update speech util version 0.2.26 (#508)

* support sonic-3

* update supported languages

README.md

@@ -35,6 +35,7 @@ Configuration is provided via environment variables:
 |K8S_FEATURE_SERVER_SERVICE_PORT| feature server port(required for K8S) |no|
 |JAMBONZ_RECORD_WS_USERNAME| recording websocket username|no|
 |JAMBONZ_RECORD_WS_PASSWORD| recording websocket password|no|
+|DISABLE_RATE_LIMITS| disable rate limiting|no
 
 #### Database dependency
 A mysql database is used to store long-lived objects such as Accounts, Applications, etc. To create the database schema, use or review the scripts in the 'db' folder, particularly:

app.js

@@ -48,7 +48,8 @@ const {
   retrieveKey,
   deleteKey,
   incrKey,
-  listConferences
+  listConferences,
+  getCallCount
 } = require('./lib/helpers/realtimedb-helpers');
 const {
   getTtsVoices,
@@ -118,7 +119,8 @@ app.locals = {
   queryAlertsSP,
   writeCdrs,
   writeAlerts,
-  AlertType
+  AlertType,
+  getCallCount
 };
 
 const unless = (paths, middleware) => {
@@ -168,7 +170,12 @@ if (process.env.JAMBONES_TRUST_PROXY) {
     });
   }
 }
-app.use(limiter);
+
+const disableRateLimit = process.env.DISABLE_RATE_LIMITS === 'true' || process.env.DISABLE_RATE_LIMITS === '1';
+
+if (!disableRateLimit) {
+  app.use(limiter);
+}
 app.use(helmet());
 app.use(helmet.hidePoweredBy());
 app.use(nocache());

db/jambones-sql.sql

@@ -1,5 +1,4 @@
 /* SQLEditor (MySQL (2))*/
-
 SET FOREIGN_KEY_CHECKS=0;
 
 DROP TABLE IF EXISTS account_static_ips;
@@ -418,6 +417,7 @@ register_public_ip_in_contact BOOLEAN NOT NULL DEFAULT false,
 register_status VARCHAR(4096),
 dtmf_type ENUM('rfc2833','tones','info') NOT NULL DEFAULT 'rfc2833',
 outbound_sip_proxy VARCHAR(255),
+trunk_type ENUM('static_ip','auth','reg') NOT NULL DEFAULT 'static_ip',
 PRIMARY KEY (voip_carrier_sid)
 ) COMMENT='A Carrier or customer PBX that can send or receive calls';
 

db/jambones.sqs

@@ -546,12 +546,12 @@
         <schema><![CDATA[]]></schema>
         <comment><![CDATA[A Carrier or customer PBX that can send or receive calls]]></comment>
         <location>
-            <x>16.00</x>
-            <y>427.00</y>
+            <x>20.00</x>
+            <y>418.00</y>
         </location>
         <size>
             <width>293.00</width>
-            <height>580.00</height>
+            <height>600.00</height>
         </size>
         <zorder>6</zorder>
         <SQLField>
@@ -749,6 +749,13 @@
             <type><![CDATA[VARCHAR(255)]]></type>
             <uid><![CDATA[556ABA45-BC63-444D-8CB1-973EFCCF9FE7]]></uid>
         </SQLField>
+        <SQLField>
+            <name><![CDATA[trunk_type]]></name>
+            <type><![CDATA[ENUM('static_ip','auth','reg')]]></type>
+            <defaultValue><![CDATA[static_ip]]></defaultValue>
+            <notNull><![CDATA[1]]></notNull>
+            <uid><![CDATA[CCF1560C-349E-4DB9-91E5-120F1EDB7CDE]]></uid>
+        </SQLField>
         <labelWindowIndex><![CDATA[28]]></labelWindowIndex>
         <objectComment><![CDATA[A Carrier or customer PBX that can send or receive calls]]></objectComment>
         <ui.treeExpanded><![CDATA[1]]></ui.treeExpanded>
@@ -1293,7 +1300,7 @@
         <comment><![CDATA[a regex-based pattern match for call routing]]></comment>
         <location>
             <x>16.00</x>
-            <y>1007.00</y>
+            <y>1039.00</y>
         </location>
         <size>
             <width>254.00</width>
@@ -2165,8 +2172,8 @@
         <schema><![CDATA[]]></schema>
         <comment><![CDATA[A phone number that has been assigned to an account]]></comment>
         <location>
-            <x>16.00</x>
-            <y>1128.00</y>
+            <x>11.00</x>
+            <y>1162.00</y>
         </location>
         <size>
             <width>522.00</width>
@@ -2608,6 +2615,7 @@
         <SQLField>
             <name><![CDATA[speech_synthesis_voice]]></name>
             <type><![CDATA[VARCHAR(256)]]></type>
+            <defaultValue><![CDATA[en-US-Standard-C]]></defaultValue>
             <notNull><![CDATA[0]]></notNull>
             <uid><![CDATA[929D66F0-64B9-4D7C-AB4B-24F131E1178F]]></uid>
         </SQLField>
@@ -3161,17 +3169,17 @@
         <overviewPanelHidden><![CDATA[0]]></overviewPanelHidden>
         <pageBoundariesVisible><![CDATA[0]]></pageBoundariesVisible>
         <PageGridVisible><![CDATA[0]]></PageGridVisible>
-        <RightSidebarWidth><![CDATA[1393.000000]]></RightSidebarWidth>
+        <RightSidebarWidth><![CDATA[1643.000000]]></RightSidebarWidth>
         <sidebarIndex><![CDATA[2]]></sidebarIndex>
         <snapToGrid><![CDATA[0]]></snapToGrid>
-        <SourceSidebarWidth><![CDATA[0.000000]]></SourceSidebarWidth>
+        <SourceSidebarWidth><![CDATA[312.000000]]></SourceSidebarWidth>
         <SQLEditorFileFormatVersion><![CDATA[4]]></SQLEditorFileFormatVersion>
         <uid><![CDATA[58C99A00-06C9-478C-A667-C63842E088F3]]></uid>
         <windowHeight><![CDATA[1055.000000]]></windowHeight>
-        <windowLocationX><![CDATA[1845.000000]]></windowLocationX>
+        <windowLocationX><![CDATA[1728.000000]]></windowLocationX>
         <windowLocationY><![CDATA[37.000000]]></windowLocationY>
-        <windowScrollOrigin><![CDATA[{0, 544}]]></windowScrollOrigin>
-        <windowWidth><![CDATA[1670.000000]]></windowWidth>
+        <windowScrollOrigin><![CDATA[{0, 376}]]></windowScrollOrigin>
+        <windowWidth><![CDATA[1920.000000]]></windowWidth>
     </SQLDocumentInfo>
     <AllowsIndexRenamingOnInsert><![CDATA[1]]></AllowsIndexRenamingOnInsert>
     <defaultLabelExpanded><![CDATA[1]]></defaultLabelExpanded>

db/upgrade-jambonz-db.js

@@ -231,7 +231,8 @@ const sql = {
   ],
   9005: [
     'UPDATE applications SET speech_synthesis_voice = \'en-US-Standard-C\' WHERE speech_synthesis_voice IS NULL AND speech_synthesis_vendor = \'google\' AND speech_synthesis_language = \'en-US\'',
-    'ALTER TABLE applications MODIFY COLUMN speech_synthesis_voice VARCHAR(255) DEFAULT \'en-US-Standard-C\''
+    'ALTER TABLE applications MODIFY COLUMN speech_synthesis_voice VARCHAR(255) DEFAULT \'en-US-Standard-C\'',
+    'ALTER TABLE voip_carriers ADD COLUMN trunk_type ENUM(\'static_ip\',\'auth\',\'reg\') NOT NULL DEFAULT \'static_ip\'',
   ]
 };
 
@@ -267,6 +268,8 @@ const doIt = async() => {
         if (val < 9000) upgrades.push(...sql['9000']);
         if (val < 9002) upgrades.push(...sql['9002']);
         if (val < 9003) upgrades.push(...sql['9003']);
+        if (val < 9004) upgrades.push(...sql['9004']);
+        if (val < 9005) upgrades.push(...sql['9005']);
 
         // perform all upgrades
         logger.info({upgrades}, 'applying schema upgrades..');

lib/helpers/realtimedb-helpers.js

@@ -13,7 +13,8 @@ const {
   deleteKey,
   incrKey,
   client: redisClient,
-  listConferences
+  listConferences,
+  getCallCount
 } = require('@jambonz/realtimedb-helpers')({}, logger);
 
 module.exports = {
@@ -29,5 +30,6 @@ module.exports = {
   deleteKey,
   redisClient,
   incrKey,
-  listConferences
+  listConferences,
+  getCallCount
 };

lib/models/api-key.js

@@ -47,10 +47,14 @@ class ApiKey extends Model {
   }
 
   /**
-  * update last_used api key for an account
-  */
+   * update last_used api key for an account
+   * (only if last_used is null or more than a minute ago)
+   */
   static updateLastUsed(account_sid) {
-    const sql = 'UPDATE api_keys SET last_used = NOW() WHERE account_sid = ?';
+    const sql = `UPDATE api_keys 
+                SET last_used = NOW() 
+                WHERE account_sid = ? 
+                AND (last_used IS NULL OR last_used < NOW() - INTERVAL 1 MINUTE)`;
     const args = [account_sid];
 
     return new Promise((resolve, reject) => {

lib/record/s3-multipart-upload-stream.js

@@ -15,7 +15,9 @@ class S3MultipartUploadStream extends Writable {
     this.uploadId = null;
     this.partNumber = 1;
     this.multipartETags = [];
-    this.buffer = Buffer.alloc(0);
+    // accumulate incoming chunks to avoid O(n^2) Buffer.concat on every write
+    this.chunks = [];
+    this.bufferedBytes = 0;
     this.minPartSize = 5 * 1024 * 1024; // 5 MB
     this.s3 = new S3Client(opts.bucketCredential);
     this.metadata = opts.metadata;
@@ -31,13 +33,13 @@ class S3MultipartUploadStream extends Writable {
     return response.UploadId;
   }
 
