(maxResults && {HOUNDIFY_MAX_RESULTS: maxResults}),
+        ...(enableWordTimestamps && {HOUNDIFY_ENABLE_WORD_TIMESTAMPS: enableWordTimestamps}),
+        ...(maxAlternatives && {HOUNDIFY_MAX_ALTERNATIVES: maxAlternatives}),
+        ...(partialTranscriptInterval && {HOUNDIFY_PARTIAL_TRANSCRIPT_INTERVAL: partialTranscriptInterval}),
+        ...(sessionTimeout && {HOUNDIFY_SESSION_TIMEOUT: sessionTimeout}),
+        ...(connectionTimeout && {HOUNDIFY_CONNECTION_TIMEOUT: connectionTimeout}),
+        ...(latitude && {HOUNDIFY_LATITUDE: latitude}),
+        ...(longitude && {HOUNDIFY_LONGITUDE: longitude}),
+        ...(city && {HOUNDIFY_CITY: city}),
+        ...(state && {HOUNDIFY_STATE: state}),
+        ...(country && {HOUNDIFY_COUNTRY: country}),
+        ...(timeZone && {HOUNDIFY_TIMEZONE: timeZone}),
+        ...(domain && {HOUNDIFY_DOMAIN: domain}),
+        ...(audioEndpointUri && {HOUNDIFY_AUDIO_ENDPOINT: audioEndpointUri}),
+        ...(customVocabulary && {HOUNDIFY_CUSTOM_VOCABULARY:
+          Array.isArray(customVocabulary) ? customVocabulary.join(',') : customVocabulary}),
+        ...(languageModel && {HOUNDIFY_LANGUAGE_MODEL: languageModel}),
+        ...(requestInfo && {HOUNDIFY_REQUEST_INFO: JSON.stringify(requestInfo)}),
+        ...(sampleRate && {HOUNDIFY_SAMPLING_RATE: sampleRate}),
+      };
+    }
     else if ('voxist' === vendor) {
       opts = {
         ...opts,
@@ -1008,6 +1280,16 @@ module.exports = (logger) => {
           {VOXIST_API_KEY: sttCredentials.api_key},
       };
     }
+    else if ('cartesia' === vendor) {
+      opts = {
+        ...opts,
+        ...(sttCredentials.api_key &&
+          {CARTESIA_API_KEY: sttCredentials.api_key}),
+        ...(sttCredentials.stt_model_id && {
+          CARTESIA_MODEL_ID: sttCredentials.stt_model_id
+        })
+      };
+    }
     else if ('openai' === vendor) {
       const {openaiOptions = {}} = rOpts;
       const model = openaiOptions.model || rOpts.model || sttCredentials.model_id || 'whisper-1';

lib/utils/tts-streaming-buffer.js

@@ -80,7 +80,7 @@ class TtsStreamingBuffer extends Emitter {
     clearTimeout(this.timer);
     this.removeCustomEventListeners();
     if (this.ep) {
-      this._api(this.ep, [this.ep.uuid, 'close'])
+      this._api(this.ep, [this.ep.uuid, 'stop'])
         .catch((err) =>
           this.logger.info({ err }, 'TtsStreamingBuffer:stop Error closing TTS streaming')
         );
@@ -163,7 +163,6 @@ class TtsStreamingBuffer extends Emitter {
   }
 
