0.1.2

Improve handling of TTS cache by adding the file extension to the cac…

.github/workflows/ci.yml

@@ -8,8 +8,8 @@ jobs:
   build:
     runs-on: ubuntu-latest
     steps:
-      - uses: actions/checkout@v3
-      - uses: actions/setup-node@v3
+      - uses: actions/checkout@v4
+      - uses: actions/setup-node@v4
         with:
           node-version: lts/*
       - run: npm install
@@ -32,5 +32,7 @@ jobs:
           ELEVENLABS_API_KEY: ${{ secrets.ELEVENLABS_API_KEY }}
           ELEVENLABS_VOICE_ID: ${{ secrets.ELEVENLABS_VOICE_ID }}
           ELEVENLABS_MODEL_ID: ${{ secrets.ELEVENLABS_MODEL_ID }}
+          PLAYHT_USER_ID: ${{ secrets.PLAYHT_USER_ID }}
+          PLAYHT_API_KEY: ${{ secrets.PLAYHT_API_KEY }}
           JAMBONES_HTTP_PROXY_IP: 127.0.0.1
           JAMBONES_HTTP_PROXY_PORT: 3128

.gitignore

@@ -39,3 +39,5 @@ node_modules
 examples/*
 
 .vscode
+
+.env

index.js

@@ -13,10 +13,11 @@ module.exports = (opts, logger) => {
     getTtsSize: require('./lib/get-tts-size').bind(null, client, logger),
     purgeTtsCache: require('./lib/purge-tts-cache').bind(null, client, logger),
     addFileToCache: require('./lib/add-file-to-cache').bind(null, client, logger),
-    synthAudio: require('./lib/synth-audio').bind(null, client, logger),
+    synthAudio: require('./lib/synth-audio').bind(null, client, createHash, retrieveHash, logger),
+    getVerbioAccessToken: require('./lib/get-verbio-token').bind(null, client, logger),
     getNuanceAccessToken: require('./lib/get-nuance-access-token').bind(null, client, logger),
     getIbmAccessToken: require('./lib/get-ibm-access-token').bind(null, client, logger),
     getAwsAuthToken: require('./lib/get-aws-sts-token').bind(null, logger, createHash, retrieveHash),
-    getTtsVoices: require('./lib/get-tts-voices').bind(null, client, logger),
+    getTtsVoices: require('./lib/get-tts-voices').bind(null, client, createHash, retrieveHash, logger),
   };
 };

lib/add-file-to-cache.js

@@ -1,6 +1,7 @@
 const fs = require('fs/promises');
 const {noopLogger, makeSynthKey} = require('./utils');
-const EXPIRES = (process.env.JAMBONES_TTS_CACHE_DURATION_MINS || 4 * 60) * 60; // cache tts for 4 hours
+const {JAMBONES_TTS_CACHE_DURATION_MINS} = require('./config');
+const EXPIRES = JAMBONES_TTS_CACHE_DURATION_MINS;
 
 async function addFileToCache(client, logger, path,
   {account_sid, vendor, language, voice, deploymentId, engine, text}) {

lib/config.js

@@ -0,0 +1,20 @@
+const JAMBONES_TTS_TRIM_SILENCE = process.env.JAMBONES_TTS_TRIM_SILENCE;
+const JAMBONES_DISABLE_TTS_STREAMING = process.env.JAMBONES_DISABLE_TTS_STREAMING;
+
+const JAMBONES_HTTP_PROXY_IP = process.env.JAMBONES_HTTP_PROXY_IP;
+const JAMBONES_HTTP_PROXY_PORT = process.env.JAMBONES_HTTP_PROXY_PORT;
+const JAMBONES_TTS_CACHE_DURATION_MINS = (parseInt(process.env.JAMBONES_TTS_CACHE_DURATION_MINS) || 4 * 60) * 60; // cache tts for 4 hours
+
+const TMP_FOLDER = '/tmp';
+
+const HTTP_TIMEOUT = 5000;
+
+module.exports = {
+  JAMBONES_TTS_TRIM_SILENCE,
+  JAMBONES_DISABLE_TTS_STREAMING,
+  JAMBONES_HTTP_PROXY_IP,
+  JAMBONES_HTTP_PROXY_PORT,
+  JAMBONES_TTS_CACHE_DURATION_MINS,
+  TMP_FOLDER,
+  HTTP_TIMEOUT
+};