-  async _uploadBuffer() {
+  async _uploadPart(bodyBuffer) {
     const uploadPartCommand = new UploadPartCommand({
       Bucket: this.bucketName,
       Key: this.objectKey,
       PartNumber: this.partNumber,
       UploadId: this.uploadId,
-      Body: this.buffer,
+      Body: bodyBuffer,
     });
 
     const uploadPartResponse = await this.s3.send(uploadPartCommand);
@@ -54,11 +56,16 @@ class S3MultipartUploadStream extends Writable {
         this.uploadId = await this._initMultipartUpload();
       }
 
-      this.buffer = Buffer.concat([this.buffer, chunk]);
+      // accumulate without concatenating on every write
+      this.chunks.push(chunk);
+      this.bufferedBytes += chunk.length;
 
-      if (this.buffer.length >= this.minPartSize) {
-        await this._uploadBuffer();
-        this.buffer = Buffer.alloc(0);
+      if (this.bufferedBytes >= this.minPartSize) {
+        const partBuffer = Buffer.concat(this.chunks, this.bufferedBytes);
+        // reset accumulators before awaiting upload to allow GC
+        this.chunks = [];
+        this.bufferedBytes = 0;
+        await this._uploadPart(partBuffer);
       }
 
       callback(null);
@@ -69,8 +76,11 @@ class S3MultipartUploadStream extends Writable {
 
   async _finalize(err) {
     try {
-      if (this.buffer.length > 0) {
-        await this._uploadBuffer();
+      if (this.bufferedBytes > 0) {
+        const finalBuffer = Buffer.concat(this.chunks, this.bufferedBytes);
+        this.chunks = [];
+        this.bufferedBytes = 0;
+        await this._uploadPart(finalBuffer);
       }
 
       const completeMultipartUploadCommand = new CompleteMultipartUploadCommand({

lib/record/upload.js

@@ -51,8 +51,10 @@ async function upload(logger, socket) {
 
           /**encoder */
           let encoder;
+          let recordFormat;
           if (account[0].record_format === 'wav') {
             encoder = new wav.Writer({ channels: 2, sampleRate, bitDepth: 16 });
+            recordFormat = 'wav';
           } else {
             // default is mp3
             encoder = new PCMToMP3Encoder({
@@ -60,7 +62,9 @@ async function upload(logger, socket) {
               sampleRate: sampleRate,
               bitrate: 128
             }, logger);
+            recordFormat = 'mp3';
           }
+          logger.info({ record_format: recordFormat, channels: 2, sampleRate }, 'record upload: selected encoder');
 
           /* start streaming data */
           pipeline(

lib/routes/api/accounts.js

@@ -161,6 +161,9 @@ router.post('/:sid/VoipCarriers', async(req, res) => {
   try {
     const account_sid = parseAccountSid(req);
     await validateRequest(req, account_sid);
+    // Set the service_provder_sid to the relevent value for the account
+    const account = await Account.retrieve(req.user.account_sid);
+    payload.service_provider_sid = account[0].service_provider_sid;
 
     logger.debug({payload}, 'POST /:sid/VoipCarriers');
     const uuid = await VoipCarrier.make({
@@ -1220,4 +1223,23 @@ router.get('/:sid/Conferences', async(req, res) => {
   }
 });
 
+/**
+ * retrieve counts of calls under an account
+ */
+router.get('/:sid/CallCount', async(req, res) => {
+  const {logger, getCallCount} = req.app.locals;
+  try {
+    const accountSid = parseAccountSid(req);
+    await validateRequest(req, accountSid);
+    const count = await getCallCount(accountSid);
+    count.outbound = Number(count.outbound);
+    count.inbound = Number(count.inbound);
+    logger.debug(`retrieved, outbound: ${count.outbound}, inbound: ${count.inbound}, for account sid ${accountSid}`);
+    res.status(200).json(count);
+    updateLastUsed(logger, accountSid, req).catch((err) => {});
+  } catch (err) {
+    sysError(logger, res, err);
+  }
+});
+
 module.exports = router;

lib/routes/api/forgot-password.js

@@ -52,20 +52,21 @@ router.post('/', async(req, res) => {
   let obj;
   try {
     if (!email || !validateEmail(email)) {
+      logger.info({email}, 'Bad POST to /forgot-password is missing email or invalid email');
       return res.status(400).json({error: 'invalid or missing email'});
     }
 
     const [r] = await promisePool.query({sql, nestTables: true}, email);
     if (0 === r.length) {
-      logger.info('user not found');
+      logger.info(`user not found: ${email}`);
       return res.status(400).json({error: 'failed to reset your password'});
     }
     obj = r[0];
     if (!obj.user.is_active) {
-      logger.info(obj.user.name, 'user is inactive');
+      logger.info({user: obj.user.name, obj}, 'user is inactive');
       return res.status(400).json({error: 'failed to reset your password'});
     } else if (obj.acc.account_sid !== null && !obj.acc.is_active) {
-      logger.info(obj.acc.account_sid, 'account is inactive');
+      logger.info({account_sid: obj.acc.account_sid, obj}, 'account is inactive');
       return res.status(400).json({error: 'failed to reset your password'});
     }
     res.sendStatus(204);

lib/routes/api/service-providers.js

@@ -46,10 +46,16 @@ async function validateRetrieve(req) {
       return;
     }
 
-    if (req.user.hasScope('service_provider') || req.user.hasScope('account')) {
+    if (req.user.hasScope('service_provider')) {
       if (service_provider_sid === req.user.service_provider_sid) return;
     }
 
+    if (req.user.hasScope('account')) {
+      const results = await Account.retrieve(req.user.account_sid);
+      if (service_provider_sid === results[0].service_provider_sid) return;
+    }
+
+
     throw new DbErrorForbidden('insufficient permissions');
   } catch (error) {
     throw error;

lib/routes/api/sip-gateways.js

@@ -18,8 +18,7 @@ const checkUserScope = async(req, voip_carrier_sid) => {
     const carrier = await lookupCarrierBySid(voip_carrier_sid);
     if (!carrier) throw new DbErrorBadRequest('invalid voip_carrier_sid');
 
-    if ((!carrier.service_provider_sid || carrier.service_provider_sid === req.user.service_provider_sid) &&
-      (!carrier.account_sid || carrier.account_sid === req.user.account_sid)) {
+    if (!carrier.account_sid || carrier.account_sid === req.user.account_sid) {
 
       if (req.method !== 'GET' && !carrier.account_sid) {
         throw new DbErrorForbidden('insufficient privileges');

lib/routes/api/speech-credentials.js

@@ -15,7 +15,10 @@ const {decryptCredential, testWhisper, testDeepgramTTS,
   testCartesia,
   testVoxistStt,
   testOpenAiStt,
-  testInworld} = require('../../utils/speech-utils');
+  testInworld,
+  testResembleTTS,
+  testHoundifyStt,
+  testGladiaStt} = require('../../utils/speech-utils');
 const {DbErrorUnprocessableRequest, DbErrorForbidden, DbErrorBadRequest} = require('../../utils/errors');
 const {
   testGoogleTts,
@@ -123,6 +126,7 @@ const encryptCredential = (obj) => {
     role_arn,
     region,
     client_id,
+    client_key,
     client_secret,
     secret,
     nuance_tts_uri,
@@ -132,6 +136,8 @@ const encryptCredential = (obj) => {
     deepgram_stt_use_tls,
     deepgram_tts_uri,
     playht_tts_uri,
+    resemble_tts_uri,
+    resemble_tts_use_tls,
     use_custom_tts,
     custom_tts_endpoint,
     custom_tts_endpoint_url,
@@ -159,6 +165,7 @@ const encryptCredential = (obj) => {
     voice_engine,
     engine_version,
     service_version,
+    api_uri,
     options
   } = obj;
 
@@ -226,6 +233,24 @@ const encryptCredential = (obj) => {
         deepgram_stt_use_tls, deepgram_tts_uri, model_id});
       return encrypt(deepgramData);
 
+    case 'gladia':
+      const gladiaData = JSON.stringify({api_key, region});
+      return encrypt(gladiaData);
+
+    case 'resemble':
+      assert(api_key, 'invalid resemble speech credential: api_key is required');
+      const resembleData = JSON.stringify({
+        api_key,
+        ...(resemble_tts_uri && {resemble_tts_uri}),
+        ...(resemble_tts_use_tls && {resemble_tts_use_tls})
+      });
+      return encrypt(resembleData);
+
+    case 'deepgramflux':
+      assert(api_key, 'invalid deepgram flux speech credential: api_key is required');
+      const deepgramfluxData = JSON.stringify({api_key});
+      return encrypt(deepgramfluxData);
+
     case 'ibm':
       const ibmData = JSON.stringify({tts_api_key, tts_region, stt_api_key, stt_region, instance_id});
       return encrypt(ibmData);
@@ -248,7 +273,11 @@ const encryptCredential = (obj) => {
     case 'elevenlabs':
       assert(api_key, 'invalid elevenLabs speech credential: api_key is required');
       assert(model_id, 'invalid elevenLabs speech credential: model_id is required');
-      const elevenlabsData = JSON.stringify({api_key, model_id, options});
+      const elevenlabsData = JSON.stringify({
+        api_key,
+        model_id,
+        ...(api_uri && {api_uri}),
+        options});
       return encrypt(elevenlabsData);
 
     case 'speechmatics':
@@ -296,6 +325,13 @@ const encryptCredential = (obj) => {
       const assemblyaiData = JSON.stringify({api_key, service_version});
       return encrypt(assemblyaiData);
 