   clear() {
-    this.logger.debug('TtsStreamingBuffer:clear');
     if (this._connectionStatus !== TtsStreamingConnectionStatus.Connected) return;
     clearTimeout(this.timer);
     this._api(this.ep, [this.ep.uuid, 'clear']).catch((err) =>
@@ -193,10 +192,7 @@ class TtsStreamingBuffer extends Emitter {
         this.logger.debug('TtsStreamingBuffer:_feedQueue TTS stream is not open or no endpoint available');
         return;
       }
-      if (
-        this._connectionStatus === TtsStreamingConnectionStatus.NotConnected ||
-        this._connectionStatus === TtsStreamingConnectionStatus.Failed
-      ) {
+      if (this._connectionStatus !== TtsStreamingConnectionStatus.Connected) {
         this.logger.debug('TtsStreamingBuffer:_feedQueue TTS stream is not connected');
         return;
       }
@@ -224,7 +220,8 @@ class TtsStreamingBuffer extends Emitter {
           this.queue.shift();
         }
         // Immediately send all accumulated text (ignoring sentence boundaries).
-        if (flushText.length > 0) {
+        // Skip sending if flushText is only whitespace.
+        if (flushText.length > 0 && !isWhitespace(flushText)) {
           const modifiedFlushText = flushText.replace(/\n\n/g, '\n \n');
           try {
             await this._api(this.ep, [this.ep.uuid, 'send', modifiedFlushText]);
@@ -278,6 +275,14 @@ class TtsStreamingBuffer extends Emitter {
       }
       const chunk = combinedText.slice(0, chunkEnd);
 
+      // Check if the chunk is only whitespace before processing the queue
+      // If so, wait for more meaningful text
+      if (isWhitespace(chunk)) {
+        this.logger.debug('TtsStreamingBuffer:_feedQueue chunk is only whitespace, waiting for more text');
+        this._setTimerIfNeeded();
+        return;
+      }
+
       // Now we iterate over the queue items
       // and deduct their lengths until we've accounted for chunkEnd characters.
       let remaining = chunkEnd;
@@ -301,6 +306,14 @@ class TtsStreamingBuffer extends Emitter {
       this.bufferedLength -= chunkEnd;
 
       const modifiedChunk = chunk.replace(/\n\n/g, '\n \n');
+
+      if (isWhitespace(modifiedChunk)) {
+        this.logger.debug('TtsStreamingBuffer:_feedQueue modified chunk is only whitespace, restoring queue');
+        this.queue.unshift({ type: 'text', value: chunk });
+        this.bufferedLength += chunkEnd;
+        this._setTimerIfNeeded();
+        return;
+      }
       this.logger.debug(`TtsStreamingBuffer:_feedQueue sending chunk to tts: ${modifiedChunk}`);
 
       try {
@@ -349,6 +362,7 @@ class TtsStreamingBuffer extends Emitter {
     if (this.queue.length > 0) {
       await this._feedQueue();
     }
+    this.emit(TtsStreamingEvents.Connected, { vendor });
   }
 
   _onConnectFailure(vendor) {
@@ -399,6 +413,7 @@ class TtsStreamingBuffer extends Emitter {
 
   removeCustomEventListeners() {
     this.eventHandlers.forEach((h) => h.ep.removeCustomEventListener(h.event, h.handler));
+    this.eventHandlers.length = 0;
   }
 
   _initHandlers(ep) {
@@ -422,7 +437,15 @@ class TtsStreamingBuffer extends Emitter {
 
 const findSentenceBoundary = (text, limit) => {
   // Look for punctuation or double newline that signals sentence end.
-  const sentenceEndRegex = /[.!?](?=\s|$)|\n\n/g;
+  // Includes:
+  //   - ASCII: . ! ?
+  //   - Arabic: ؟ (question mark), ۔ (full stop)
+  //   - Japanese: 。 (full stop), ！, ？ (full-width exclamation/question)
+  //
+  // For languages that use spaces between sentences, we still require
+  // whitespace or end-of-string after the mark. For Japanese (no spaces),
+  // we treat the punctuation itself as a boundary regardless of following char.
+  const sentenceEndRegex = /[.!?؟۔](?=\s|$)|[。！？]|\n\n/g;
   let lastSentenceBoundary = -1;
   let match;
   while ((match = sentenceEndRegex.exec(text)) && match.index < limit) {

lib/utils/ws-requestor.js

@@ -293,7 +293,7 @@ class WsRequestor extends BaseRequestor {
 
       /* send the message */
       this.ws.send(JSON.stringify(obj), async() => {
-        this.logger.debug({obj}, `WsRequestor:request websocket: sent (${url})`);
+        if (obj.type !== 'llm:event') this.logger.debug({obj}, `WsRequestor:request websocket: sent (${url})`);
         // If session:reconnect is waiting for ack, hold here until ack to send queuedMsgs
         if (this._reconnectPromise) {
           try {