lib/get-aws-sts-token.js

@@ -1,32 +1,41 @@
-const { STSClient, GetSessionTokenCommand } = require('@aws-sdk/client-sts');
+const { STSClient, GetSessionTokenCommand, AssumeRoleCommand } = require('@aws-sdk/client-sts');
 const {makeAwsKey, noopLogger} = require('./utils');
 const debug = require('debug')('jambonz:speech-utils');
 const EXPIRY = 3600;
 
 async function getAwsAuthToken(
-  logger,
-  createHash, retrieveHash,
-  awsAccessKeyId, awsSecretAccessKey, awsRegion) {
+  logger, createHash, retrieveHash,
+  awsAccessKeyId, awsSecretAccessKey, awsRegion, roleArn = null) {
   logger = logger || noopLogger;
   try {
-    const key = makeAwsKey(awsAccessKeyId);
+    const key = makeAwsKey(roleArn || awsAccessKeyId);
     const obj = await retrieveHash(key);
     if (obj) return {...obj, servedFromCache: true};
 
-    /* access token not found in cache, so generate it using STS */
-    const stsClient = new STSClient({
-      region: awsRegion,
-      credentials: {
-        accessKeyId: awsAccessKeyId,
-        secretAccessKey: awsSecretAccessKey,
-      }
-    });
-    const command = new GetSessionTokenCommand({DurationSeconds: EXPIRY});
-    const data = await stsClient.send(command);
+    let data;
+    if (roleArn) {
+      const stsClient = new STSClient({ region: awsRegion});
+      const roleToAssume = { RoleArn: roleArn, RoleSessionName: 'Jambonz_Speech', DurationSeconds: EXPIRY};
+      const command = new AssumeRoleCommand(roleToAssume);
+
+      data = await stsClient.send(command);
+    } else {
+      /* access token not found in cache, so generate it using STS */
+      const stsClient = new STSClient({
+        region: awsRegion,
+        credentials: {
+          accessKeyId: awsAccessKeyId,
+          secretAccessKey: awsSecretAccessKey,
+        }
+      });
+      const command = new GetSessionTokenCommand({DurationSeconds: EXPIRY});
+      data = await stsClient.send(command);
+    }
 
     const credentials = {
       accessKeyId: data.Credentials.AccessKeyId,
       secretAccessKey: data.Credentials.SecretAccessKey,
+      sessionToken: data.Credentials.SessionToken,
       securityToken: data.Credentials.SessionToken
     };
 

lib/get-ibm-access-token.js

@@ -2,8 +2,8 @@ const formurlencoded = require('form-urlencoded');
 const {Pool} = require('undici');
 const pool = new Pool('https://iam.cloud.ibm.com');
 const {makeIbmKey, noopLogger} = require('./utils');
+const { HTTP_TIMEOUT } = require('./config');
 const debug = require('debug')('jambonz:realtimedb-helpers');
-const HTTP_TIMEOUT = 5000;
 
 async function getIbmAccessToken(client, logger, apiKey) {
   logger = logger || noopLogger;

lib/get-nuance-access-token.js

@@ -2,8 +2,8 @@ const formurlencoded = require('form-urlencoded');
 const {Pool} = require('undici');
 const pool = new Pool('https://auth.crt.nuance.com');
 const {makeNuanceKey, makeBasicAuthHeader, noopLogger} = require('./utils');
+const { HTTP_TIMEOUT } = require('./config');
 const debug = require('debug')('jambonz:realtimedb-helpers');
-const HTTP_TIMEOUT = 5000;
 
 async function getNuanceAccessToken(client, logger, clientId, secret, scope) {
   logger = logger || noopLogger;

