update to latest speech-utils

* support playht3.0 languages

* update speech utils version

.github/workflows/ci.yml

@@ -7,11 +7,21 @@ jobs:
     runs-on: ubuntu-latest
     steps:
       - uses: actions/checkout@v3
+      - name: Install Docker Compose
+        run: |
+          sudo curl -L "https://github.com/docker/compose/releases/download/1.29.2/docker-compose-$(uname -s)-$(uname -m)" -o /usr/local/bin/docker-compose
+          sudo chmod +x /usr/local/bin/docker-compose
+          docker-compose --version
       - uses: actions/setup-node@v3
         with:
           node-version: lts/*
       - run: npm install
       - run: npm run jslint
+      - name: Install Docker Compose
+        run: |
+          sudo curl -L "https://github.com/docker/compose/releases/download/1.29.2/docker-compose-$(uname -s)-$(uname -m)" -o /usr/local/bin/docker-compose
+          sudo chmod +x /usr/local/bin/docker-compose
+          docker-compose --version
       - run: npm test
 
  

db/jambones-sql.sql

@@ -358,7 +358,9 @@ CREATE TABLE system_information
 (
 domain_name VARCHAR(255),
 sip_domain_name VARCHAR(255),
-monitoring_domain_name VARCHAR(255)
+monitoring_domain_name VARCHAR(255),
+private_network_cidr VARCHAR(8192),
+log_level ENUM('info', 'debug') NOT NULL DEFAULT 'info'
 );
 
 CREATE TABLE users
@@ -552,6 +554,7 @@ siprec_hook_sid CHAR(36),
 record_all_calls BOOLEAN NOT NULL DEFAULT false,
 record_format VARCHAR(16) NOT NULL DEFAULT 'mp3',
 bucket_credential VARCHAR(8192) COMMENT 'credential used to authenticate with storage service',
+enable_debug_log BOOLEAN NOT NULL DEFAULT false,
 PRIMARY KEY (account_sid)
 ) COMMENT='An enterprise that uses the platform for comm services';
 

db/upgrade-jambonz-db.js

@@ -197,6 +197,11 @@ const sql = {
     'ALTER TABLE applications MODIFY COLUMN speech_synthesis_voice VARCHAR(256)',
     'ALTER TABLE applications MODIFY COLUMN fallback_speech_synthesis_voice VARCHAR(256)',
     'ALTER TABLE sip_gateways ADD COLUMN use_sips_scheme BOOLEAN NOT NULL DEFAULT 0',
+  ],
+  9002: [
+    'ALTER TABLE system_information ADD COLUMN private_network_cidr VARCHAR(8192)',
+    'ALTER TABLE system_information ADD COLUMN log_level ENUM(\'info\', \'debug\') NOT NULL DEFAULT \'info\'',
+    'ALTER TABLE accounts ADD COLUMN enable_debug_log BOOLEAN NOT NULL DEFAULT false',
   ]
 };
 
@@ -230,6 +235,7 @@ const doIt = async() => {
         if (val < 8004) upgrades.push(...sql['8004']);
         if (val < 8005) upgrades.push(...sql['8005']);
         if (val < 9000) upgrades.push(...sql['9000']);
+        if (val < 9002) upgrades.push(...sql['9002']);
 
         // perform all upgrades
         logger.info({upgrades}, 'applying schema upgrades..');

lib/models/speech-credential.js

@@ -23,10 +23,11 @@ class SpeechCredential extends Model {
   static async getSpeechCredentialsByVendorAndLabel(service_provider_sid, account_sid, vendor, label) {
     let sql;
     if (account_sid) {
-      sql = `SELECT * FROM speech_credentials WHERE account_sid = ? AND vendor = ? ${label ? 'AND label = ?' : ''}`;
+      sql = `SELECT * FROM speech_credentials WHERE account_sid = ? AND vendor = ?
+        AND label ${label ? '= ?' : 'is NULL'}`;
     } else {
       sql = `SELECT * FROM speech_credentials WHERE service_provider_sid = ? AND vendor = ?
-        ${label ? 'AND label = ?' : ''}`;
+        AND label ${label ? '= ?' : 'is NULL'}`;
     }
     const [rows] = await promisePool.query(sql, [account_sid ? account_sid : service_provider_sid, vendor, label]);
     return rows;

lib/models/system-information.js

@@ -33,6 +33,10 @@ SystemInformation.fields = [
     name: 'monitoring_domain_name',
     type: 'string',
   },
+  {
+    name: 'private_network_cidr',
+    type: 'string',
+  },
 ];
 
 module.exports = SystemInformation;

lib/routes/api/sip-gateways.js

@@ -41,6 +41,7 @@ const checkUserScope = async(req, voip_carrier_sid) => {
 
 const validate = async(req, sid) => {
   const {lookupSipGatewayBySid} = req.app.locals;
+  const {netmask} = req.body;
   let voip_carrier_sid;
 