+    case 'houndify':
+      assert(client_id, 'invalid houndify speech credential: client_id is required');
+      assert(client_key, 'invalid houndify speech credential: client_key is required');
+      assert(user_id, 'invalid houndify speech credential: user_id is required');
+      const houndifyData = JSON.stringify({client_id, client_key, user_id});
+      return encrypt(houndifyData);
+
     case 'voxist':
       assert(api_key, 'invalid voxist speech credential: api_key is required');
       const voxistData = JSON.stringify({api_key});
@@ -518,7 +554,10 @@ router.put('/:sid', async(req, res) => {
           playht_tts_uri,
           engine_version,
           service_version,
-          speechmatics_stt_uri
+          speechmatics_stt_uri,
+          resemble_tts_use_tls,
+          resemble_tts_uri,
+          api_uri
         } = req.body;
 
         const newCred = {
@@ -551,7 +590,10 @@ router.put('/:sid', async(req, res) => {
           playht_tts_uri,
           engine_version,
           service_version,
-          speechmatics_stt_uri
+          speechmatics_stt_uri,
+          resemble_tts_uri,
+          resemble_tts_use_tls,
+          api_uri
         };
         logger.info({o, newCred}, 'updating speech credential with this new credential');
         obj.credential = encryptCredential(newCred);
@@ -751,6 +793,17 @@ router.get('/:sid/test', async(req, res) => {
           SpeechCredential.sttTestResult(sid, false);
         }
       }
+    } else if (cred.vendor === 'resemble') {
+      if (cred.use_for_tts) {
+        try {
+          await testResembleTTS(logger, synthAudio, credential);
+          results.tts.status = 'ok';
+          SpeechCredential.ttsTestResult(sid, true);
+        } catch (err) {
+          results.tts = {status: 'fail', reason: err.message};
+          SpeechCredential.ttsTestResult(sid, false);
+        }
+      }
     } else if (cred.vendor === 'deepgram') {
       const {api_key} = credential;
       if (cred.use_for_tts) {
@@ -774,6 +827,31 @@ router.get('/:sid/test', async(req, res) => {
         }
       }
     }
+    else if (cred.vendor === 'deepgramflux') {
+      const {api_key} = credential;
+      if (cred.use_for_stt && api_key) {
+        try {
+          await testDeepgramStt(logger, {api_key});
+          results.stt.status = 'ok';
+          SpeechCredential.sttTestResult(sid, true);
+        } catch (err) {
+          results.stt = {status: 'fail', reason: err.message};
+          SpeechCredential.sttTestResult(sid, false);
+        }
+      }
+    }
+    else if (cred.vendor === 'gladia') {
+      if (cred.use_for_stt) {
+        try {
+          await testGladiaStt(logger, credential);
+          results.stt.status = 'ok';
+          SpeechCredential.sttTestResult(sid, true);
+        } catch (err) {
+          results.stt = {status: 'fail', reason: err.message};
+          SpeechCredential.sttTestResult(sid, false);
+        }
+      }
+    }
     else if (cred.vendor === 'ibm') {
       const {getTtsVoices} = req.app.locals;
 
@@ -820,10 +898,10 @@ router.get('/:sid/test', async(req, res) => {
         }
       }
     } else if (cred.vendor === 'elevenlabs') {
-      const {api_key, model_id} = credential;
+      const {api_key, model_id, api_uri} = credential;
       if (cred.use_for_tts) {
         try {
-          await testElevenlabs(logger, {api_key, model_id});
+          await testElevenlabs(logger, {api_key, model_id, api_uri});
           results.tts.status = 'ok';
           SpeechCredential.ttsTestResult(sid, true);
         } catch (err) {
@@ -918,6 +996,17 @@ router.get('/:sid/test', async(req, res) => {
           SpeechCredential.sttTestResult(sid, false);
         }
       }
+    } else if (cred.vendor === 'houndify') {
+      if (cred.use_for_stt) {
+        try {
+          await testHoundifyStt(logger, credential);
+          results.stt.status = 'ok';
+          SpeechCredential.sttTestResult(sid, true);
+        } catch (err) {
+          results.stt = {status: 'fail', reason: err.message};
+          SpeechCredential.sttTestResult(sid, false);
+        }
+      }
     } else if (cred.vendor === 'voxist') {
       const {api_key} = credential;
       if (cred.use_for_stt) {

lib/utils/speech-data/stt-deepgram.js

@@ -1,56 +1,153 @@
 module.exports = [
   { name: 'Multilingual', value: 'multi' },
+  { name: 'Afrikaans', value: 'af' },
+  { name: 'Amharic', value: 'am' },
+  { name: 'Arabic', value: 'ar' },
+  { name: 'Assamese', value: 'as' },
+  { name: 'Azerbaijani', value: 'az' },
+  { name: 'Bashkir', value: 'ba' },
+  { name: 'Belarusian', value: 'be' },
   { name: 'Bulgarian', value: 'bg' },
+  { name: 'Bengali', value: 'bn' },
+  { name: 'Tibetan', value: 'bo' },
+  { name: 'Breton', value: 'br' },
+  { name: 'Bosnian', value: 'bs' },
   { name: 'Catalan', value: 'ca' },
-  { name: 'Chinese (Mandarin, Simplified)', value: 'zh' },
-  { name: 'Chinese (Mandarin, Simplified - China)', value: 'zh-CN' },
-  { name: 'Chinese (Mandarin, Simplified - Hans)', value: 'zh-Hans' },
-  { name: 'Chinese (Mandarin, Traditional)', value: 'zh-TW' },
-  { name: 'Chinese (Mandarin, Traditional - Hant)', value: 'zh-Hant' },
-  { name: 'Chinese (Cantonese, Traditional - Hong Kong)', value: 'zh-HK' },
   { name: 'Czech', value: 'cs' },
+  { name: 'Welsh', value: 'cy' },
   { name: 'Danish', value: 'da' },
   { name: 'Danish (Denmark)', value: 'da-DK' },
-  { name: 'Dutch', value: 'nl' },
-  { name: 'English', value: 'en' },
-  { name: 'English (United States)', value: 'en-US' },
-  { name: 'English (Australia)', value: 'en-AU' },
-  { name: 'English (United Kingdom)', value: 'en-GB' },
-  { name: 'English (New Zealand)', value: 'en-NZ' },
-  { name: 'English (India)', value: 'en-IN' },
-  { name: 'Estonian', value: 'et' },
-  { name: 'Finnish', value: 'fi' },
-  { name: 'Flemish', value: 'nl-BE' },
-  { name: 'French', value: 'fr' },
-  { name: 'French (Canada)', value: 'fr-CA' },
   { name: 'German', value: 'de' },
+  { name: 'German (Austria)', value: 'de-AT' },
   { name: 'German (Switzerland)', value: 'de-CH' },
-  { name: 'Greek', value: 'el' },
+  { name: 'German (Germany)', value: 'de-DE' },
+  { name: 'Modern Greek', value: 'el' },
+  { name: 'English', value: 'en' },
+  { name: 'English (Australia)', value: 'en-AU' },
+  { name: 'English (Canada)', value: 'en-CA' },
+  { name: 'English (United Kingdom)', value: 'en-GB' },
+  { name: 'English (Ireland)', value: 'en-IE' },
+  { name: 'English (India)', value: 'en-IN' },
+  { name: 'English (Malaysia)', value: 'en-MY' },
+  { name: 'English (New Zealand)', value: 'en-NZ' },
+  { name: 'English (Philippines)', value: 'en-PH' },
+  { name: 'English (United States)', value: 'en-US' },
+  { name: 'English (South Africa)', value: 'en-ZA' },
+  { name: 'Spanish', value: 'es' },
+  { name: 'Spanish (Latin America and the Caribbean)', value: 'es-419' },
+  { name: 'Spanish (Argentina)', value: 'es-AR' },
+  { name: 'Spanish (Colombia)', value: 'es-CO' },
+  { name: 'Spanish (Spain)', value: 'es-ES' },
+  { name: 'Spanish (Latin America)', value: 'es-LATAM' },
+  { name: 'Spanish (Mexico)', value: 'es-MX' },
+  { name: 'Spanish (United States)', value: 'es-US' },
+  { name: 'Estonian', value: 'et' },
+  { name: 'Basque', value: 'eu' },
+  { name: 'Persian', value: 'fa' },
+  { name: 'Finnish', value: 'fi' },
+  { name: 'Faroese', value: 'fo' },
+  { name: 'French', value: 'fr' },
+  { name: 'French (Belgium)', value: 'fr-BE' },
+  { name: 'French (Canada)', value: 'fr-CA' },
+  { name: 'French (Switzerland)', value: 'fr-CH' },
+  { name: 'French (France)', value: 'fr-FR' },
+  { name: 'French (Canada)', value: 'fr-ca' },
+  { name: 'Galician', value: 'gl' },
+  { name: 'Gujarati', value: 'gu' },
+  { name: 'Hausa', value: 'ha' },
+  { name: 'Hawaiian', value: 'haw' },
+  { name: 'Hebrew', value: 'he' },
   { name: 'Hindi', value: 'hi' },
+  { name: 'Hindi (Latin)', value: 'hi-Latn' },
+  { name: 'Croatian', value: 'hr' },
+  { name: 'Haitian', value: 'ht' },
   { name: 'Hungarian', value: 'hu' },
+  { name: 'Armenian', value: 'hy' },
   { name: 'Indonesian', value: 'id' },
+  { name: 'Indonesian (Indonesia)', value: 'id-ID' },
+  { name: 'Icelandic', value: 'is' },
   { name: 'Italian', value: 'it' },
+  { name: 'Italian (Italy)', value: 'it-IT' },
   { name: 'Japanese', value: 'ja' },
+  { name: 'Japanese (Japan)', value: 'ja-JP' },
+  { name: 'Javanese', value: 'jw' },
+  { name: 'Georgian', value: 'ka' },
+  { name: 'Kazakh', value: 'kk' },
+  { name: 'Khmer', value: 'km' },
+  { name: 'Kannada', value: 'kn' },
   { name: 'Korean', value: 'ko' },
-  { name: 'Korean (South Korea)', value: 'ko-KR' },
-  { name: 'Latvian', value: 'lv' },
+  { name: 'Korean (Republic of Korea)', value: 'ko-KR' },
+  { name: 'Latin', value: 'la' },
+  { name: 'Luxembourgish', value: 'lb' },
+  { name: 'Lingala', value: 'ln' },
+  { name: 'Lao', value: 'lo' },
   { name: 'Lithuanian', value: 'lt' },
+  { name: 'Latvian', value: 'lv' },
+  { name: 'Malagasy', value: 'mg' },
+  { name: 'Maori', value: 'mi' },
+  { name: 'Macedonian', value: 'mk' },
+  { name: 'Malayalam', value: 'ml' },
+  { name: 'Mongolian', value: 'mn' },
+  { name: 'Marathi', value: 'mr' },
   { name: 'Malay', value: 'ms' },
+  { name: 'Malay (Malaysia)', value: 'ms-MY' },
+  { name: 'Malay (Singapore)', value: 'ms-SG' },
+  { name: 'Maltese', value: 'mt' },
+  { name: 'Burmese', value: 'my' },
+  { name: 'Nepali', value: 'ne' },
+  { name: 'Dutch', value: 'nl' },
+  { name: 'Dutch (Belgium)', value: 'nl-BE' },
+  { name: 'Dutch (Netherlands)', value: 'nl-NL' },
+  { name: 'Norwegian Nynorsk', value: 'nn' },
   { name: 'Norwegian', value: 'no' },
+  { name: 'Norwegian (Norway)', value: 'no-NO' },
+  { name: 'Occitan', value: 'oc' },
+  { name: 'Panjabi', value: 'pa' },
   { name: 'Polish', value: 'pl' },
+  { name: 'Polish (Poland)', value: 'pl-PL' },
+  { name: 'Pushto', value: 'ps' },
   { name: 'Portuguese', value: 'pt' },
   { name: 'Portuguese (Brazil)', value: 'pt-BR' },
   { name: 'Portuguese (Portugal)', value: 'pt-PT' },
   { name: 'Romanian', value: 'ro' },
+  { name: 'Romanian (Moldova)', value: 'ro-MD' },
   { name: 'Russian', value: 'ru' },
+  { name: 'Russian (Latin)', value: 'ru-Latn' },
+  { name: 'Russian (Russian Federation)', value: 'ru-RU' },
+  { name: 'Sanskrit', value: 'sa' },
+  { name: 'Sindhi', value: 'sd' },
+  { name: 'Sinhala', value: 'si' },
   { name: 'Slovak', value: 'sk' },
-  { name: 'Spanish', value: 'es' },
-  { name: 'Spanish (Latin America)', value: 'es-419' },
+  { name: 'Slovenian', value: 'sl' },
+  { name: 'Shona', value: 'sn' },
+  { name: 'Somali', value: 'so' },
+  { name: 'Albanian', value: 'sq' },
+  { name: 'Serbian', value: 'sr' },
+  { name: 'Sundanese', value: 'su' },
   { name: 'Swedish', value: 'sv' },
   { name: 'Swedish (Sweden)', value: 'sv-SE' },
+  { name: 'Swahili', value: 'sw' },
+  { name: 'Tamil', value: 'ta' },
+  { name: 'Tamasheq', value: 'taq' },
+  { name: 'Telugu', value: 'te' },
+  { name: 'Tajik', value: 'tg' },
   { name: 'Thai', value: 'th' },
   { name: 'Thai (Thailand)', value: 'th-TH' },
+  { name: 'Turkmen', value: 'tk' },
+  { name: 'Tagalog', value: 'tl' },
   { name: 'Turkish', value: 'tr' },
+  { name: 'Turkish (Türkiye)', value: 'tr-TR' },
+  { name: 'Tatar', value: 'tt' },
   { name: 'Ukrainian', value: 'uk' },
-  { name: 'Vietnamese', value: 'vi' }
+  { name: 'Urdu', value: 'ur' },
+  { name: 'Uzbek', value: 'uz' },
+  { name: 'Vietnamese', value: 'vi' },
+  { name: 'Yiddish', value: 'yi' },
+  { name: 'Yoruba', value: 'yo' },
+  { name: 'Chinese (Mandarin, Mainland)', value: 'zh' },
+  { name: 'Chinese (China)', value: 'zh-CN' },
+  { name: 'Chinese (Cantonese, Hong Kong)', value: 'zh-HK' },
+  { name: 'Chinese (Han (Simplified variant))', value: 'zh-Hans' },
+  { name: 'Chinese (Han (Traditional variant))', value: 'zh-Hant' },
+  { name: 'Chinese (Traditional, Taiwan)', value: 'zh-TW' }
 ];