lib/get-tts-voices.js

@@ -1,11 +1,16 @@
 const assert = require('assert');
 const {noopLogger, createNuanceClient, createKryptonClient} = require('./utils');
 const getNuanceAccessToken = require('./get-nuance-access-token');
+const getVerbioAccessToken = require('./get-verbio-token');
 const {GetVoicesRequest, Voice} = require('../stubs/nuance/synthesizer_pb');
 const TextToSpeechV1 = require('ibm-watson/text-to-speech/v1');
 const { IamAuthenticator } = require('ibm-watson/auth');
 const ttsGoogle = require('@google-cloud/text-to-speech');
 const { PollyClient, DescribeVoicesCommand } = require('@aws-sdk/client-polly');
+const getAwsAuthToken = require('./get-aws-sts-token');
+const {Pool} = require('undici');
+const { HTTP_TIMEOUT } = require('./config');
+const verbioVoicePool = new Pool('https://us.rest.speechcenter.verbio.com');
 
 const getIbmVoices = async(client, logger, credentials) => {
   const {tts_region, tts_api_key} = credentials;
@@ -87,16 +92,26 @@ const getGoogleVoices = async(_client, logger, credentials) => {
   return await client.listVoices();
 };
 
-const getAwsVoices = async(_client, logger, credentials) => {
+const getAwsVoices = async(_client, createHash, retrieveHash, logger, credentials) => {
   try {
-    const {region, accessKeyId, secretAccessKey} = credentials;
-    const client = new PollyClient({
-      region,
-      credentials: {
-        accessKeyId,
-        secretAccessKey
-      }
-    });
+    const {region, accessKeyId, secretAccessKey, roleArn} = credentials;
+    let client = null;
+    if (accessKeyId && secretAccessKey) {
+      client = new PollyClient({
+        region,
+        credentials: {
+          accessKeyId,
+          secretAccessKey
+        }
+      });
+    } else if (roleArn) {
+      client = new PollyClient({
+        region,
+        credentials: await getAwsAuthToken(logger, createHash, retrieveHash, null, null, region, roleArn),
+      });
+    } else {
+      client = new PollyClient({region});
+    }
     const command = new DescribeVoicesCommand({});
     const response = await client.send(command);
     return response;
@@ -106,6 +121,26 @@ const getAwsVoices = async(_client, logger, credentials) => {
   }
 };
 
+const getVerbioVoices = async(client, logger, credentials) => {
+  try {
+    const access_token = await getVerbioAccessToken(client, logger, credentials);
+    const { body} =  await verbioVoicePool.request({
+      path: '/api/v1/voices',
+      method: 'GET',
+      headers: {
+        'Authorization': `Bearer ${access_token.access_token}`,
+        'User-Agent': 'jambonz'
+      },
+      timeout: HTTP_TIMEOUT,
+      followRedirects: false
+    });
+    return await body.json();
+  } catch (err) {
+    logger.info({err}, 'getVerbioVoices - failed to list voices for Verbio');
+    throw err;
+  }
+};
+
 /**
  * Synthesize speech to an mp3 file, and also cache the generated speech
  * in redis (base64 format) for 24 hours so as to avoid unnecessarily paying
@@ -122,10 +157,10 @@ const getAwsVoices = async(_client, logger, credentials) => {
  * @returns object containing filepath to an mp3 file in the /tmp folder containing
  * the synthesized audio, and a variable indicating whether it was served from cache
  */
-async function getTtsVoices(client, logger, {vendor, credentials}) {
+async function getTtsVoices(client, createHash, retrieveHash, logger, {vendor, credentials}) {
   logger = logger || noopLogger;
 
-  assert.ok(['nuance', 'ibm', 'google', 'aws', 'polly'].includes(vendor),
+  assert.ok(['nuance', 'ibm', 'google', 'aws', 'polly', 'verbio'].includes(vendor),
     `getTtsVoices not supported for vendor ${vendor}`);
 
   switch (vendor) {
@@ -137,7 +172,9 @@ async function getTtsVoices(client, logger, {vendor, credentials}) {
       return getGoogleVoices(client, logger, credentials);
     case 'aws':
     case 'polly':
-      return getAwsVoices(client, logger, credentials);
+      return getAwsVoices(client, createHash, retrieveHash, logger, credentials);
+    case 'verbio':
+      return getVerbioVoices(client, logger, credentials);
     default:
       break;
   }