   if (sid) {
@@ -52,6 +53,12 @@ const validate = async(req, sid) => {
     voip_carrier_sid = req.body.voip_carrier_sid;
     if (!voip_carrier_sid) throw new DbErrorBadRequest('missing voip_carrier_sid');
   }
+  if (netmask &&
+    process.env.JAMBONZ_MIN_GATEWAY_NETMASK &&
+    parseInt(netmask) < process.env.JAMBONZ_MIN_GATEWAY_NETMASK) {
+    throw new DbErrorBadRequest(
+      `netmask required to have value equal or greater than ${process.env.JAMBONZ_MIN_GATEWAY_NETMASK}`);
+  }
   await checkUserScope(req, voip_carrier_sid);
 };
 

lib/routes/api/speech-credentials.js

@@ -10,7 +10,8 @@ const {decryptCredential, testWhisper, testDeepgramTTS,
   testPlayHT,
   testRimelabs,
   testVerbioTts,
-  testVerbioStt} = require('../../utils/speech-utils');
+  testVerbioStt,
+  testSpeechmaticsStt} = require('../../utils/speech-utils');
 const {DbErrorUnprocessableRequest, DbErrorForbidden, DbErrorBadRequest} = require('../../utils/errors');
 const {
   testGoogleTts,
@@ -122,8 +123,10 @@ const encryptCredential = (obj) => {
     secret,
     nuance_tts_uri,
     nuance_stt_uri,
+    speechmatics_stt_uri,
     deepgram_stt_uri,
     deepgram_stt_use_tls,
+    deepgram_tts_uri,
     use_custom_tts,
     custom_tts_endpoint,
     custom_tts_endpoint_url,
@@ -204,10 +207,10 @@ const encryptCredential = (obj) => {
 
     case 'deepgram':
       // API key is optional if onprem
-      if (!deepgram_stt_uri) {
+      if (!deepgram_stt_uri || !deepgram_tts_uri) {
         assert(api_key, 'invalid deepgram speech credential: api_key is required');
       }
-      const deepgramData = JSON.stringify({api_key, deepgram_stt_uri, deepgram_stt_use_tls});
+      const deepgramData = JSON.stringify({api_key, deepgram_stt_uri, deepgram_stt_use_tls, deepgram_tts_uri});
       return encrypt(deepgramData);
 
     case 'ibm':
@@ -235,6 +238,12 @@ const encryptCredential = (obj) => {
       const elevenlabsData = JSON.stringify({api_key, model_id, options});
       return encrypt(elevenlabsData);
 
+    case 'speechmatics':
+      assert(api_key, 'invalid speechmatics speech credential: api_key is required');
+      assert(speechmatics_stt_uri, 'invalid speechmatics speech credential: speechmatics_stt_uri is required');
+      const speechmaticsData = JSON.stringify({api_key, speechmatics_stt_uri, options});
+      return encrypt(speechmaticsData);
+
     case 'playht':
       assert(api_key, 'invalid playht speech credential: api_key is required');
       assert(user_id, 'invalid playht speech credential: user_id is required');
@@ -458,6 +467,7 @@ router.put('/:sid', async(req, res) => {
           options,
           deepgram_stt_uri,
           deepgram_stt_use_tls,
+          deepgram_tts_uri,
           engine_version
         } = req.body;
 
@@ -485,6 +495,7 @@ router.put('/:sid', async(req, res) => {
           options,
           deepgram_stt_uri,
           deepgram_stt_use_tls,
+          deepgram_tts_uri,
           engine_version
         };
         logger.info({o, newCred}, 'updating speech credential with this new credential');
@@ -765,6 +776,18 @@ router.get('/:sid/test', async(req, res) => {
           SpeechCredential.ttsTestResult(sid, false);
         }
       }
+    } else if (cred.vendor === 'speechmatics') {
+      const {api_key} = credential;
+      if (cred.use_for_stt) {
+        try {
+          await testSpeechmaticsStt(logger, {api_key});
+          results.stt.status = 'ok';
+          SpeechCredential.ttsTestResult(sid, true);
+        } catch (err) {
+          results.stt = {status: 'fail', reason: err.message};
+          SpeechCredential.ttsTestResult(sid, false);
+        }
+      }
     } else if (cred.vendor === 'playht') {
       if (cred.use_for_tts) {
         try {
@@ -852,7 +875,7 @@ router.get('/:sid/test', async(req, res) => {
 router.get('/speech/supportedLanguagesAndVoices', async(req, res) => {
   const {logger, getTtsVoices} = req.app.locals;
   try {
-    const {vendor, label} = req.query;
+    const {vendor, label, create_new} = req.query;
     if (!vendor) {
       throw new DbErrorBadRequest('vendor is required');
     }
@@ -860,7 +883,7 @@ router.get('/speech/supportedLanguagesAndVoices', async(req, res) => {
     const service_provider_sid = req.user.service_provider_sid ||
       req.body.service_provider_sid || parseServiceProviderSid(req);
 
-    const credentials = await SpeechCredential.getSpeechCredentialsByVendorAndLabel(
+    const credentials = create_new ? null : await SpeechCredential.getSpeechCredentialsByVendorAndLabel(
       service_provider_sid, account_sid, vendor, label);
     const tmp = credentials && credentials.length > 0 ? credentials[0] : null;
     const cred = tmp ? JSON.parse(decrypt(tmp.credential)) : null;