package.json

@@ -1,6 +1,6 @@
 {
   "name": "jambonz-feature-server",
-  "version": "0.9.4",
+  "version": "0.9.5",
   "main": "app.js",
   "engines": {
     "node": ">= 18.x"
@@ -27,14 +27,14 @@
   "dependencies": {
     "@aws-sdk/client-auto-scaling": "^3.549.0",
     "@aws-sdk/client-sns": "^3.549.0",
-    "@jambonz/db-helpers": "^0.9.12",
+    "@jambonz/db-helpers": "^0.9.18",
     "@jambonz/http-health-check": "^0.0.1",
     "@jambonz/mw-registrar": "^0.2.7",
-    "@jambonz/realtimedb-helpers": "^0.8.13",
-    "@jambonz/speech-utils": "^0.2.12",
+    "@jambonz/realtimedb-helpers": "^0.8.15",
+    "@jambonz/speech-utils": "^0.2.26",
     "@jambonz/stats-collector": "^0.1.10",
-    "@jambonz/time-series": "^0.2.13",
-    "@jambonz/verb-specifications": "^0.0.105",
+    "@jambonz/time-series": "^0.2.15",
+    "@jambonz/verb-specifications": "^0.0.122",
     "@modelcontextprotocol/sdk": "^1.9.0",
     "@opentelemetry/api": "^1.8.0",
     "@opentelemetry/exporter-jaeger": "^1.23.0",
@@ -48,13 +48,13 @@
     "bent": "^7.3.12",
     "debug": "^4.3.4",
     "deepcopy": "^2.1.0",
-    "drachtio-fsmrf": "^4.0.3",
-    "drachtio-srf": "^5.0.5",
+    "drachtio-fsmrf": "^4.1.2",
+    "drachtio-srf": "^5.0.14",
     "express": "^4.19.2",
     "express-validator": "^7.0.1",
     "moment": "^2.30.1",
     "parse-url": "^9.2.0",
-    "pino": "^8.20.0",
+    "pino": "^10.1.0",
     "polly-ssml-split": "^0.1.0",
     "sdp-transform": "^2.15.0",
     "short-uuid": "^5.1.0",

test/index.js

@@ -4,6 +4,7 @@ require('./ws-requestor-unit-test');
 require('./http-requestor-retry-test');
 require('./http-requestor-unit-test');
 require('./unit-tests');
+require('./tts-streaming-buffer-test');
 require('./docker_start');
 require('./create-test-db');
 require('./account-validation-tests');
@@ -16,6 +17,7 @@ require('./sip-request-tests');
 require('./create-call-test');
 require('./play-tests');
 require('./sip-refer-tests');
+require('./sip-refer-handler-tests');
 require('./listen-tests');
 require('./config-test');
 require('./queue-test');