lib/utils/speech-data/stt-gladia.js

@@ -0,0 +1,103 @@
+module.exports = [
+  { name: 'Afrikaans', value: 'af' },
+  { name: 'Albanian', value: 'sq' },
+  { name: 'Amharic', value: 'am' },
+  { name: 'Arabic', value: 'ar' },
+  { name: 'Armenian', value: 'hy' },
+  { name: 'Assamese', value: 'as' },
+  { name: 'Azerbaijani', value: 'az' },
+  { name: 'Bashkir', value: 'ba' },
+  { name: 'Basque', value: 'eu' },
+  { name: 'Belarusian', value: 'be' },
+  { name: 'Bengali', value: 'bn' },
+  { name: 'Bosnian', value: 'bs' },
+  { name: 'Breton', value: 'br' },
+  { name: 'Bulgarian', value: 'bg' },
+  { name: 'Cantonese', value: 'yue' },
+  { name: 'Catalan', value: 'ca' },
+  { name: 'Chinese', value: 'zh' },
+  { name: 'Croatian', value: 'hr' },
+  { name: 'Czech', value: 'cs' },
+  { name: 'Danish', value: 'da' },
+  { name: 'Dutch', value: 'nl' },
+  { name: 'English', value: 'en' },
+  { name: 'Estonian', value: 'et' },
+  { name: 'Faroese', value: 'fo' },
+  { name: 'Finnish', value: 'fi' },
+  { name: 'French', value: 'fr' },
+  { name: 'Galician', value: 'gl' },
+  { name: 'Georgian', value: 'ka' },
+  { name: 'German', value: 'de' },
+  { name: 'Greek', value: 'el' },
+  { name: 'Gujarati', value: 'gu' },
+  { name: 'Haitian Creole', value: 'ht' },
+  { name: 'Hausa', value: 'ha' },
+  { name: 'Hawaiian', value: 'haw' },
+  { name: 'Hebrew', value: 'he' },
+  { name: 'Hindi', value: 'hi' },
+  { name: 'Hungarian', value: 'hu' },
+  { name: 'Icelandic', value: 'is' },
+  { name: 'Indonesian', value: 'id' },
+  { name: 'Italian', value: 'it' },
+  { name: 'Japanese', value: 'ja' },
+  { name: 'Javanese', value: 'jw' },
+  { name: 'Kannada', value: 'kn' },
+  { name: 'Kazakh', value: 'kk' },
+  { name: 'Khmer', value: 'km' },
+  { name: 'Korean', value: 'ko' },
+  { name: 'Lao', value: 'lo' },
+  { name: 'Latin', value: 'la' },
+  { name: 'Latvian', value: 'lv' },
+  { name: 'Lingala', value: 'ln' },
+  { name: 'Lithuanian', value: 'lt' },
+  { name: 'Luxembourgish', value: 'lb' },
+  { name: 'Macedonian', value: 'mk' },
+  { name: 'Malagasy', value: 'mg' },
+  { name: 'Malay', value: 'ms' },
+  { name: 'Malayalam', value: 'ml' },
+  { name: 'Maltese', value: 'mt' },
+  { name: 'Maori', value: 'mi' },
+  { name: 'Marathi', value: 'mr' },
+  { name: 'Mongolian', value: 'mn' },
+  { name: 'Myanmar', value: 'my' },
+  { name: 'Nepali', value: 'ne' },
+  { name: 'Norwegian', value: 'no' },
+  { name: 'Nynorsk', value: 'nn' },
+  { name: 'Occitan', value: 'oc' },
+  { name: 'Pashto', value: 'ps' },
+  { name: 'Persian', value: 'fa' },
+  { name: 'Polish', value: 'pl' },
+  { name: 'Portuguese', value: 'pt' },
+  { name: 'Punjabi', value: 'pa' },
+  { name: 'Romanian', value: 'ro' },
+  { name: 'Russian', value: 'ru' },
+  { name: 'Sanskrit', value: 'sa' },
+  { name: 'Serbian', value: 'sr' },
+  { name: 'Shona', value: 'sn' },
+  { name: 'Sindhi', value: 'sd' },
+  { name: 'Sinhala', value: 'si' },
+  { name: 'Slovak', value: 'sk' },
+  { name: 'Slovenian', value: 'sl' },
+  { name: 'Somali', value: 'so' },
+  { name: 'Spanish', value: 'es' },
+  { name: 'Sundanese', value: 'su' },
+  { name: 'Swahili', value: 'sw' },
+  { name: 'Swedish', value: 'sv' },
+  { name: 'Tagalog', value: 'tl' },
+  { name: 'Tajik', value: 'tg' },
+  { name: 'Tamil', value: 'ta' },
+  { name: 'Tatar', value: 'tt' },
+  { name: 'Telugu', value: 'te' },
+  { name: 'Thai', value: 'th' },
+  { name: 'Tibetan', value: 'bo' },
+  { name: 'Turkish', value: 'tr' },
+  { name: 'Turkmen', value: 'tk' },
+  { name: 'Ukrainian', value: 'uk' },
+  { name: 'Urdu', value: 'ur' },
+  { name: 'Uzbek', value: 'uz' },
+  { name: 'Vietnamese', value: 'vi' },
+  { name: 'Welsh', value: 'cy' },
+  { name: 'Wolof', value: 'wo' },
+  { name: 'Yiddish', value: 'yi' },
+  { name: 'Yoruba', value: 'yo' }
+];

lib/utils/speech-data/stt-houndify.js

@@ -0,0 +1,19 @@
+module.exports = [
+  { name: 'English', value: 'en' },
+  { name: 'Spanish', value: 'es' },
+  { name: 'Portuguese', value: 'pt' },
+  { name: 'French', value: 'fr' },
+  { name: 'Indian-accented English', value: 'en-IN' },
+  { name: 'German', value: 'de' },
+  { name: 'Dutch', value: 'nl' },
+  { name: 'Italian', value: 'it' },
+  { name: 'Korean', value: 'ko' },
+  { name: 'Japanese', value: 'ja' },
+  { name: 'Mandarin', value: 'zh-CN' },
+  { name: 'Russian', value: 'ru' },
+  { name: 'Polish', value: 'pl' },
+  { name: 'Swedish', value: 'sv' },
+  { name: 'Arabic', value: 'ar' },
+  { name: 'Turkish', value: 'tr' },
+  { name: 'Hebrew', value: 'he' },
+];