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

@@ -0,0 +1,218 @@
+module.exports = [
+  {
+    name: 'Automatic',
+    value: 'auto',
+  },
+  {
+    name: 'Arabic',
+    value: 'ar',
+  },
+  {
+    name: 'Bashkir',
+    value: 'ba',
+  },
+  {
+    name: 'Basque',
+    value: 'eu',
+  },
+  {
+    name: 'Belarusian',
+    value: 'be',
+  },
+  {
+    name: 'Bulgarian',
+    value: 'bg',
+  },
+  {
+    name: 'Cantonese',
+    value: 'yue',
+  },
+  {
+    name: 'Catalan',
+    value: 'ca',
+  },
+  {
+    name: 'Croatian',
+    value: 'hr',
+  },
+  {
+    name: 'Czech',
+    value: 'cs',
+  },
+  {
+    name: 'Danish',
+    value: 'da',
+  },
+  {
+    name: 'Dutch',
+    value: 'nl',
+  },
+  {
+    name: 'English',
+    value: 'en',
+  },
+  {
+    name: 'Esperanto',
+    value: 'eo',
+  },
+  {
+    name: 'Estonian',
+    value: 'et',
+  },
+  {
+    name: 'Finnish',
+    value: 'fi',
+  },
+  {
+    name: 'French',
+    value: 'fr',
+  },
+  {
+    name: 'Galician',
+    value: 'gl',
+  },
+  {
+    name: 'German',
+    value: 'de',
+  },
+  {
+    name: 'Greek',
+    value: 'el',
+  },
+  {
+    name: 'Hebrew',
+    value: 'he',
+  },
+  {
+    name: 'Hindi',
+    value: 'hi',
+  },
+  {
+    name: 'Hungarian',
+    value: 'hu',
+  },
+  {
+    name: 'Irish',
+    value: 'ga',
+  },
+  {
+    name: 'Interlingua',
+    value: 'ia',
+  },
+  {
+    name: 'Italian',
+    value: 'it',
+  },
+  {
+    name: 'Indonesian',
+    value: 'id',
+  },
+  {
+    name: 'Japanese',
+    value: 'ja',
+  },
+  {
+    name: 'Korean',
+    value: 'ko',
+  },
+  {
+    name: 'Latvian',
+    value: 'lv',
+  },
+  {
+    name: 'Lithuanian',
+    value: 'lt',
+  },
+  {
+    name: 'Maltese',
+    value: 'mt',
+  },
+  {
+    name: 'Malay',
+    value: 'ms',
+  },
+  {
+    name: 'Mandarin',
+    value: 'cmn',
+  },
+  {
+    name: 'Marathi',
+    value: 'mr',
+  },
+  {
+    name: 'Mongolian',
+    value: 'mn',
+  },
+  {
+    name: 'Norwegian',
+    value: 'no',
+  },
+  {
+    name: 'Persian',
+    value: 'fa',
+  },
+  {
+    name: 'Polish',
+    value: 'pl',
+  },
+  {
+    name: 'Portuguese',
+    value: 'pt',
+  },
+  {
+    name: 'Romanian',
+    value: 'ro',
+  },
+  {
+    name: 'Russian',
+    value: 'ru',
+  },
+  {
+    name: 'Slovakian',
+    value: 'sk',
+  },
+  {
+    name: 'Slovenian',
+    value: 'sl',
+  },
+  {
+    name: 'Spanish',
+    value: 'es',
+  },
+  {
+    name: 'Spanish & English bilingual',
+    value: 'es',
+  },
+  {
+    name: 'Swedish',
+    value: 'sv',
+  },
+  {
+    name: 'Tamil',
+    value: 'ta',
+  },
+  {
+    name: 'Thai',
+    value: 'th',
+  },
+  {
+    name: 'Turkish',
+    value: 'tr',
+  },
+  {
+    name: 'Uyghur',
+    value: 'ug',
+  },
+  {
+    name: 'Ukrainian',
+    value: 'uk',
+  },
+  {
+    name: 'Vietnamese',
+    value: 'vi',
+  },
+  {
+    name: 'Welsh',
+    value: 'cy',
+  },
+];