test/scenarios/uas-dial-refer.xml

@@ -0,0 +1,117 @@
+<?xml version="1.0" encoding="ISO-8859-1" ?>
+<!DOCTYPE scenario SYSTEM "sipp.dtd">
+
+<scenario name="UAS that accepts call and sends REFER">
+  <!-- Receive incoming INVITE -->
+  <recv request="INVITE" crlf="true">
+    <action>
+      <ereg regexp=".*" search_in="hdr" header="Subject:" assign_to="1" />
+      <ereg regexp=".*" search_in="hdr" header="From:" assign_to="2" />
+    </action>
+  </recv>
+
+  <!-- Send 180 Ringing -->
+  <send>
+    <![CDATA[
+      SIP/2.0 180 Ringing
+      [last_Via:]
+      [last_From:]
+      [last_To:];tag=[pid]SIPpTag01[call_number]
+      [last_Call-ID:]
+      [last_CSeq:]
+      [last_Record-Route:]
+      Subject:[$1]
+      Content-Length: 0
+    ]]>
+  </send>
+
+  <!-- Send 200 OK with SDP -->
+  <send>
+    <![CDATA[
+      SIP/2.0 200 OK
+      [last_Via:]
+      [last_From:]
+      [last_To:];tag=[pid]SIPpTag01[call_number]
+      [last_Call-ID:]
+      [last_CSeq:]
+      [last_Record-Route:]
+      Subject:[$1]
+      Contact: <sip:[local_ip]:[local_port];transport=[transport]>
+      Content-Type: application/sdp
+      Content-Length: [len]
+
+      v=0
+      o=user1 53655765 2353687637 IN IP[local_ip_type] [local_ip]
+      s=-
+      c=IN IP[media_ip_type] [media_ip]
+      t=0 0
+      m=audio [media_port] RTP/AVP 0
+      a=rtpmap:0 PCMU/8000
+    ]]>
+  </send>
+
+<recv request="ACK" rtd="true" crlf="true">
+  <action>
+    <!-- Check if this is NOT the first call (tag ends with 012 or higher) -->
+    <ereg regexp="tag=1SIPpTag01[2-9]" search_in="hdr" header="To:" assign_to="3" />
+    <log message="Not first call check result: [$3]"/>
+  </action>
+</recv>
+
+<!-- Skip REFER if we found a non-first call tag -->
+<nop next="skip_refer" test="3" value="" compare="not_equal">
+  <action>
+    <log message="Found non-first call tag [$3], skipping REFER"/>
+  </action>
+</nop>
+
+<!-- Wait a moment, then send REFER (only on first call) -->
+<pause milliseconds="1000"/>
+
+<nop>
+  <action>
+    <log message="Sending REFER for first call"/>
+  </action>
+</nop>
+
+<!-- Send REFER (only on first iteration) -->
+  <send retrans="500">
+    <![CDATA[
+      REFER sip:service@[remote_ip]:[remote_port] SIP/2.0
+      Via: SIP/2.0/[transport] [local_ip]:[local_port];branch=[branch]
+      From: <sip:[local_ip]:[local_port]>;tag=[pid]SIPpTag01[call_number]
+      To: [$2]
+      [last_Call-ID:]
+      CSeq: 2 REFER
+      Contact: <sip:[local_ip]:[local_port];transport=[transport]>
+      Max-Forwards: 70
+      X-Call-Number: [call_number]
+      Refer-To: <sip:+15551234567@example.com>
+      Referred-By: <sip:[local_ip]:[local_port]>
+      Content-Length: 0
+    ]]>
+  </send>
+
+  <!-- Expect 202 Accepted (only on first iteration) -->
+  <recv response="202"/>
+  
+  <label id="skip_refer"/>
+
+  <!-- Wait for BYE from feature server -->
+  <recv request="BYE"/>
+
+  <!-- Send 200 OK to BYE -->
+  <send>
+    <![CDATA[
+      SIP/2.0 200 OK
+      [last_Via:]
+      [last_From:]
+      [last_To:]
+      [last_Call-ID:]
+      [last_CSeq:]
+      Contact: <sip:[local_ip]:[local_port];transport=[transport]>
+      Content-Length: 0
+    ]]>
+  </send>
+
+</scenario>