lib/utils/speech-data/tts-model-cartesia.js

@@ -9,6 +9,15 @@ module.exports = [
     value: 'sonic-2',
     languages: ['en', 'fr', 'de', 'es', 'pt', 'zh', 'ja', 'hi', 'it', 'ko', 'nl', 'pl', 'ru', 'sv', 'tr']
   },
+  {
+    name: 'Sonic 3',
+    value: 'sonic-3',
+    languages: [
+      'en', 'fr', 'de', 'es', 'pt', 'zh', 'ja', 'hi', 'it', 'ko', 'nl', 'pl', 'ru', 'sv', 'tr',
+      'tl', 'bg', 'ro', 'ar', 'cs', 'el', 'fi', 'hr', 'ms', 'sk', 'da', 'ta', 'uk', 'hu', 'no',
+      'vi', 'bn', 'th', 'he', 'ka', 'id', 'te', 'gu', 'kn', 'ml', 'mr', 'pa'
+    ]
+  },
   {
     name: 'Sonic Turbo',
     value: 'sonic-turbo',

lib/utils/speech-data/tts-resemble.js

@@ -0,0 +1,438 @@
+module.exports = [
+  {
+    value: 'en-gb',
+    name: 'En-gb',
+    voices: [
+      {
+        name: 'Seth (Legacy) (professional) - Resemble Voice',
+        value: 'a52c4efc',
+      },
+      {
+        name: 'Seth (professional) - Resemble Voice',
+        value: 'd3e61caf',
+      },
+    ],
+  },
+  {
+    value: 'en-GB',
+    name: 'En-GB',
+    voices: [
+      {
+        name: 'Beatrice Pendergast (professional) - Resemble Voice',
+        value: '00b1fd4e',
+      },
+      {
+        name: 'Ed Smart (professional) - Resemble Voice',
+        value: '0c755526',
+      },
+      {
+        name: 'Paula J (professional) - Resemble Voice',
+        value: '33e64cd2',
+      },
+    ],
+  },
+  {
+    value: 'en-us',
+    name: 'En-us',
+    voices: [
+      {
+        name: 'David (professional) - Resemble Voice',
+        value: '5bb13f03',
+      },
+    ],
+  },
+  {
+    value: 'en-US',
+    name: 'En-US',
+    voices: [
+      {
+        name: 'Adam Lofbomm (professional) - Resemble Voice',
+        value: '4e228dba',
+      },
+      {
+        name: 'Alex (professional) - Resemble Voice',
+        value: '41b99669',
+      },
+      {
+        name: 'Amelia (professional) - Resemble Voice',
+        value: 'ecbe5d97',
+      },
+      {
+        name: 'Andrew (rapid) - Resemble Marketplace',
+        value: 'd2f26a3e',
+      },
+      {
+        name: 'Annika (professional) - Resemble Voice',
+        value: 'b27f3cc0',
+      },
+      {
+        name: 'Arthur (professional) - Resemble Voice',
+        value: '9de11312',
+      },
+      {
+        name: 'Ash (professional) - Resemble Voice',
+        value: 'ee322483',
+      },
+      {
+        name: 'Aurora (professional) - Resemble Voice',
+        value: 'a72d9fca',
+      },
+      {
+        name: 'Austin (professional) - Resemble Voice',
+        value: '82a67e58',
+      },
+      {
+        name: 'Beth (Legacy) (professional) - Resemble Voice',
+        value: '25c7823f',
+      },
+      {
+        name: 'Beth (professional) - Resemble Voice',
+        value: 'fa66d263',
+      },
+      {
+        name: 'Blade (professional) - Resemble Voice',
+        value: '8bedd793',
+      },
+      {
+        name: 'Brandy Sky (professional) - Resemble Voice',
+        value: '79e2f1dc',
+      },
+      {
+        name: 'Brenley (professional) - Resemble Voice',
+        value: 'e6ec3ca4',
+      },
+      {
+        name: 'Britney (professional) - Resemble Voice',
+        value: 'e57e23ff',
+      },
+      {
+        name: 'Broadcast Joe (professional) - Resemble Voice',
+        value: '21e49584',
+      },
+      {
+        name: 'Carl Bishop (Angry) (professional) - Resemble Voice',
+        value: 'f06cd770',
+      },
+      {
+        name: 'Carl Bishop (Conversational) (professional) - Resemble Voice',
+        value: '7f40ff35',
+      },
+      {
+        name: 'Carl Bishop (Happy) (professional) - Resemble Voice',
+        value: '99751e42',
+      },
+      {
+        name: 'Carl Bishop (professional) - Resemble Voice',
+        value: '01bcc102',
+      },
+      {
+        name: 'Carl Bishop (Scared) (Legacy) (professional) - Resemble Voice',
+        value: '1dcf0222',
+      },
+      {
+        name: 'Carl Bishop (Scared) (professional) - Resemble Voice',
+        value: 'eacbc44f',
+      },
+      {
+        name: 'Charles (Legacy) (professional) - Resemble Voice',
+        value: '4c6d3da5',
+      },
+      {
+        name: 'Charles (professional) - Resemble Voice',
+        value: 'd79a5198',
+      },
+      {
+        name: 'Charlotte (professional) - Resemble Voice',
+        value: '96b91cf9',
+      },
+      {
+        name: 'Chris Whiting (professional) - Resemble Voice',
+        value: '95b7560a',
+      },
+      {
+        name: 'Cliff (professional) - Resemble Voice',
+        value: 'fcf8490c',
+      },
+      {
+        name: 'Connor (professional) - Resemble Voice',
+        value: 'a6131acf',
+      },
+      {
+        name: 'Deanna (professional) - Resemble Voice',
+        value: '0842fdf9',
+      },
+      {
+        name: 'Ember (professional) - Resemble Voice',
+        value: '55592656',
+      },
+      {
+        name: 'Gene Amore (professional) - Resemble Voice',
+        value: 'f2ea7aa0',
+      },
+      {
+        name: 'Harry Robinson (professional) - Resemble Voice',
+        value: '3c36d67d',
+      },
+      {
+        name: 'Helena (professional) - Resemble Voice',
+        value: 'ac948df2',
+      },
+      {
+        name: 'Hem (professional) - Resemble Voice',
+        value: 'b6edbe5f',
+      },
+      {
+        name: 'John (professional) - Resemble Voice',
+        value: 'ac48daeb',
+      },
+      {
+        name: 'Josh (professional) - Resemble Voice',
+        value: '987c99e9',
+      },
+      {
+        name: 'Julie Hoverson (professional) - Resemble Voice',
+        value: 'b119524c',
+      },
+      {
+        name: 'Justin (Legacy) (professional) - Resemble Voice',
+        value: 'b2d1bb75',
+      },
+      {
+        name: 'Justin (Meditative) (Legacy) (professional) - Resemble Voice',
+        value: '93ce0920',
+      },
+      {
+        name: 'Justin (Meditative) (professional) - Resemble Voice',
+        value: '2570000e',
+      },
+      {
+        name: 'Justin (professional) - Resemble Voice',
+        value: '9d513c17',
+      },
+      {
+        name: 'Karl Nordman (professional) - Resemble Voice',
+        value: 'da67f17e',
+      },
+      {
+        name: 'Kate (professional) - Resemble Voice',
+        value: '28b4cc5a',
+      },
+      {
+        name: 'Katya (professional) - Resemble Voice',
+        value: 'c9ee13b4',
+      },
+      {
+        name: 'Ken (professional) - Resemble Voice',
+        value: '3dbfbf3d',
+      },
+      {
+        name: 'Kessi (professional) - Resemble Voice',
+        value: '2211cb8c',
+      },
+      {
+        name: 'Little Ari (professional) - Resemble Voice',
+        value: '805adead',
+      },
+      {
+        name: 'Little Brittle (professional) - Resemble Voice',
+        value: '8a73f115',
+      },
+      {
+        name: 'Liz (professional) - Resemble Voice',
+        value: '4884d94a',
+      },
+      {
+        name: 'Lothar (professional) - Resemble Voice',
+        value: '78671217',
+      },
+      {
+        name: 'Luna (professional) - Resemble Voice',
+        value: 'ae8223ca',
+      },
+      {
+        name: 'Matt Weller (professional) - Resemble Voice',
+        value: 'f4da4639',
+      },
+      {
+        name: 'Maureen (Angry) (professional) - Resemble Voice',
+        value: '482babfc',
+      },
+      {
+        name: 'Maureen (Caring) (professional) - Resemble Voice',
+        value: 'b15e550f',
+      },
+      {
+        name: 'Maureen (Happy) (professional) - Resemble Voice',
+        value: '91947e5c',
+      },
+      {
+        name: 'Maureen (professional) - Resemble Voice',
+        value: '7d94218f',
+      },
+      {
+        name: 'Maureen (Sad) (professional) - Resemble Voice',
+        value: 'bca7481c',
+      },
+      {
+        name: 'Maureen (Scared) (professional) - Resemble Voice',
+        value: '251c9439',
+      },
+      {
+        name: 'Mauren (Announcer) (professional) - Resemble Voice',
+        value: 'e984fb89',
+      },
+      {
+        name: 'Melody (Legacy) (professional) - Resemble Voice',
+        value: '15be93bd',
+      },
+      {
+        name: 'Melody (professional) - Resemble Voice',
+        value: '1c49e774',
+      },
+      {
+        name: 'Mike (professional) - Resemble Voice',
+        value: '3a02dc40',
+      },
+      {
+        name: 'Niki (professional) - Resemble Voice',
+        value: 'db37643c',
+      },
+      {
+        name: 'Olga (professional) - Resemble Voice',
+        value: '07c1d6b5',
+      },
+      {
+        name: 'Olivia (Legacy) (professional) - Resemble Voice',
+        value: '405b58e3',
+      },
+      {
+        name: 'Olivia (professional) - Resemble Voice',
+        value: 'ef49f972',
+      },
+      {
+        name: 'Orion (professional) - Resemble Voice',
+        value: 'aa8053cc',
+      },
+      {
+        name: 'Pete (professional) - Resemble Voice',
+        value: '1864fd63',
+      },
+      {
+        name: 'Primrose (Legacy) (professional) - Resemble Voice',
+        value: '7c8e47ca',
+      },
+      {
+        name: 'Primrose (professional) - Resemble Voice',
+        value: '33eecc17',
+      },
+      {
+        name: 'Primrose (Whispering) (Legacy) (professional) - Resemble Voice',
+        value: 'a56c5c6f',
+      },
+      {
+        name: 'Primrose (Whispering) (professional) - Resemble Voice',
+        value: '28fcdf76',
+      },
+      {
+        name: 'Primrose (Winded) (Legacy) (professional) - Resemble Voice',
+        value: '6f9a77a4',
+      },
+      {
+        name: 'Primrose (Winded) (professional) - Resemble Voice',
+        value: '0097f246',
+      },
+      {
+        name: 'Professor Shaposhnikov (professional) - Resemble Voice',
+        value: '3f5fb9f1',
+      },
+      {
+        name: 'Radio Nikole (professional) - Resemble Voice',
+        value: '19eae884',
+      },
+      {
+        name: 'Richard Garifo (professional) - Resemble Voice',
+        value: '85ba84f2',
+      },
+      {
+        name: 'Rico (professional) - Resemble Voice',
+        value: '14ca34b3',
+      },
+      {
+        name: 'Robert (professional) - Resemble Voice',
+        value: '3e907bcc',
+      },
+      {
+        name: 'Rupert (rapid) - Resemble Voice',
+        value: '28f1626c',
+      },
+      {
+        name: 'Sam (professional) - Resemble Voice',
+        value: '0f2f9a7e',
+      },
+      {
+        name: 'Samantha (Legacy) (professional) - Resemble Voice',
+        value: '266bfae9',
+      },
+      {
+        name: 'Samantha (professional) - Resemble Voice',
+        value: 'e28236ee',
+      },
+      {
+        name: 'Siobhan (professional) - Resemble Voice',
+        value: 'af72c1ac',
+      },
+      {
+        name: 'Steve (Scared) (professional) - Resemble Voice',
+        value: 'aaa56e79',
+      },
+      {
+        name: 'Tanja (professional) - Resemble Voice',
+        value: 'adb84c77',
+      },
+      {
+        name: 'Tanja (Telephonic) (professional) - Resemble Voice',
+        value: '4f5a470b',
+      },
+      {
+        name: 'Tanja (Warm Word Weaver) (professional) - Resemble Voice',
+        value: 'abbbc383',
+      },
+      {
+        name: 'Tarkos (professional) - Resemble Voice',
+        value: '779842bf',
+      },
+      {
+        name: 'Tyler (professional) - Resemble Voice',
+        value: 'ff225977',
+      },
+      {
+        name: 'Vicky (professional) - Resemble Voice',
+        value: 'f453b918',
+      },
+      {
+        name: 'Vivian (Legacy) (professional) - Resemble Voice',
+        value: 'bed1044d',
+      },
+      {
+        name: 'Vivian (professional) - Resemble Voice',
+        value: '1ff0045f',
+      },
+      {
+        name: 'William (Whispering) (Legacy) (professional) - Resemble Voice',
+        value: '79eb7953',
+      },
+      {
+        name: 'William (Whispering) (professional) - Resemble Voice',
+        value: 'e2180df0',
+      },
+      {
+        name: 'Willow (Whispering) (professional) - Resemble Voice',
+        value: 'f2906c4a',
+      },
+      {
+        name: 'Willow II (Whispering) (professional) - Resemble Voice',
+        value: 'c815cd7a',
+      },
+    ],
+  },
+];

