fix #1466: (#1467)

* fix #1466:

* do not send tts streaming events when we are not doing tts streaming

lib/config.js

@@ -139,6 +139,10 @@ 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;
+
+/* 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;
@@ -231,5 +235,6 @@ module.exports = {
   JAMBONES_DIAL_SBC_FOR_REGISTERED_USER,
   JAMBONES_MEDIA_TIMEOUT_MS,
   JAMBONES_MEDIA_HOLD_TIMEOUT_MS,
+  JAMBONES_SAY_CHUNK_SIZE,
   JAMBONES_TRANSCRIBE_EP_DESTROY_DELAY_MS,
 };

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

@@ -291,7 +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/session/call-session.js

@@ -927,7 +927,7 @@ 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');
@@ -941,9 +941,11 @@ 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() {
@@ -951,7 +953,7 @@ class CallSession extends Emitter {
   }
 
   stopTtsStream() {
-    if (this.appIsUsingWebsockets) {
+    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();

lib/tasks/dial.js

@@ -158,7 +158,7 @@ class TaskDial extends Task {
 
   get canReleaseMedia() {
     const keepAnchor = this.data.anchorMedia ||
-      this.weAreTranscoding ||
+      this.isTranscoding ||
       this.cs.isBackGroundListen ||
       this.cs.onHoldMusic ||
       ANCHOR_MEDIA_ALWAYS ||
@@ -773,6 +773,15 @@ 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) {
@@ -930,13 +939,6 @@ class TaskDial extends Task {
         this.logger.info({err}, 'Dial:_selectSingleDial - Error boosting audio signal');
       }
     }
-    /* basic determination to see if call is being transcoded */
-    const codecA = getLeadingCodec(this.epOther.local.sdp);
-    const codecB = getLeadingCodec(this.ep.remote.sdp);
-    this.weAreTranscoding = (codecA !== codecB);
-    if (this.weAreTranscoding) {
-      this.logger.info(`Dial:_selectSingleDial - transcoding from ${codecA} (A leg) to ${codecB} (B leg)`);
-    }
     /* 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);

lib/tasks/gather.js

@@ -881,7 +881,7 @@ 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;
     }

lib/tasks/say.js

@@ -1,6 +1,7 @@
 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, NonFatalTaskError } = require('../utils/error');
 const { sleepFor } = require('../utils/helpers');
@@ -31,7 +32,7 @@ const isMatchingEvent = (logger, filename, playbackId, evt) => {
 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,

lib/utils/stt-latency-calculator.js

@@ -127,7 +127,6 @@ class SttLatencyCalculator extends Emitter {
 
   calculateLatency() {
     if (!this.isRunning) {
-      this.logger.debug('Latency calculator is not running, cannot calculate latency, returning default values');
       return null;
     }
 

lib/utils/tts-streaming-buffer.js

@@ -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) =>
@@ -437,7 +436,15 @@ 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) {