test/sip-refer-handler-tests.js

@@ -0,0 +1,90 @@
+const test = require('tape');
+const { sippUac } = require('./sipp')('test_fs');
+const bent = require('bent');
+const getJSON = bent('json')
+const clearModule = require('clear-module');
+const {provisionCallHook} = require('./utils');
+const { sleepFor } = require('../lib/utils/helpers');
+
+process.on('unhandledRejection', (reason, p) => {
+  console.log('Unhandled Rejection at: Promise', p, 'reason:', reason);
+});
+  
+function connect(connectable) {
+  return new Promise((resolve, reject) => {
+    connectable.on('connect', () => {
+      return resolve();
+    });
+  });
+}
+
+test('when parent leg recvs REFER it should end the dial after adulting child leg', async(t) => {
+    clearModule.all();
+    const {srf, disconnect} = require('../app');
+    try {
+      await connect(srf);
+      // wait for fs connected to drachtio server.
+      await sleepFor(1000);
+      
+      // GIVEN
+      const from = "dial_refer_handler";
+      let verbs = [
+        {
+          "verb": "dial",
+          "callerId": from,
+          "actionHook": "/actionHook",
+          "referHook": "/referHook",
+          "anchorMedia": true,
+          "target": [
+              {
+                  "type": "phone",
+                  "number": "15083084809"
+              }
+          ]
+        }
+      ];
+  
+      await provisionCallHook(from, verbs);
+  
+      // THEN
+      //const p = sippUac('uas-dial.xml', '172.38.0.10', undefined, undefined, 2);
+      const p = sippUac('uas-dial-refer.xml', '172.38.0.10', undefined, undefined, 2);
+      await sleepFor(1000);
+  
+      let account_sid = '622f62e4-303a-49f2-bbe0-eb1e1714e37a';
+  
+      let post = bent('http://127.0.0.1:3000/', 'POST', 'json', 201);
+      post('v1/createCall', {
+        'account_sid':account_sid,
+        "call_hook": {
+          "url": "http://127.0.0.1:3100/",
+          "method": "POST",
+        },
+        "from": from,
+        "to": {
+          "type": "phone",
+          "number": "15583084808"
+      }});
+  
+      await p;
+  
+      // Verify that the referHook was called
+      const obj = await getJSON(`http://127.0.0.1:3100/lastRequest/${from}_referHook`);
+      t.ok(obj.body.from === from,
+        'dial-refer-handler: referHook was called with correct from');
+      t.ok(obj.body.refer_details && obj.body.refer_details.sip_refer_to,
+        'dial-refer-handler: refer_details included in referHook');
+      t.ok(obj.body.refer_details.refer_to_user === '+15551234567',
+        'dial-refer-handler: refer_to_user correctly parsed');
+      t.ok(obj.body.refer_details.referring_call_sid,
+        'dial-refer-handler: referring_call_sid included');
+      t.ok(obj.body.refer_details.referred_call_sid,
+        'dial-refer-handler: referred_call_sid included');
+  
+      disconnect();
+    } catch (err) {
+      console.log(`error received: ${err}`);
+      disconnect();
+      t.error(err);
+    }
+});

test/sip-refer-tests.js

@@ -58,6 +58,46 @@ test('\'refer\' tests w/202 and NOTIFY', {timeout: 25000}, async(t) => {
   }
 });
 
+test('\'refer\' tests tel:', {timeout: 25000}, async(t) => {
+  clearModule.all();
+  const {srf, disconnect} = require('../app');
+
+  try {
+    await connect(srf);
+
+    // GIVEN
+    const verbs = [
+      {
+        verb: 'say',
+        text: 'silence_stream://100'
+      },
+      {
+        verb: 'sip:refer',
+        referTo: 'tel:+1234567890',
+        actionHook: '/actionHook'
+      }
+    ];
+    const noVerbs = [];
+
+    const from = 'refer_with_tel';
+    await provisionCallHook(from, verbs);
+    await provisionActionHook(from, noVerbs)
+
+    // THEN
+    await sippUac('uac-refer-with-notify.xml', '172.38.0.10', from);
+    t.pass('refer: successfully received 202 Accepted');
+    await sleepFor(1000);
+    const obj  = await getJSON(`http:127.0.0.1:3100/lastRequest/${from}_actionHook`);
+    t.ok(obj.body.final_referred_call_status === 200, 'refer: successfully received NOTIFY with 200 OK');
+    // console.log(`obj: ${JSON.stringify(obj)}`);
+    disconnect();
+  } catch (err) {
+    console.log(`error received: ${err}`);
+    disconnect();
+    t.error(err);
+  }
+});
+
 test('\'refer\' tests w/202 but no NOTIFY', {timeout: 25000}, async(t) => {
   clearModule.all();
   const {srf, disconnect} = require('../app');