lib/utils/speech-utils.js

@@ -5,6 +5,8 @@ const sdk = require('microsoft-cognitiveservices-speech-sdk');
 const { SpeechClient } = require('@soniox/soniox-node');
 const fs = require('fs');
 const { AssemblyAI } = require('assemblyai');
+const Houndify = require('houndify');
+const { GladiaClient } = require('@gladiaio/sdk');
 const {decrypt, obscureKey} = require('./encrypt-decrypt');
 const { RealtimeSession } = require('speechmatics');
 
@@ -21,6 +23,7 @@ const TtsPlayHtLanguagesVoices = require('./speech-data/tts-playht');
 const TtsVerbioLanguagesVoices = require('./speech-data/tts-verbio');
 const TtsInworldLanguagesVoices = require('./speech-data/tts-inworld');
 const ttsCartesia = require('./speech-data/tts-cartesia');
+const TtsResembleLanguagesVoices = require('./speech-data/tts-resemble');
 
 const TtsModelDeepgram = require('./speech-data/tts-model-deepgram');
 const TtsLanguagesDeepgram = require('./speech-data/tts-deepgram');
@@ -44,9 +47,11 @@ const SttCobaltLanguagesVoices = require('./speech-data/stt-cobalt');
 const SttSonioxLanguagesVoices = require('./speech-data/stt-soniox');
 const SttSpeechmaticsLanguagesVoices = require('./speech-data/stt-speechmatics');
 const SttAssemblyaiLanguagesVoices = require('./speech-data/stt-assemblyai');
+const SttHoundifyLanguagesVoices = require('./speech-data/stt-houndify');
 const SttVoxistLanguagesVoices = require('./speech-data/stt-voxist');
 const SttVerbioLanguagesVoices = require('./speech-data/stt-verbio');
 const SttOpenaiLanguagesVoices = require('./speech-data/stt-openai');
+const SttGladiaLanguagesVoices = require('./speech-data/stt-gladia');
 
 
 const SttModelOpenai = require('./speech-data/stt-model-openai');
@@ -167,6 +172,65 @@ const testGoogleStt = async(logger, credentials) => {
   }
 };
 
+const testGladiaStt = async(logger, credentials) => {
+  const {api_key} = credentials;
+
+  try {
+    const gladiaClient = new GladiaClient({
+      apiKey: api_key,
+    });
+    const gladiaConfig = {
+      model: 'solaria-1',
+      encoding: 'wav/pcm',
+      sample_rate: 16000,
+      bit_depth: 16,
+      channels: 1,
+      language_config: {
+        languages: ['en'],
+        code_switching: false,
+      },
+    };
+    // Start the live session
+    const liveSession = gladiaClient.liveV2().startSession(gladiaConfig);
+    // Read the test audio file
+    const audioBuffer = fs.readFileSync(`${__dirname}/../../data/test_audio.wav`);
+
+    // Wait for final transcript
+    return new Promise((resolve, reject) => {
+      liveSession.on('message', (message) => {
+        if (message.type === 'transcript' && message.data.is_final) {
+          logger.debug(`${message.data.id}: ${message.data.utterance.text}`);
+          liveSession.stopRecording();
+          resolve(message.data.utterance.text);
+        }
+      });
+
+      liveSession.on('error', (error) => {
+        logger.error({error}, 'Gladia Live STT error');
+        reject(error);
+      });
+
+      // Send audio in chunks
+      const chunkSize = 1024;
+      for (let i = 0; i < audioBuffer.length; i += chunkSize) {
+        const chunk = audioBuffer.slice(i, i + chunkSize);
+        liveSession.sendAudio(chunk);
+      }
+      // Stop recording after sending all audio
+      liveSession.stopRecording();
+
+      // Set a timeout to prevent hanging
+      setTimeout(() => {
+        reject(new Error('Gladia STT test timeout'));
+      }, 30000); // 30 second timeout
+    });
+
+  } catch (error) {
+    logger.error({error}, 'Failed to create Gladia Live STT session');
+    throw error;
+  }
+};
+
 const testDeepgramStt = async(logger, credentials) => {
   const {api_key, deepgram_stt_uri, deepgram_stt_use_tls} = credentials;
   const deepgram = new Deepgram(api_key, deepgram_stt_uri, deepgram_stt_uri && deepgram_stt_use_tls);
@@ -315,8 +379,8 @@ const testWellSaidTts = async(logger, credentials) => {
 };
 
 const testElevenlabs = async(logger, credentials) => {
-  const {api_key, model_id} = credentials;
-  const response = await fetch('https://api.elevenlabs.io/v1/text-to-speech/21m00Tcm4TlvDq8ikWAM', {
+  const {api_key, model_id, api_uri} = credentials;
+  const response = await fetch(`https://${api_uri || 'api.elevenlabs.io'}/v1/text-to-speech/21m00Tcm4TlvDq8ikWAM`, {
     method: 'POST',
     headers: {
       'xi-api-key': api_key,
@@ -424,6 +488,24 @@ const testWhisper = async(logger, synthAudio, credentials) => {
   }
 };
 
+const testResembleTTS = async(logger, synthAudio, credentials) => {
+  try {
+    await synthAudio({increment: () => {}, histogram: () => {}},
+      {
+        vendor: 'resemble',
+        credentials,
+        language: 'en-US',
+        voice: '3f5fb9f1',
+        text: 'Hi there and welcome to jambones!',
+        renderForCaching: true
+      }
+    );
+  } catch (err) {
+    logger.info({err}, 'synth resemble returned error');
+    throw err;
+  }
+};
+
 const testDeepgramTTS = async(logger, synthAudio, credentials) => {
   try {
     await synthAudio({increment: () => {}, histogram: () => {}},
@@ -576,6 +658,72 @@ const testAssemblyStt = async(logger, credentials) => {
   });
 };
 