lib/get-verbio-token.js

@@ -0,0 +1,51 @@
+const {Pool} = require('undici');
+const { noopLogger, makeVerbioKey } = require('./utils');
+const { HTTP_TIMEOUT } = require('./config');
+const pool = new Pool('https://auth.speechcenter.verbio.com:444');
+const debug = require('debug')('jambonz:realtimedb-helpers');
+
+async function getVerbioAccessToken(client, logger, credentials) {
+  logger = logger || noopLogger;
+  const { client_id, client_secret } = credentials;
+  try {
+    const key = makeVerbioKey(client_id);
+    const access_token = await client.get(key);
+    if (access_token) {
+      return {access_token, servedFromCache: true};
+    }
+
+    const payload = {
+      client_id,
+      client_secret
+    };
+
+    const {statusCode, headers, body} =  await pool.request({
+      path: '/api/v1/token',
+      method: 'POST',
+      headers: {
+        'Content-Type': 'application/json',
+        'User-Agent': 'jambonz'
+      },
+      body: JSON.stringify(payload),
+      timeout: HTTP_TIMEOUT,
+      followRedirects: false
+    });
+
+    if (200 !== statusCode) {
+      logger.debug({statusCode, headers, body: await body.text()}, 'error fetching access token from Verbio');
+      const err = new Error();
+      err.statusCode = statusCode;
+      throw err;
+    }
+    const json = await body.json();
+    const expiry =  Math.floor(json.expiration_time - Date.now() / 1000 - 30);
+    await client.set(key, json.access_token, 'EX', expiry);
+    return {...json, servedFromCache: false};
+  } catch (err) {
+    debug(err, `getVerbioAccessToken: Error retrieving Verbio access token for client_id ${client_id}`);
+    logger.error(err, `getVerbioAccessToken: Error retrieving Verbio access token for client_id ${client_id}`);
+    throw err;
+  }
+}
+
+module.exports = getVerbioAccessToken;

lib/synth-audio.js

@@ -19,9 +19,11 @@ const {
   createNuanceClient,
   createKryptonClient,
   createRivaClient,
-  noopLogger
+  noopLogger,
+  makeFilePath
 } = require('./utils');
 const getNuanceAccessToken = require('./get-nuance-access-token');
+const getVerbioAccessToken = require('./get-verbio-token');
 const {
   SynthesisRequest,
   Voice,
@@ -36,9 +38,15 @@ const {
 const {SynthesizeSpeechRequest} = require('../stubs/riva/proto/riva_tts_pb');
 const {AudioEncoding} = require('../stubs/riva/proto/riva_audio_pb');
 const debug = require('debug')('jambonz:realtimedb-helpers');
-const EXPIRES = (process.env.JAMBONES_TTS_CACHE_DURATION_MINS || 4 * 60) * 60; // cache tts for 4 hours
-const TMP_FOLDER = '/tmp';
+const {
+  JAMBONES_DISABLE_TTS_STREAMING,
+  JAMBONES_HTTP_PROXY_IP,
+  JAMBONES_HTTP_PROXY_PORT,
+  JAMBONES_TTS_CACHE_DURATION_MINS,
+} = require('./config');
+const EXPIRES = JAMBONES_TTS_CACHE_DURATION_MINS;
 const OpenAI = require('openai');
+const getAwsAuthToken = require('./get-aws-sts-token');
 
 
 const trimTrailingSilence = (buffer) => {
@@ -75,7 +83,7 @@ const trimTrailingSilence = (buffer) => {
  * @returns object containing filepath to an mp3 file in the /tmp folder containing
  * the synthesized audio, and a variable indicating whether it was served from cache
  */
-async function synthAudio(client, logger, stats, { account_sid,
+async function synthAudio(client, createHash, retrieveHash, logger, stats, { account_sid,
   vendor, language, voice, gender, text, engine, salt, model, credentials, deploymentId,
   disableTtsCache, renderForCaching, disableTtsStreaming, options
 }) {
@@ -84,10 +92,10 @@ async function synthAudio(client, logger, stats, { account_sid,
   let rtt;
   logger = logger || noopLogger;
 