lib/utils/speech-data/tts-languages-playht.js

@@ -0,0 +1,152 @@
+// languages.js
+
+module.exports = [
+  {
+    name: 'English',
+    value: 'english'
+  },
+  {
+    name: 'Mandarin',
+    value: 'mandarin'
+  },
+  {
+    name: 'Hindi',
+    value: 'hindi'
+  },
+  {
+    name: 'Japanese',
+    value: 'japanese'
+  },
+  {
+    name: 'Korean',
+    value: 'korean'
+  },
+  {
+    name: 'Arabic',
+    value: 'arabic'
+  },
+  {
+    name: 'Spanish',
+    value: 'spanish'
+  },
+  {
+    name: 'French',
+    value: 'french'
+  },
+  {
+    name: 'Italian',
+    value: 'italian'
+  },
+  {
+    name: 'Portuguese',
+    value: 'portuguese'
+  },
+  {
+    name: 'German',
+    value: 'german'
+  },
+  {
+    name: 'Dutch',
+    value: 'dutch'
+  },
+  {
+    name: 'Swedish',
+    value: 'swedish'
+  },
+  {
+    name: 'Czech',
+    value: 'czech'
+  },
+  {
+    name: 'Polish',
+    value: 'polish'
+  },
+  {
+    name: 'Russian',
+    value: 'russian'
+  },
+  {
+    name: 'Bulgarian',
+    value: 'bulgarian'
+  },
+  {
+    name: 'Hebrew',
+    value: 'hebrew'
+  },
+  {
+    name: 'Greek',
+    value: 'greek'
+  },
+  {
+    name: 'Turkish',
+    value: 'turkish'
+  },
+  {
+    name: 'Afrikaans',
+    value: 'afrikaans'
+  },
+  {
+    name: 'Xhosa',
+    value: 'xhosa'
+  },
+  {
+    name: 'Tagalog',
+    value: 'tagalog'
+  },
+  {
+    name: 'Malay',
+    value: 'malay'
+  },
+  {
+    name: 'Indonesian',
+    value: 'indonesian'
+  },
+  {
+    name: 'Bengali',
+    value: 'bengali'
+  },
+  {
+    name: 'Serbian',
+    value: 'serbian'
+  },
+  {
+    name: 'Thai',
+    value: 'thai'
+  },
+  {
+    name: 'Urdu',
+    value: 'urdu'
+  },
+  {
+    name: 'Croatian',
+    value: 'croatian'
+  },
+  {
+    name: 'Hungarian',
+    value: 'hungarian'
+  },
+  {
+    name: 'Danish',
+    value: 'danish'
+  },
+  {
+    name: 'Amharic',
+    value: 'amharic'
+  },
+  {
+    name: 'Albanian',
+    value: 'albanian'
+  },
+  {
+    name: 'Catalan',
+    value: 'catalan'
+  },
+  {
+    name: 'Ukrainian',
+    value: 'ukrainian'
+  },
+  {
+    name: 'Galician',
+    value: 'galician'
+  }
+];

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

@@ -1,5 +1,6 @@
 module.exports = [
   { name: 'Turbo v2', value: 'eleven_turbo_v2' },
+  { name: 'Turbo v2.5', value: 'eleven_turbo_v2_5' },
   { name: 'Multilingual v2', value: 'eleven_multilingual_v2' },
   { name: 'Multilingual v1', value: 'eleven_multilingual_v1' },
   { name: 'English v1', value: 'eleven_monolingual_v1' },

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

@@ -1,4 +1,5 @@
 module.exports = [
+  { name: 'Play3.0', value: 'Play3.0' },
   { name: 'PlayHT2.0-turbo', value: 'PlayHT2.0-turbo' },
   { name: 'PlayHT2.0', value: 'PlayHT2.0' },
   { name: 'PlayHT1.0', value: 'PlayHT1.0' },

lib/utils/speech-utils.js

@@ -7,6 +7,7 @@ const bent = require('bent');
 const fs = require('fs');
 const { AssemblyAI } = require('assemblyai');
 const {decrypt, obscureKey} = require('./encrypt-decrypt');
+const { RealtimeSession } = require('speechmatics');
 
 const TtsGoogleLanguagesVoices = require('./speech-data/tts-google');
 const TtsAwsLanguagesVoices = require('./speech-data/tts-aws');
@@ -24,6 +25,7 @@ const TtsModelDeepgram = require('./speech-data/tts-model-deepgram');
 const TtsModelElevenLabs = require('./speech-data/tts-model-elevenlabs');
 const TtsModelWhisper = require('./speech-data/tts-model-whisper');
 const TtsModelPlayHT = require('./speech-data/tts-model-playht');
+const ttsLanguagesPlayHt = require('./speech-data/tts-languages-playht');
 const TtsModelRimelabs = require('./speech-data/tts-model-rimelabs');
 
 const SttGoogleLanguagesVoices = require('./speech-data/stt-google');
@@ -35,9 +37,11 @@ const SttIbmLanguagesVoices = require('./speech-data/stt-ibm');
 const SttNvidiaLanguagesVoices = require('./speech-data/stt-nvidia');
 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 SttVerbioLanguagesVoices = require('./speech-data/stt-verbio');
 
+
 const testSonioxStt = async(logger, credentials) => {
   const api_key = credentials;
   const soniox = new SpeechClient(api_key);
@@ -54,6 +58,61 @@ const testSonioxStt = async(logger, credentials) => {
   });
 };
 
+const testSpeechmaticsStt = async(logger, credentials) => {
+  const {api_key, speechmatics_stt_uri} = credentials;
+  return new Promise(async(resolve, reject) => {
+    try {
+      const session = new RealtimeSession({ apiKey: api_key, realtimeUrl: speechmatics_stt_uri });
+      let transcription = '';
+      session.addListener('Error', (error) => {
+        reject(error);
+      });
+
+      session.addListener('AddTranscript', (message) => {
+        transcription += message.metadata.transcript;
+      });
+
+      session.addListener('EndOfTranscript', () => {
+        resolve(transcription);
+      });
+
+      session
+        .start({
+          transcription_config: {
+            language: 'en',
+            operating_point: 'enhanced',
+            enable_partials: true,
+            max_delay: 2,
+          },
+          audio_format: { type: 'file' },
+        })
+        .then(() => {
+          //prepare file stream
+          const fileStream = fs.createReadStream(`${__dirname}/../../data/test_audio.wav`);
+
+          //send it
+          fileStream.on('data', (sample) => {
+            session.sendAudio(sample);
+          });
+
+          //end the session
+          fileStream.on('end', () => {
+            session.stop();
+          });
+
+          return;
+
+        })
+        .catch((error) => {
+          reject(error);
+        });
+    } catch (error) {
+      logger.info({error}, 'failed to get speechmatics transcript');
+      reject(error);
+    }
+  });
+};
+
 const testNuanceTts = async(logger, getTtsVoices, credentials) => {
   const voices = await getTtsVoices({vendor: 'nuance', credentials});
   return voices;
@@ -92,8 +151,8 @@ const testGoogleStt = async(logger, credentials) => {
 };
 