+const testHoundifyStt = async(logger, credentials) => {
+  const {client_id, client_key, user_id} = credentials;
+
+  return new Promise((resolve, reject) => {
+    try {
+      // Read the test audio file
+      const audioBuffer = fs.readFileSync(`${__dirname}/../../data/test_audio.wav`);
+
+      // Create VoiceRequest for speech-to-text testing
+      const voiceRequest = new Houndify.VoiceRequest({
+        // Your Houndify Client ID and Key
+        clientId: client_id,
+        clientKey: client_key,
+
+        // Request info
+        requestInfo: {
+          UserID: user_id || 'test_user',
+          Latitude: 37.388309,
+          Longitude: -121.973968
+        },
+
+        // Audio format configuration
+        sampleRate: 16000,
+        enableVAD: true,
+
+        // Response and error handlers
+        onResponse: function(response, info) {
+          logger.debug({response, info}, 'Houndify STT response received');
+          if (response && response.AllResults && response.AllResults.length > 0) {
+            resolve(response);
+          } else {
+            reject(new Error('No transcription results received'));
+          }
+        },
+
+        onError: function(err, info) {
+          logger.error({err, info}, 'Houndify STT error');
+          reject(err);
+        },
+
+        onRecordingStarted: function() {
+          logger.debug('Houndify recording started');
+        },
+
+        onRecordingStopped: function() {
+          logger.debug('Houndify recording stopped');
+        }
+      });
+
+      // Send audio in chunks (VoiceRequest automatically starts when you write data)
+      const chunkSize = 1024;
+      for (let i = 0; i < audioBuffer.length; i += chunkSize) {
+        const chunk = audioBuffer.slice(i, i + chunkSize);
+        voiceRequest.write(chunk);
+      }
+
+      // End the request
+      voiceRequest.end();
+
+    } catch (error) {
+      logger.error({error}, 'Failed to create Houndify VoiceRequest');
+      reject(error);
+    }
+  });
+};
+
 const testVoxistStt = async(logger, credentials) => {
   const {api_key} = credentials;
   const response = await fetch('https://api-asr.voxist.com/clients', {
@@ -636,7 +784,6 @@ function decryptCredential(obj, credential, logger, isObscureKey = true) {
     obj.role_arn = o.role_arn;
     obj.secret_access_key = isObscureKey ? obscureKey(o.secret_access_key) : o.secret_access_key;
     obj.aws_region = o.aws_region;
-    logger.info({obj, o}, 'retrieving aws speech credential');
   }
   else if ('microsoft' === obj.vendor) {
     const o = JSON.parse(decrypt(credential));
@@ -648,7 +795,6 @@ function decryptCredential(obj, credential, logger, isObscureKey = true) {
     obj.use_custom_stt = o.use_custom_stt;
     obj.custom_stt_endpoint = o.custom_stt_endpoint;
     obj.custom_stt_endpoint_url = o.custom_stt_endpoint_url;
-    logger.info({obj, o}, 'retrieving azure speech credential');
   }
   else if ('wellsaid' === obj.vendor) {
     const o = JSON.parse(decrypt(credential));
@@ -669,6 +815,14 @@ function decryptCredential(obj, credential, logger, isObscureKey = true) {
     obj.deepgram_tts_uri = o.deepgram_tts_uri;
     obj.model_id = o.model_id;
   }
+  else if ('deepgramflux' === obj.vendor) {
+    const o = JSON.parse(decrypt(credential));
+    obj.api_key = isObscureKey ? obscureKey(o.api_key) : o.api_key;
+  }
+  else if ('gladia' === obj.vendor) {
+    const o = JSON.parse(decrypt(credential));
+    obj.api_key = isObscureKey ? obscureKey(o.api_key) : o.api_key;
+  }
   else if ('ibm' === obj.vendor) {
     const o = JSON.parse(decrypt(credential));
     obj.tts_api_key = isObscureKey ? obscureKey(o.tts_api_key) : o.tts_api_key;
@@ -693,6 +847,7 @@ function decryptCredential(obj, credential, logger, isObscureKey = true) {
     const o = JSON.parse(decrypt(credential));
     obj.api_key = isObscureKey ? obscureKey(o.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) {
     const o = JSON.parse(decrypt(credential));
@@ -727,6 +882,16 @@ function decryptCredential(obj, credential, logger, isObscureKey = true) {
     const o = JSON.parse(decrypt(credential));
     obj.api_key = isObscureKey ? obscureKey(o.api_key) : o.api_key;
     obj.service_version = o.service_version;
+  } else if ('houndify' === obj.vendor) {
+    const o = JSON.parse(decrypt(credential));
+    obj.client_key = isObscureKey ? obscureKey(o.client_key) : o.client_key;
+    obj.client_id = o.client_id;
+    obj.user_id = o.user_id;
+  } else if ('resemble' === obj.vendor) {
+    const o = JSON.parse(decrypt(credential));
+    obj.api_key = isObscureKey ? obscureKey(o.api_key) : o.api_key;
+    obj.resemble_tts_uri = o.resemble_tts_uri;
+    obj.resemble_tts_use_tls = o.resemble_tts_use_tls;
   } else if ('voxist' === obj.vendor) {
     const o = JSON.parse(decrypt(credential));
     obj.api_key = isObscureKey ? obscureKey(o.api_key) : o.api_key;
@@ -781,6 +946,8 @@ async function getLanguagesAndVoicesForVendor(logger, vendor, credential, getTts
       return await getLanguagesVoicesForNuane(credential, getTtsVoices, logger);
     case 'deepgram':
       return await getLanguagesVoicesForDeepgram(credential, getTtsVoices, logger);
+    case 'gladia':
+      return await getLanguagesVoicesForGladia(credential, getTtsVoices, logger);
     case 'ibm':
       return await getLanguagesVoicesForIbm(credential, getTtsVoices, logger);
     case 'nvidia':
@@ -797,8 +964,12 @@ async function getLanguagesAndVoicesForVendor(logger, vendor, credential, getTts
       return await getLanguagesVoicesForRimelabs(credential, getTtsVoices, logger);
     case 'inworld':
       return await getLanguagesVoicesForInworld(credential, getTtsVoices, logger);
+    case 'resemble':
+      return await getLanguagesAndVoicesForResemble(credential, getTtsVoices, logger);
     case 'assemblyai':
       return await getLanguagesVoicesForAssemblyAI(credential, getTtsVoices, logger);
+    case 'houndify':
+      return await getLanguagesVoicesForHoundify(credential, getTtsVoices, logger);
     case 'voxist':
       return await getLanguagesVoicesForVoxist(credential, getTtsVoices, logger);
     case 'whisper':
@@ -907,8 +1078,30 @@ async function getLanguagesVoicesForDeepgram(credential, getTtsVoices, logger) {
         logger.error({response}, 'Error fetching Deepgram voices');
         throw new Error('failed to list voices');
       }
-      const {stt, tts} = await response.json();
-      let sttLangs = SttDeepgramLanguagesVoices;
+      const {stt, tts, languages} = await response.json();
+      // Helper function to get language name
+      const getLanguageName = (langCode) => {
+        if (languages && languages[langCode]) {
+          return languages[langCode];
+        }
+        const existingLang = SttDeepgramLanguagesVoices.find((l) => l.value === langCode);
+        return existingLang ? existingLang.name : capitalizeFirst(langCode);
+      };
+      // Collect unique languages from selected models
+      const allSttLanguages = new Set();
+      const modelsToProcess = model_id ?
+        stt.filter((m) => m.canonical_name === model_id) :
+        stt;
+      modelsToProcess.forEach((model) => {
+        if (model.languages && Array.isArray(model.languages)) {
+          model.languages.forEach((lang) => allSttLanguages.add(lang));
+        }
+      });
+      // Convert to expected format
+      const sttLangs = Array.from(allSttLanguages).map((langCode) => ({
+        name: getLanguageName(langCode),
+        value: langCode
+      })).sort((a, b) => a.name.localeCompare(b.name));
       const sttModels = Array.from(
         new Map(
           stt.map((m) => [m.canonical_name, { name: capitalizeFirst(m.canonical_name), value: m.canonical_name }])
@@ -919,16 +1112,6 @@ async function getLanguagesVoicesForDeepgram(credential, getTtsVoices, logger) {
           tts.map((m) => [m.canonical_name, { name: capitalizeFirst(m.canonical_name), value: m.canonical_name }])
         ).values()
       ).sort((a, b) => a.name.localeCompare(b.name));
-      // if model_id is not provided, return all models, all voices, all languages
-      if (!model_id) {
-        return tranform(TtsLanguagesDeepgram, sttLangs, ttsModels, sttModels);
-      }
-
-      const selectedSttModel = stt.find((m) => m.canonical_name === model_id);
-      const selectedSttLangs = selectedSttModel ? selectedSttModel.languages : [];
-      sttLangs = SttDeepgramLanguagesVoices.filter((l) => {
-        return selectedSttLangs.includes(l.value);
-      });
       return tranform(TtsLanguagesDeepgram, sttLangs, ttsModels, sttModels);
     }
   }
@@ -936,6 +1119,11 @@ async function getLanguagesVoicesForDeepgram(credential, getTtsVoices, logger) {
     TtsModelDeepgram, sttModelDeepgram.sort((a, b) => a.name.localeCompare(b.name)));
 }
 
+async function getLanguagesVoicesForGladia(credential, getTtsVoices, logger) {
+  return tranform(undefined, SttGladiaLanguagesVoices.sort((a, b) => a.name.localeCompare(b.name)),
+    undefined, undefined);
+}
+
 async function getLanguagesVoicesForIbm(credential, getTtsVoices, logger) {
   if (credential) {
     try {
@@ -974,10 +1162,12 @@ async function getLanguagesVoicesForElevenlabs(credential) {
       'xi-api-key': credential.api_key
     };
 
-    const getModelPromise = fetch('https://api.elevenlabs.io/v1/models', {
+    const api_uri = credential.api_uri || 'api.elevenlabs.io';
+
+    const getModelPromise = fetch(`https://${api_uri}/v1/models`, {
       headers
     });
-    const getVoicePromise = fetch('https://api.elevenlabs.io/v1/voices', {
+    const getVoicePromise = fetch(`https://${api_uri}/v1/voices`, {
       headers
     });
     const [langResp, voiceResp] = await Promise.all([getModelPromise, getVoicePromise]);
@@ -1206,6 +1396,10 @@ async function getLanguagesVoicesForAssemblyAI(credential) {
   return tranform(undefined, SttAssemblyaiLanguagesVoices);
 }
 