-  assert.ok(['google', 'aws', 'polly', 'microsoft',
-    'wellsaid', 'nuance', 'nvidia', 'ibm', 'elevenlabs', 'whisper', 'deepgram'].includes(vendor) ||
+  assert.ok(['google', 'aws', 'polly', 'microsoft', 'wellsaid', 'nuance', 'nvidia', 'ibm', 'elevenlabs',
+    'whisper', 'deepgram', 'playht', 'rimelabs', 'verbio'].includes(vendor) ||
   vendor.startsWith('custom'),
-  `synthAudio supported vendors are google, aws, microsoft, nuance, nvidia and wellsaid, not ${vendor}`);
+  `synthAudio supported vendors are google, aws, microsoft, nuance, nvidia and wellsaid ..etc, not ${vendor}`);
   if ('google' === vendor) {
     assert.ok(language, 'synthAudio requires language when google is used');
   }
@@ -123,12 +131,25 @@ async function synthAudio(client, logger, stats, { account_sid,
     assert.ok(voice, 'synthAudio requires voice when elevenlabs is used');
     assert.ok(credentials.api_key, 'synthAudio requires api_key when elevenlabs is used');
     assert.ok(credentials.model_id, 'synthAudio requires model_id when elevenlabs is used');
+  } else if ('playht' === vendor) {
+    assert.ok(voice, 'synthAudio requires voice when playht is used');
+    assert.ok(credentials.api_key, 'synthAudio requires api_key when playht is used');
+    assert.ok(credentials.user_id, 'synthAudio requires user_id when playht is used');
+    assert.ok(credentials.voice_engine, 'synthAudio requires voice_engine when playht is used');
+  } else if ('rimelabs' === vendor) {
+    assert.ok(voice, 'synthAudio requires voice when rimelabs is used');
+    assert.ok(credentials.api_key, 'synthAudio requires api_key when rimelabs is used');
+    assert.ok(credentials.model_id, 'synthAudio requires model_id when rimelabs is used');
   } else if ('whisper' === vendor) {
     assert.ok(voice, 'synthAudio requires voice when whisper is used');
     assert.ok(credentials.model_id, 'synthAudio requires model when whisper is used');
     assert.ok(credentials.api_key, 'synthAudio requires api_key when whisper is used');
   } else  if (vendor.startsWith('custom')) {
     assert.ok(credentials.custom_tts_url, `synthAudio requires custom_tts_url in credentials when ${vendor} is used`);
+  } else if ('verbio' === vendor) {
+    assert.ok(voice, 'synthAudio requires voice when verbio is used');
+    assert.ok(credentials.client_id, 'synthAudio requires client_id when verbio is used');
+    assert.ok(credentials.client_secret, 'synthAudio requires client_secret when verbio is used');
   }
   const key = makeSynthKey({
     account_sid,
@@ -139,19 +160,7 @@ async function synthAudio(client, logger, stats, { account_sid,
     text
   });
   let filePath;
-  if (['nuance', 'nvidia'].includes(vendor) ||
-    (
-      process.env.JAMBONES_TTS_TRIM_SILENCE &&
-      ['microsoft', 'azure'].includes(vendor)
-    ) ||
-    (
-      !process.env.JAMBONES_DISABLE_TTS_STREAMING &&
-      vendor === 'elevenlabs'
-    )
-  ) {
-    filePath = `${TMP_FOLDER}/${key.replace('tts:', `tts-${salt || ''}`)}.r8`;
-  }
-  else filePath = `${TMP_FOLDER}/${key.replace('tts:', `tts-${salt || ''}`)}.mp3`;
+  filePath = makeFilePath(vendor, key, salt);
   debug(`synth key is ${key}`);
   let cached;
   if (!disableTtsCache) {
@@ -178,12 +187,14 @@ async function synthAudio(client, logger, stats, { account_sid,
       case 'aws':
       case 'polly':
         vendorLabel = 'aws';
-        audioBuffer = await synthPolly(logger, {credentials, stats, language, voice, text, engine});
+        audioBuffer = await synthPolly(createHash, retrieveHash, logger,
+          {credentials, stats, language, voice, text, engine});
         break;
       case 'azure':
       case 'microsoft':
         vendorLabel = 'microsoft';
-        audioBuffer = await synthMicrosoft(logger, {credentials, stats, language, voice, text, deploymentId, filePath});
+        audioBuffer = await synthMicrosoft(logger, {credentials, stats, language, voice, text, deploymentId,
+          filePath, renderForCaching, disableTtsStreaming});        
         break;
       case 'nuance':
         model = model || 'enhanced';
@@ -201,16 +212,30 @@ async function synthAudio(client, logger, stats, { account_sid,
       case 'elevenlabs':
         audioBuffer = await synthElevenlabs(logger, {
           credentials, options, stats, language, voice, text, renderForCaching, disableTtsStreaming, filePath
-        });
-        if (audioBuffer?.filePath) return audioBuffer;
+        });        
+        break;
+      case 'playht':
+        audioBuffer = await synthPlayHT(logger, {
+          credentials, options, stats, language, voice, text, renderForCaching, disableTtsStreaming, filePath
+        });        
+        break;
+      case 'rimelabs':
+        audioBuffer = await synthRimelabs(logger, {
+          credentials, options, stats, language, voice, text, renderForCaching, disableTtsStreaming, filePath
+        });        
         break;
       case 'whisper':
         audioBuffer = await synthWhisper(logger, {
+          credentials, stats, voice, text, renderForCaching, disableTtsStreaming});        
+        break;
+      case 'verbio':
+        audioBuffer = await synthVerbio(client, logger, {
           credentials, stats, voice, text, renderForCaching, disableTtsStreaming});
         if (audioBuffer?.filePath) return audioBuffer;
         break;
       case 'deepgram':
-        audioBuffer = await synthDeepgram(logger, {credentials, stats, model, text});
+        audioBuffer = await synthDeepgram(logger, {credentials, stats, model, text,
+          renderForCaching, disableTtsStreaming});        
         break;
       case vendor.startsWith('custom') ? vendor : 'cant_match_value':
         ({ audioBuffer, filePath } = await synthCustomVendor(logger,
@@ -219,6 +244,7 @@ async function synthAudio(client, logger, stats, { account_sid,
       default:
         assert(`synthAudio: unsupported speech vendor ${vendor}`);
     }
+    if('filePath' in audioBuffer) return audioBuffer;
     const diff = process.hrtime(startAt);
     const time = diff[0] * 1e3 + diff[1] * 1e-6;
     rtt = time.toFixed(0);
@@ -238,16 +264,28 @@ async function synthAudio(client, logger, stats, { account_sid,
   });
 }
 