 const testDeepgramStt = async(logger, credentials) => {
-  const {api_key} = credentials;
-  const deepgram = new Deepgram(api_key);
+  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);
 
   const mimetype = 'audio/wav';
   const source = {
@@ -270,9 +329,10 @@ const testPlayHT = async(logger, synthAudio, credentials) => {
       {
         vendor: 'playht',
         credentials,
-        language: 'en-US',
+        language: 'english',
         voice: 's3://voice-cloning-zero-shot/d9ff78ba-d016-47f6-b0ef-dd630f59414e/female-cs/manifest.json',
-        text: 'Hi there and welcome to jambones!'
+        text: 'Hi there and welcome to jambones!',
+        renderForCaching: true
       }
     );
     // Test if playHT can fetch voices
@@ -295,7 +355,8 @@ const testRimelabs = async(logger, synthAudio, credentials) => {
         credentials,
         language: 'en-US',
         voice: 'amber',
-        text: 'Hi there and welcome to jambones!'
+        text: 'Hi there and welcome to jambones!',
+        renderForCaching: true
       }
     );
   } catch (err) {
@@ -312,7 +373,8 @@ const testWhisper = async(logger, synthAudio, credentials) => {
         credentials,
         language: 'en-US',
         voice: 'alloy',
-        text: 'Hi there and welcome to jambones!'
+        text: 'Hi there and welcome to jambones!',
+        renderForCaching: true
       }
     );
   } catch (err) {
@@ -328,7 +390,8 @@ const testDeepgramTTS = async(logger, synthAudio, credentials) => {
         vendor: 'deepgram',
         credentials,
         model: 'aura-asteria-en',
-        text: 'Hi there and welcome to jambones!'
+        text: 'Hi there and welcome to jambones!',
+        renderForCaching: true
       }
     );
   } catch (err) {
@@ -383,7 +446,8 @@ const testVerbioTts = async(logger, synthAudio, credentials) => {
         credentials,
         language: 'en-US',
         voice: 'tommy_en-us',
-        text: 'Hi there and welcome to jambones!'
+        text: 'Hi there and welcome to jambones!',
+        renderForCaching: true
       }
     );
   } catch (err) {
@@ -509,6 +573,7 @@ function decryptCredential(obj, credential, logger, isObscureKey = true) {
     obj.api_key = isObscureKey ? obscureKey(o.api_key) : o.api_key;
     obj.deepgram_stt_uri = o.deepgram_stt_uri;
     obj.deepgram_stt_use_tls = o.deepgram_stt_use_tls;
+    obj.deepgram_tts_uri = o.deepgram_tts_uri;
   }
   else if ('ibm' === obj.vendor) {
     const o = JSON.parse(decrypt(credential));
@@ -526,6 +591,10 @@ function decryptCredential(obj, credential, logger, isObscureKey = true) {
   } else if ('soniox' === obj.vendor) {
     const o = JSON.parse(decrypt(credential));
     obj.api_key = isObscureKey ? obscureKey(o.api_key) : o.api_key;
+  } else if ('speechmatics' === obj.vendor) {
+    const o = JSON.parse(decrypt(credential));
+    obj.api_key = isObscureKey ? obscureKey(o.api_key) : o.api_key;
+    obj.speechmatics_stt_uri = o.speechmatics_stt_uri;
   } else if ('elevenlabs' === obj.vendor) {
     const o = JSON.parse(decrypt(credential));
     obj.api_key = isObscureKey ? obscureKey(o.api_key) : o.api_key;
@@ -617,6 +686,8 @@ async function getLanguagesAndVoicesForVendor(logger, vendor, credential, getTts
       return await getLanguagesVoicesForWhisper(credential, getTtsVoices, logger);
     case 'verbio':
       return await getLanguagesVoicesForVerbio(credential, getTtsVoices, logger);
+    case 'speechmatics':
+      return await getLanguagesVoicesForSpeechmatics(credential, getTtsVoices, logger);
     default:
       logger.info(`invalid vendor ${vendor}, return empty result`);
       throw new Error(`Invalid vendor ${vendor}`);
@@ -729,6 +800,10 @@ async function getLanguagesVoicesForSoniox(credential) {
   return tranform(undefined, SttSonioxLanguagesVoices);
 }
 