+async function getLanguagesVoicesForHoundify(credential) {
+  return tranform(undefined, SttHoundifyLanguagesVoices);
+}
+
 async function getLanguagesVoicesForVoxist(credential) {
   return tranform(undefined, SttVoxistLanguagesVoices);
 }
@@ -1238,6 +1432,82 @@ async function getLanguagesVoicesForVerbio(credentials, getTtsVoices, logger) {
   }
 }
 
+async function getLanguagesAndVoicesForResemble(credential, getTtsVoices, logger) {
+  if (credential) {
+    try {
+      const {api_key} = credential;
+      let allVoices = [];
+      let page = 1;
+      let hasMorePages = true;
+      // Fetch all pages of voices
+      while (hasMorePages) {
+        const response = await fetch(`https://app.resemble.ai/api/v2/voices?page=${page}&page_size=100`, {
+          headers: {
+            'Authorization': `Token token=${api_key}`,
+            'Accept': 'application/json'
+          }
+        });
+
+        if (!response.ok) {
+          throw new Error('failed to list voices');
+        }
+
+        const data = await response.json();
+
+        if (!data.success) {
+          throw new Error('API returned unsuccessful response');
+        }
+
+        allVoices = allVoices.concat(data.items);
+
+        // Check if there are more pages
+        hasMorePages = page < data.num_pages;
+        page++;
+      }
+
+      // Filter only finished voices that support text_to_speech
+      const availableVoices = allVoices.filter((voice) =>
+        voice.status === 'finished' &&
+        voice.component_status?.text_to_speech?.status === 'ready'
+      );
+
+      // Group voices by language
+      const ttsVoices = availableVoices.reduce((acc, voice) => {
+        const languageCode = voice.default_language || 'en-US';
+        const existingLanguage = acc.find((lang) => lang.value === languageCode);
+
+        const voiceEntry = {
+          name: `${voice.name} (${voice.voice_type}) - ${voice.source}`,
+          value: voice.uuid
+        };
+
+        if (existingLanguage) {
+          existingLanguage.voices.push(voiceEntry);
+        } else {
+
+          acc.push({
+            value: languageCode,
+            name: capitalizeFirst(languageCode),
+            voices: [voiceEntry]
+          });
+        }
+
+        return acc;
+      }, []);
+      // Sort languages and voices
+      ttsVoices.sort((a, b) => a.name.localeCompare(b.name));
+      ttsVoices.forEach((lang) => {
+        lang.voices.sort((a, b) => a.name.localeCompare(b.name));
+      });
+      return tranform(ttsVoices);
+    } catch (err) {
+      logger.info('Error while fetching Resemble languages, voices, return predefined values', err);
+    }
+  }
+
+  return tranform(TtsResembleLanguagesVoices);
+}
+
 function tranform(tts, stt, models, sttModels) {
   return {
     ...(tts && {tts}),
@@ -1508,6 +1778,7 @@ module.exports = {
   testNuanceTts,
   testNuanceStt,
   testDeepgramStt,
+  testGladiaStt,
   testIbmTts,
   testIbmStt,
   testSonioxStt,
@@ -1526,5 +1797,7 @@ module.exports = {
   testSpeechmaticsStt,
   testCartesia,
   testVoxistStt,
-  testOpenAiStt
+  testOpenAiStt,
+  testResembleTTS,
+  testHoundifyStt
 };

package.json

@@ -1,6 +1,6 @@
 {
   "name": "jambonz-api-server",
-  "version": "0.9.4",
+  "version": "0.9.5",
   "description": "",
   "main": "app.js",
   "scripts": {
@@ -25,15 +25,16 @@
     "@aws-sdk/client-transcribe": "^3.549.0",
     "@azure/storage-blob": "^12.17.0",
     "@deepgram/sdk": "^1.21.0",
+    "@gladiaio/sdk": "^0.5.2",
     "@google-cloud/speech": "^6.5.0",
     "@google-cloud/storage": "^7.9.0",
-    "@jambonz/db-helpers": "^0.9.12",
+    "@jambonz/db-helpers": "^0.9.18",
     "@jambonz/lamejs": "^1.2.2",
     "@jambonz/mw-registrar": "^0.2.7",
-    "@jambonz/realtimedb-helpers": "^0.8.14",
-    "@jambonz/speech-utils": "^0.2.13",
+    "@jambonz/realtimedb-helpers": "^0.8.15",
+    "@jambonz/speech-utils": "^0.2.26",
     "@jambonz/time-series": "^0.2.8",
-    "@jambonz/verb-specifications": "^0.0.106",
+    "@jambonz/verb-specifications": "^0.0.118",
     "@soniox/soniox-node": "^1.2.2",
     "ajv": "^8.17.1",
     "argon2": "^0.40.1",
@@ -44,6 +45,7 @@
     "express-rate-limit": "^7.2.0",
     "form-data": "^4.0.0",
     "helmet": "^7.1.0",
+    "houndify": "^3.1.14",
     "ibm-watson": "^9.0.1",
     "is-valid-hostname": "^1.0.2",
     "jsonwebtoken": "^9.0.2",

test/speech-credentials.js

@@ -656,7 +656,8 @@ test('speech credentials tests', async(t) => {
         use_for_stt: true,
         use_for_tts: false,
         api_key: 'asdasdasdasddsadasda',
-        model_id: 'eleven_multilingual_v2'
+        model_id: 'eleven_multilingual_v2',
+        api_uri: 'api.elevenlabs.io'
       }
     });
     t.ok(result.statusCode === 201, 'successfully added speech credential for elevenlabs');
@@ -805,6 +806,29 @@ test('speech credentials tests', async(t) => {
     });
     t.ok(result.statusCode === 204, 'successfully deleted speech credential');
 
+    /* add a credential for houndify */
+    result = await request.post(`/Accounts/${account_sid}/SpeechCredentials`, {
+      resolveWithFullResponse: true,
+      auth: authUser,
+      json: true,
+      body: {
+        vendor: 'houndify',
+        use_for_stt: true,
+        client_key: "ClientKey",
+        client_id: "ClientID",
+        user_id: "test_user"
+      }
+    });
+    t.ok(result.statusCode === 201, 'successfully added speech credential for houndify');
+    const houndifySid = result.body.sid;
+
+    /* delete the credential */
+    result = await request.delete(`/Accounts/${account_sid}/SpeechCredentials/${houndifySid}`, {
+      auth: authUser,
+      resolveWithFullResponse: true,
+    });
+    t.ok(result.statusCode === 204, 'successfully deleted speech credential');
+
     /* add a credential for Voxist */
     result = await request.post(`/Accounts/${account_sid}/SpeechCredentials`, {
       resolveWithFullResponse: true,
@@ -902,6 +926,72 @@ test('speech credentials tests', async(t) => {
     });
     t.ok(result.statusCode === 204, 'successfully deleted speech credential');
 
+    /* add a credential for resemble */
+    result = await request.post(`/Accounts/${account_sid}/SpeechCredentials`, {
+      resolveWithFullResponse: true,
+      auth: authUser,
+      json: true,
+      body: {
+        vendor: 'resemble',
+        use_for_tts: true,
+        use_for_stt: false,
+        api_key: 'api_key',
+      }
+    });
+    t.ok(result.statusCode === 201, 'successfully added speech credential for Resemble');
+    const resembleSid = result.body.sid;
+
+    /* delete the credential */
+    result = await request.delete(`/Accounts/${account_sid}/SpeechCredentials/${resembleSid}`, {
+      auth: authUser,
+      resolveWithFullResponse: true,
+    });
+    t.ok(result.statusCode === 204, 'successfully deleted speech credential for Resemble');
+
+    /* add a credential for deepgram river */
+    result = await request.post(`/Accounts/${account_sid}/SpeechCredentials`, {
+      resolveWithFullResponse: true,
+      auth: authUser,
+      json: true,
+      body: {
+        vendor: 'deepgramflux',
+        use_for_tts: false,
+        use_for_stt: true,
+        api_key: 'api_key',
+      }
+    });
+    t.ok(result.statusCode === 201, 'successfully added speech credential for Verbio');
+    const deepgramfluxSid = result.body.sid;
+
+    /* delete the credential */
+    result = await request.delete(`/Accounts/${account_sid}/SpeechCredentials/${deepgramfluxSid}`, {
+      auth: authUser,
+      resolveWithFullResponse: true,
+    });
+    t.ok(result.statusCode === 204, 'successfully deleted speech credential deepgramflux');
+
+    /* add a credential for gladia */
+    result = await request.post(`/Accounts/${account_sid}/SpeechCredentials`, {
+      resolveWithFullResponse: true,
+      auth: authUser,
+      json: true,
+      body: {
+        vendor: 'gladia',
+        use_for_tts: false,
+        use_for_stt: true,
+        api_key: 'api_key',
+      }
+    });
+    t.ok(result.statusCode === 201, 'successfully added speech credential for Gladia');
+    const gladiaSid = result.body.sid;
+
+    /* delete the credential */
+    result = await request.delete(`/Accounts/${account_sid}/SpeechCredentials/${gladiaSid}`, {
+      auth: authUser,
+      resolveWithFullResponse: true,
+    });
+    t.ok(result.statusCode === 204, 'successfully deleted speech credential for Gladia');
+
     /* Check google supportedLanguagesAndVoices */
     result = await request.get(`/Accounts/${account_sid}/SpeechCredentials/speech/supportedLanguagesAndVoices?vendor=google`, {
       resolveWithFullResponse: true,
@@ -1037,6 +1127,15 @@ test('speech credentials tests', async(t) => {
     t.ok(result.body.tts.length !== 0, 'successfully get whisper supported languages and voices');
     t.ok(result.body.models.length !== 0, 'successfully get whisper supported languages and voices');
 
+    /* Check gladia supportedLanguagesAndVoices */
+    result = await request.get(`/Accounts/${account_sid}/SpeechCredentials/speech/supportedLanguagesAndVoices?vendor=gladia`, {
+      resolveWithFullResponse: true,
+      simple: false,
+      auth: authAdmin,
+      json: true,
+    });
+    t.ok(result.body.stt.length !== 0, 'successfully get gladia supported languages and voices');
+
     await deleteObjectBySid(request, '/Accounts', account_sid);
     await deleteObjectBySid(request, '/ServiceProviders', service_provider_sid);
     t.end();