-const synthPolly = async(logger, {credentials, stats, language, voice, engine, text}) => {
+const synthPolly = async(createHash, retrieveHash, logger,
+  {credentials, stats, language, voice, engine, text}) => {
   try {
-    const {region, accessKeyId, secretAccessKey} = credentials;
-    const polly = new PollyClient({
-      region,
-      credentials: {
-        accessKeyId,
-        secretAccessKey
-      }
-    });
+    const {region, accessKeyId, secretAccessKey, roleArn} = credentials;
+    let polly;
+    if (accessKeyId && secretAccessKey) {
+      polly = new PollyClient({
+        region,
+        credentials: {
+          accessKeyId,
+          secretAccessKey
+        }
+      });
+    } else if (roleArn) {
+      polly = new PollyClient({
+        region,
+        credentials: await getAwsAuthToken(logger, createHash, retrieveHash, null, null, region, roleArn),
+      });
+    } else {
+      // AWS RoleArn assigned to Instance profile
+      polly = new PollyClient({region});
+    }
     const opts = {
       Engine: engine,
       OutputFormat: 'mp3',
@@ -341,7 +379,7 @@ async function _synthOnPremMicrosoft(logger, {
   text,
   filePath
 }) {
-  const {use_custom_tts, custom_tts_endpoint_url} = credentials;
+  const {use_custom_tts, custom_tts_endpoint_url, api_key} = credentials;
   let content = text;
 