test/tts-streaming-buffer-test.js

@@ -0,0 +1,177 @@
+const test = require('tape');
+const sinon = require('sinon');
+
+const noop = () => {};
+const logger = {
+  error: noop,
+  info: noop,
+  debug: noop
+};
+
+const {
+  TtsStreamingConnectionStatus
+} = require('../lib/utils/constants.json');
+
+const TtsStreamingBuffer = require('../lib/utils/tts-streaming-buffer');
+
+// Helper to create a mock CallSession
+function createMockCs(options = {}) {
+  const mockEp = {
+    uuid: 'test-uuid-1234',
+    api: sinon.stub().resolves({ body: '+OK' }),
+    addCustomEventListener: sinon.stub(),
+    removeCustomEventListener: sinon.stub()
+  };
+
+  return {
+    logger,
+    ep: mockEp,
+    isTtsStreamOpen: options.isTtsStreamOpen !== undefined ? options.isTtsStreamOpen : true,
+    getTsStreamingVendor: () => options.vendor || 'deepgram'
+  };
+}
+
+/**
+ * BUG REPRODUCTION TEST
+ *
+ * This test reproduces the exact issue from production logs:
+ * {
+ *   "args": ["uuid", "send", " "],
+ *   "msg": "Error calling uuid_deepgram_tts_streaming: -USAGE: <uuid> connect|send|clear|close [tokens]"
+ * }
+ *
+ * Root cause: When multiple flushes are queued while connecting, and a space token
+ * gets buffered between flushes, Phase 1 of _feedQueue sends that space to the TTS vendor.
+ *
+ * Sequence:
+ * 1. bufferTokens('Hello.') while connecting
+ * 2. flush() while connecting
+ * 3. bufferTokens(' ') while connecting (passes because bufferedLength > 0)
+ * 4. flush() while connecting
+ * 5. Connection completes, _feedQueue processes: [text:Hello., flush, text:" ", flush]
+ * 6. First flush sends "Hello." - OK
+ * 7. Second flush sends " " - BUG!
+ */
+test('TtsStreamingBuffer: multiple flushes while connecting - space token sent to TTS vendor', async(t) => {
+  const cs = createMockCs();
+  const buffer = new TtsStreamingBuffer(cs);
+
+  buffer._connectionStatus = TtsStreamingConnectionStatus.Connecting;
+  buffer.vendor = 'deepgram';
+
+  const apiCalls = [];
+  const originalApi = buffer._api.bind(buffer);
+  buffer._api = async function(ep, args) {
+    apiCalls.push({ args: [...args] });
+    return originalApi(ep, args);
+  };
+
+  // First batch while connecting
+  await buffer.bufferTokens('Hello.');
+  buffer.flush();
+
+  // Second batch - just a space (passes because bufferedLength > 0)
+  await buffer.bufferTokens(' ');
+  buffer.flush();
+
+  // Verify queue state before connect
+  t.equal(buffer.queue.length, 4, 'queue should have 4 items: [text, flush, text, flush]');
+  t.equal(buffer.queue[0].type, 'text', 'first item should be text');
+  t.equal(buffer.queue[0].value, 'Hello.', 'first text should be "Hello."');
+  t.equal(buffer.queue[1].type, 'flush', 'second item should be flush');
+  t.equal(buffer.queue[2].type, 'text', 'third item should be text');
+  t.equal(buffer.queue[2].value, ' ', 'third item should be space');
+  t.equal(buffer.queue[3].type, 'flush', 'fourth item should be flush');
+
+  // Connect - triggers _feedQueue
+  buffer._connectionStatus = TtsStreamingConnectionStatus.Connected;
+  await buffer._feedQueue();
+
+  // Check API calls
+  const sendCalls = apiCalls.filter(call => call.args[1] === 'send');
+
+  // This assertion will FAIL until the bug is fixed
+  const whitespaceOnlySends = sendCalls.filter(call => /^\s*$/.test(call.args[2]));
+
+  t.equal(whitespaceOnlySends.length, 0,
+    `should not send whitespace-only tokens, but sent: ${whitespaceOnlySends.map(c => JSON.stringify(c.args[2])).join(', ')}`);
+
+  t.end();
+});
+
+/**
+ * Additional test: Verify text with trailing space in same flush is OK
+ */
+test('TtsStreamingBuffer: text with trailing space in same flush should work', async(t) => {
+  const cs = createMockCs();
+  const buffer = new TtsStreamingBuffer(cs);
+
+  buffer._connectionStatus = TtsStreamingConnectionStatus.Connecting;
+  buffer.vendor = 'deepgram';
+
+  const apiCalls = [];
+  const originalApi = buffer._api.bind(buffer);
+  buffer._api = async function(ep, args) {
+    apiCalls.push({ args: [...args] });
+    return originalApi(ep, args);
+  };
+
+  // Buffer text with trailing space, then flush
+  await buffer.bufferTokens('Hello.');
+  await buffer.bufferTokens(' ');
+  buffer.flush();
+
+  // Connect
+  buffer._connectionStatus = TtsStreamingConnectionStatus.Connected;
+  await buffer._feedQueue();
+
+  const sendCalls = apiCalls.filter(call => call.args[1] === 'send');
+
+  t.equal(sendCalls.length, 1, 'should have one send call');
+  t.equal(sendCalls[0].args[2], 'Hello. ', 'should send "Hello. " (text with trailing space)');
+
+  t.end();
+});
+
+/**
+ * Test: Leading whitespace should be discarded when buffer is empty
+ */
+test('TtsStreamingBuffer: leading whitespace discarded when buffer empty', async(t) => {
+  const cs = createMockCs();
+  const buffer = new TtsStreamingBuffer(cs);
+
+  buffer._connectionStatus = TtsStreamingConnectionStatus.Connected;
+  buffer.vendor = 'deepgram';
+
+  // Try to buffer whitespace when buffer is empty
+  const result = await buffer.bufferTokens('   ');
+
+  t.equal(result.status, 'ok', 'should return ok status');
+  t.equal(buffer.bufferedLength, 0, 'buffer should remain empty');
+  t.equal(buffer.queue.length, 0, 'queue should remain empty');
+
+  t.end();
+});
+
+/**
+ * Test: Whitespace can be buffered when buffer has content
+ */
+test('TtsStreamingBuffer: whitespace accepted when buffer has content', async(t) => {
+  const cs = createMockCs();
+  const buffer = new TtsStreamingBuffer(cs);
+
+  buffer._connectionStatus = TtsStreamingConnectionStatus.Connecting;
+  buffer.vendor = 'deepgram';
+
+  // Buffer real text first
+  await buffer.bufferTokens('Hello');
+
+  // Now buffer whitespace (should pass because bufferedLength > 0)
+  const result = await buffer.bufferTokens(' ');
+
+  t.equal(result.status, 'ok', 'should return ok status');
+  t.equal(buffer.bufferedLength, 6, 'buffer should have 6 chars');
+  t.equal(buffer.queue.length, 2, 'queue should have 2 items');
+
+  t.end();
+});