+async function getLanguagesVoicesForSpeechmatics(credential) {
+  return tranform(undefined, SttSpeechmaticsLanguagesVoices);
+}
+
 async function getLanguagesVoicesForElevenlabs(credential) {
   if (credential) {
     const get = bent('https://api.elevenlabs.io', 'GET', 'json', {
@@ -796,6 +871,7 @@ const fetchLayHTVoices = async(credential) => {
 
 async function getLanguagesVoicesForPlayHT(credential) {
   if (credential) {
+    const {voice_engine} = credential;
     const [cloned_voice, voices] = await fetchLayHTVoices(credential);
     const list_voices = [...cloned_voice, ...voices];
 
@@ -803,35 +879,57 @@ async function getLanguagesVoicesForPlayHT(credential) {
       let name = `${d.name} -${concat(d.accent)}${concat(d.age)}${concat(d.gender)}${concat(d.loudness)}` +
       `${concat(d.style)}${concat(d.tempo)}${concat(d.texture)}` ;
       name = name.endsWith(',') ? name.trim().slice(0, -1) : name;
+      name += !d.language_code ? ' - Custom Voice' : '';
+
       return {
         value: `${d.id}`,
         name
       };
     };
 
-    const ttsVoices = list_voices.reduce((acc, voice) => {
-      if (!credential.voice_engine.includes(voice.voice_engine)) {
+    const buildPlay30Payload = () => {
+      // PlayHT3.0 can play different languages with differrent voice.
+      // all voices will be added to english language by default and orther langauges will get voices from english.
+      const ttsVoices = ttsLanguagesPlayHt.map((l) => ({
+        ...l,
+        voices: l.value === 'english' ? list_voices.map((v) => buildVoice(v)) : []
+      }));
+      return tranform(ttsVoices, undefined, TtsModelPlayHT);
+    };
+
+    const buildPayload = () => {
+      const ttsVoices = list_voices.reduce((acc, voice) => {
+        if (!voice_engine.includes(voice.voice_engine)) {
+          return acc;
+        }
+        const languageCode = voice.language_code;
+        // custom voice does not have language code
+        if (!languageCode) {
+          voice.language_code = 'en';
+          voice.language = 'Custom-English';
+        }
+        const existingLanguage = acc.find((lang) => lang.value === languageCode);
+        if (existingLanguage) {
+          existingLanguage.voices.push(buildVoice(voice));
+        } else {
+          acc.push({
+            value: voice.language_code,
+            name: voice.language,
+            voices: [buildVoice(voice)]
+          });
+        }
         return acc;
-      }
-      const languageCode = voice.language_code;
-      // custom voice does not have language code
-      if (!languageCode) {
-        voice.language_code = 'en';
-        voice.language = 'Custom-English';
-      }
-      const existingLanguage = acc.find((lang) => lang.value === languageCode);
-      if (existingLanguage) {
-        existingLanguage.voices.push(buildVoice(voice));
-      } else {
-        acc.push({
-          value: voice.language_code,
-          name: voice.language,
-          voices: [buildVoice(voice)]
-        });
-      }
-      return acc;
-    }, []);
-    return tranform(ttsVoices, undefined, TtsModelPlayHT);
+      }, []);
+      return tranform(ttsVoices, undefined, TtsModelPlayHT);
+    };
+
+    switch (voice_engine) {
+      case 'Play3.0':
+        return buildPlay30Payload();
+
+      default:
+        return buildPayload();
+    }
   }
   return tranform(TtsPlayHtLanguagesVoices, undefined, TtsModelPlayHT);
 }
@@ -1057,5 +1155,6 @@ module.exports = {
   testWhisper,
   testVerbioTts,
   testVerbioStt,
-  getLanguagesAndVoicesForVendor
+  getLanguagesAndVoicesForVendor,
+  testSpeechmaticsStt
 };

package.json

@@ -1,6 +1,6 @@
 {
   "name": "jambonz-api-server",
-  "version": "0.9.0",
+  "version": "0.9.2",
   "description": "",
   "main": "app.js",
   "scripts": {
@@ -28,8 +28,8 @@
     "@jambonz/db-helpers": "^0.9.3",
     "@jambonz/lamejs": "^1.2.2",
     "@jambonz/mw-registrar": "^0.2.7",
-    "@jambonz/realtimedb-helpers": "^0.8.9",
-    "@jambonz/speech-utils": "^0.1.11",
+    "@jambonz/realtimedb-helpers": "^0.8.10",
+    "@jambonz/speech-utils": "^0.1.20",
     "@jambonz/time-series": "^0.2.8",
     "@jambonz/verb-specifications": "^0.0.72",
     "@soniox/soniox-node": "^1.2.2",
@@ -46,12 +46,13 @@
     "jsonwebtoken": "^9.0.2",
     "mailgun.js": "^10.2.1",
     "microsoft-cognitiveservices-speech-sdk": "1.36.0",
-    "mysql2": "^3.9.3",
+    "mysql2": "^3.11.0",
     "nocache": "4.0.0",
     "passport": "^0.7.0",
     "passport-http-bearer": "^1.0.1",
     "pino": "^8.20.0",
     "short-uuid": "^4.2.2",
+    "speechmatics": "^4.0.0",
     "stream-buffers": "^3.0.2",
     "stripe": "^14.24.0",
     "swagger-ui-express": "^5.0.0",

test/accounts.js

@@ -152,7 +152,9 @@ test('account tests', async(t) => {
       auth: authAdmin,
       json: true,
     });
+    console.log(result);
     t.ok(result.name === 'daveh' , 'successfully retrieved account by sid');
+    t.ok(result.enable_debug_log  === 0 , 'enable_debug_log default value ok');
 