   if (use_custom_tts && !content.startsWith('<speak')) {
@@ -365,7 +403,8 @@ async function _synthOnPremMicrosoft(logger, {
     const post = bent('POST', 'buffer', {
       'X-Microsoft-OutputFormat': trimSilence ? 'raw-8khz-16bit-mono-pcm' : 'audio-16khz-32kbitrate-mono-mp3',
       'Content-Type': 'application/ssml+xml',
-      'User-Agent': 'Jambonz'
+      'User-Agent': 'Jambonz',
+      ...(api_key && {'Ocp-Apim-Subscription-Key': api_key})
     });
     const mp3 = await post(custom_tts_endpoint_url, content);
     return mp3;
@@ -381,10 +420,48 @@ const synthMicrosoft = async(logger, {
   language,
   voice,
   text,
-  filePath
+  filePath,
+  renderForCaching,
+  disableTtsStreaming
 }) => {
   try {
     const {api_key: apiKey, region, use_custom_tts, custom_tts_endpoint, custom_tts_endpoint_url} = credentials;
+    // let clean up the text
+    let content = text;
+    if (use_custom_tts && !content.startsWith('<speak')) {
+      /**
+       * Note: it seems that to use custom voice ssml is required with the voice attribute
+       * Otherwise sending plain text we get "Voice does not match"
+       */
+      content = `<speak>${text}</speak>`;
+    }
+
+    if (content.startsWith('<speak>')) {
+      /* microsoft enforces some properties and uses voice xml element so if the user did not supply do it for them */
+      const words = content.slice(7, -8).trim().replace(/(\r\n|\n|\r)/gm, ' ');
+      // eslint-disable-next-line max-len
+      content = `<speak version="1.0" xmlns="http://www.w3.org/2001/10/synthesis" xmlns:mstts="https://www.w3.org/2001/mstts" xml:lang="${language}"><voice name="${voice}">${words}</voice></speak>`;
+      logger.info({content}, 'synthMicrosoft');
+    }
+    if (!JAMBONES_DISABLE_TTS_STREAMING && !renderForCaching && !disableTtsStreaming) {
+      let params = '';
+      params += `{api_key=${apiKey}`;
+      params += `,language=${language}`;
+      params += ',vendor=microsoft';
+      params += `,voice=${voice}`;
+      params += ',write_cache_file=1';
+      if (region) params += `,region=${region}`;
+      if (custom_tts_endpoint) params += `,endpointId=${custom_tts_endpoint}`;
+      if (custom_tts_endpoint_url) params += `,endpoint=${custom_tts_endpoint_url}`;
+      if (JAMBONES_HTTP_PROXY_IP) params += `,http_proxy_ip=${JAMBONES_HTTP_PROXY_IP}`;
+      if (JAMBONES_HTTP_PROXY_PORT) params += `,http_proxy_port=${JAMBONES_HTTP_PROXY_PORT}`;
+      params += '}';
+      return {
+        filePath: `say:${params}${content.replace(/\n/g, ' ')}`,
+        servedFromCache: false,
+        rtt: 0
+      };
+    }
     if (use_custom_tts && custom_tts_endpoint_url) {
       return await _synthOnPremMicrosoft(logger, {
         credentials,
@@ -396,39 +473,23 @@ const synthMicrosoft = async(logger, {
       });
     }
     const trimSilence = filePath.endsWith('.r8');
-    let content = text;
     const speechConfig = SpeechConfig.fromSubscription(apiKey, region);
     speechConfig.speechSynthesisLanguage = language;
     speechConfig.speechSynthesisVoiceName = voice;
     if (use_custom_tts && custom_tts_endpoint) {
       speechConfig.endpointId = custom_tts_endpoint;
     }
-    if (use_custom_tts && !content.startsWith('<speak')) {
-      /**
-       * Note: it seems that to use custom voice ssml is required with the voice attribute
-       * Otherwise sending plain text we get "Voice does not match"
-       */
-      content = `<speak>${text}</speak>`;
-    }
     speechConfig.speechSynthesisOutputFormat = trimSilence ?
       SpeechSynthesisOutputFormat.Raw8Khz16BitMonoPcm :
       SpeechSynthesisOutputFormat.Audio16Khz32KBitRateMonoMp3;
 