test/webhook/app.js

@@ -99,6 +99,24 @@ app.post('/actionHook', (req, res) => {
   return res.sendStatus(200);
 });
 
+/*
+ * referHook
+ */
+app.post('/referHook', (req, res) => {
+  console.log({payload: req.body}, 'POST /referHook');
+  let key = req.body.from + "_referHook"
+  addRequestToMap(key, req, hook_mapping);
+  return res.json([{"verb": "pause", "length": 2}]);
+});
+
+/*
+ * adultingHook
+ */
+app.post('/adulting', (req, res) => {
+  console.log({payload: req.body}, 'POST /adulting');
+  return res.sendStatus(200);
+});
+
 /*
 * customHook
 * For the hook to return

test/webhooks-tests.js

@@ -83,7 +83,8 @@ test('invalid jambonz json create alert tests', async(t) => {
       {account_sid: 'bb845d4b-83a9-4cde-a6e9-50f3743bab3f', page: 1, page_size: 25, days: 7});
     let checked = false;
     for (let i = 0; i < data.total; i++) {
-      checked = data.data[i].message === 'malformed jambonz payload: must be array'
+      checked = data.data[i].message === 'malformed jambonz payload: must be array';
+      if (checked) break;
     }
     t.ok(checked, 'alert is raised as expected');
     disconnect();