     /* update account with account level token */
     result = await request.put(`/Accounts/${sid}`, {
@@ -177,8 +179,8 @@ test('account tests', async(t) => {
           name: 'recordings',
           access_key_id: 'access_key_id',
           secret_access_key: 'secret access key'
-        }
-        
+        },
+        enable_debug_log: true
       }
     });
     t.ok(result.statusCode === 204, 'successfully updated account using account level token');
@@ -194,6 +196,7 @@ test('account tests', async(t) => {
     t.ok(result.bucket_credential.access_key_id === 'access_key_id', 'bucket_access_key_id was updated');
     t.ok(result.record_all_calls === 1, 'record_all_calls was updated');
     t.ok(result.record_format === 'wav', 'record_format was updated');
+    t.ok(result.enable_debug_log, 'enable_debug_log was updated');
 
     /* verify that account level api key last_used was updated*/
     result = await request.get(`/Accounts/${sid}/ApiKeys`, {

test/sip-gateways.js

@@ -17,6 +17,53 @@ test('sip gateway tests', async(t) => {
     let result;
     const voip_carrier_sid = await createVoipCarrier(request);
 
+    /* add a invalid sip gateway */
+    const STORED_JAMBONZ_MIN_GATEWAY_NETMASK = process.env.JAMBONZ_MIN_GATEWAY_NETMASK;
+    process.env.JAMBONZ_MIN_GATEWAY_NETMASK = 24;
+    
+    result = await request.post('/SipGateways', {
+      resolveWithFullResponse: true,
+      auth: authAdmin,
+      json: true,
+      simple: false,
+      body: {
+        voip_carrier_sid,
+        ipv4: '1.2.3.4',
+        netmask: 1,
+        inbound: true,
+        outbound: true,
+        protocol: 'tcp'
+      }
+    });
+    t.ok(result.statusCode === 400, 'successfully created sip gateway ');
+
+    result = await request.post('/SipGateways', {
+      resolveWithFullResponse: true,
+      auth: authAdmin,
+      json: true,
+      body: {
+        voip_carrier_sid,
+        ipv4: '1.2.3.4',
+        netmask: 24,
+        inbound: true,
+        outbound: true,
+        protocol: 'tcp'
+      }
+    });
+    t.ok(result.statusCode === 201, 'successfully created sip gateway ');
+
+    process.env.JAMBONZ_MIN_GATEWAY_NETMASK = STORED_JAMBONZ_MIN_GATEWAY_NETMASK;
+
+    /* delete sip gateways */
+    result = await request.delete(`/SipGateways/${result.body.sid}`, {
+      resolveWithFullResponse: true,
+      simple: false,
+      json: true,
+      auth: authAdmin
+    });
+    //console.log(`result: ${JSON.stringify(result)}`);
+    t.ok(result.statusCode === 204, 'successfully deleted sip gateway');
+
     /* add a sip gateway */
     result = await request.post('/SipGateways', {
       resolveWithFullResponse: true,

test/speech-credentials.js

@@ -371,7 +371,8 @@ test('speech credentials tests', async(t) => {
         vendor: 'deepgram',
         use_for_stt: true,
         deepgram_stt_uri: "127.0.0.1:50002",
-        deepgram_stt_use_tls: true
+        deepgram_stt_use_tls: true,
+        deepgram_tts_uri: 'https://server.com'
       }
     });
     t.ok(result.statusCode === 201, 'successfully added speech credential for deepgram');
@@ -386,6 +387,7 @@ test('speech credentials tests', async(t) => {
     t.ok(result.statusCode === 200, 'successfully get speech credential for deepgram');
     t.ok(result.body.deepgram_stt_uri === '127.0.0.1:50002', "deepgram_stt_uri is correct for deepgram");
     t.ok(result.body.deepgram_stt_use_tls === true, "deepgram_stt_use_tls is correct for deepgram");
+    t.ok(result.body.deepgram_tts_uri === 'https://server.com', "deepgram_tts_uri is correct for deepgram")
 
     result = await request.put(`/Accounts/${account_sid}/SpeechCredentials/${dg_sid}`, {
       resolveWithFullResponse: true,
@@ -395,7 +397,8 @@ test('speech credentials tests', async(t) => {
         vendor: 'deepgram',
         use_for_stt: true,
         deepgram_stt_uri: "127.0.0.2:50002",
-        deepgram_stt_use_tls: false
+        deepgram_stt_use_tls: false,
+        deepgram_tts_uri: 'https://server2.com'
       }
     });
     t.ok(result.statusCode === 204, 'successfully updated speech credential for deepgram onprem');
@@ -409,6 +412,7 @@ test('speech credentials tests', async(t) => {
     t.ok(result.statusCode === 200, 'successfully get speech credential for deepgram onprem');
     t.ok(result.body.deepgram_stt_uri === '127.0.0.2:50002', "deepgram_stt_uri is correct for deepgram onprem");
     t.ok(result.body.deepgram_stt_use_tls === false, "deepgram_stt_use_tls is correct for deepgram onprem");
+    t.ok(result.body.deepgram_tts_uri === 'https://server2.com', "deepgram_tts_uri is correct for deepgram onprem");
 
     result = await request.delete(`/Accounts/${account_sid}/SpeechCredentials/${dg_sid}`, {
       auth: authUser,
@@ -528,6 +532,39 @@ test('speech credentials tests', async(t) => {
       t.ok(result.statusCode === 204, 'successfully deleted speech credential');
     }
 
+    /* add a credential for Speechmatics */
+    if (process.env.SPEECHMATICS_API_KEY) {
+      result = await request.post(`/Accounts/${account_sid}/SpeechCredentials`, {
+        resolveWithFullResponse: true,
+        auth: authUser,
+        json: true,
+        body: {
+          vendor: 'speechmatics',
+          use_for_stt: true,
+          api_key: process.env.SPEECHMATICS_API_KEY,
+          speechmatics_stt_uri: 'eu2.rt.speechmatics.com'
+        }
+      });
+      t.ok(result.statusCode === 201, 'successfully added speech credential for speechmatics');
+      const ms_sid = result.body.sid;
+
+      /* test the speech credential */
+      result = await request.get(`/Accounts/${account_sid}/SpeechCredentials/${ms_sid}/test`, {
+        resolveWithFullResponse: true,
+        auth: authUser,
+        json: true,   
+      });
+      console.log(JSON.stringify(result));
+      t.ok(result.statusCode === 200 && result.body.stt.status === 'ok', 'successfully tested speech credential for speechmatics');
+
+      /* delete the credential */
+      result = await request.delete(`/Accounts/${account_sid}/SpeechCredentials/${ms_sid}`, {
+        auth: authUser,
+        resolveWithFullResponse: true,
+      });
+      t.ok(result.statusCode === 204, 'successfully deleted speech credential');
+    }
+
     /* add a credential for nvidia */
     result = await request.post(`/Accounts/${account_sid}/SpeechCredentials`, {
       resolveWithFullResponse: true,

test/system-information.js

@@ -16,7 +16,9 @@ test('system information test', async(t) => {
       body: {
         domain_name: 'test.com',
         sip_domain_name: 'sip.test.com',
-        monitoring_domain_name: 'monitor.test.com'
+        monitoring_domain_name: 'monitor.test.com',
+        private_network_cidr: '192.168.1.0/24, 10.10.100.1',
+        log_level: 'info'
       }
     });
     t.ok(result.statusCode === 201, 'successfully created system information ');
@@ -24,6 +26,8 @@ test('system information test', async(t) => {
     t.ok(body.domain_name === 'test.com', 'added domain_name ok');
     t.ok(body.sip_domain_name === 'sip.test.com', 'added sip_domain_name ok');
     t.ok(body.monitoring_domain_name === 'monitor.test.com', 'added monitoring_domain_name ok');
+    t.ok(body.private_network_cidr === '192.168.1.0/24, 10.10.100.1', 'added private_network_cidr ok');
+    t.ok(body.log_level === 'info', 'added log_level ok');
 
     result = await request.get('/SystemInformation', {
       auth: authAdmin,
@@ -32,6 +36,8 @@ test('system information test', async(t) => {
     t.ok(result.domain_name === 'test.com', 'get domain_name ok');
     t.ok(result.sip_domain_name === 'sip.test.com', 'get sip_domain_name ok');
     t.ok(result.monitoring_domain_name === 'monitor.test.com', 'get monitoring_domain_name ok');
+    t.ok(result.private_network_cidr === '192.168.1.0/24, 10.10.100.1', 'get private_network_cidr ok'); 
+    t.ok(result.log_level === 'info', 'added log_level ok');
 
     result = await request.post('/SystemInformation', {
       resolveWithFullResponse: true,
@@ -40,7 +46,9 @@ test('system information test', async(t) => {
       body: {
         domain_name: 'test1.com',
         sip_domain_name: 'sip1.test.com',
-        monitoring_domain_name: 'monitor1.test.com'
+        monitoring_domain_name: 'monitor1.test.com',
+        private_network_cidr: '',
+        log_level: 'debug'
       }
     });
     t.ok(result.statusCode === 201, 'successfully updated system information ');
@@ -48,6 +56,8 @@ test('system information test', async(t) => {
     t.ok(body.domain_name === 'test1.com', 'updated domain_name ok');
     t.ok(body.sip_domain_name === 'sip1.test.com', 'updated sip_domain_name ok');
     t.ok(body.monitoring_domain_name === 'monitor1.test.com', 'updated monitoring_domain_name ok');
+    t.ok(body.private_network_cidr === '', 'updated private_network_cidr ok');
+    t.ok(body.log_level === 'debug', 'updated log_level ok');
 
     result = await request.get('/SystemInformation', {
       auth: authAdmin,
@@ -56,6 +66,7 @@ test('system information test', async(t) => {
     t.ok(result.domain_name === 'test1.com', 'get domain_name ok');
     t.ok(result.sip_domain_name === 'sip1.test.com', 'get sip_domain_name ok');
     t.ok(result.monitoring_domain_name === 'monitor1.test.com', 'get monitoring_domain_name ok');
+    t.ok(result.log_level === 'debug', 'updated log_level ok');
 
   } catch(err) {
     console.error(err);