-    if (process.env.JAMBONES_HTTP_PROXY_IP && process.env.JAMBONES_HTTP_PROXY_PORT) {
+    if (JAMBONES_HTTP_PROXY_IP && JAMBONES_HTTP_PROXY_PORT) {
       logger.debug(
-        `synthMicrosoft: using proxy ${process.env.JAMBONES_HTTP_PROXY_IP}:${process.env.JAMBONES_HTTP_PROXY_PORT}`);
-      speechConfig.setProxy(process.env.JAMBONES_HTTP_PROXY_IP, process.env.JAMBONES_HTTP_PROXY_PORT);
+        `synthMicrosoft: using proxy ${JAMBONES_HTTP_PROXY_IP}:${JAMBONES_HTTP_PROXY_PORT}`);
+      speechConfig.setProxy(JAMBONES_HTTP_PROXY_IP, JAMBONES_HTTP_PROXY_PORT);
     }
     const synthesizer = new SpeechSynthesizer(speechConfig);
 
-    if (content.startsWith('<speak>')) {
-      /* microsoft enforces some properties and uses voice xml element so if the user did not supply do it for them */
-      const words = content.slice(7, -8).trim().replace(/(\r\n|\n|\r)/gm, ' ');
-      // eslint-disable-next-line max-len
-      content = `<speak version="1.0" xmlns="http://www.w3.org/2001/10/synthesis" xmlns:mstts="https://www.w3.org/2001/mstts" xml:lang="${language}"><voice name="${voice}">${words}</voice></speak>`;
-      logger.info({content}, 'synthMicrosoft');
-    }
-
     return new Promise((resolve, reject) => {
       const speakAsync = content.startsWith('<speak') ?
         synthesizer.speakSsmlAsync.bind(synthesizer) :
@@ -611,7 +672,7 @@ const synthElevenlabs = async(logger, {
   const opts = !!options && Object.keys(options).length !== 0 ? options : JSON.parse(credOpts || '{}');
 
   /* default to using the streaming interface, unless disabled by env var OR we want just a cache file */
-  if (!process.env.JAMBONES_DISABLE_TTS_STREAMING && !renderForCaching && !disableTtsStreaming) {
+  if (!JAMBONES_DISABLE_TTS_STREAMING && !renderForCaching && !disableTtsStreaming) {
     let params = '';
     params += `{api_key=${api_key}`;
     params += ',vendor=elevenlabs';
@@ -657,10 +718,151 @@ const synthElevenlabs = async(logger, {
   }
 };
 
+const synthPlayHT = async(logger, {
+  credentials, options, stats, voice, text, renderForCaching, disableTtsStreaming
+}) => {
+  const {api_key, user_id, voice_engine, options: credOpts} = credentials;
+  const opts = !!options && Object.keys(options).length !== 0 ? options : JSON.parse(credOpts || '{}');
+
+  /* default to using the streaming interface, unless disabled by env var OR we want just a cache file */
+  if (!JAMBONES_DISABLE_TTS_STREAMING && !renderForCaching && !disableTtsStreaming) {
+    let params = '';
+    params += `{api_key=${api_key}`;
+    params += `,user_id=${user_id}`;
+    params += ',vendor=playht';
+    params += `,voice=${voice}`;
+    params += `,voice_engine=${voice_engine}`;
+    params += ',write_cache_file=1';
+    if (opts.quality) params += `,quality=${opts.quality}`;
+    if (opts.speed) params += `,speed=${opts.speed}`;
+    if (opts.seed) params += `,style=${opts.seed}`;
+    if (opts.temperature) params += `,temperature=${opts.temperature}`;
+    if (opts.emotion) params += `,emotion=${opts.emotion}`;
+    if (opts.voice_guidance) params += `,voice_guidance=${opts.voice_guidance}`;
+    if (opts.style_guidance) params += `,style_guidance=${opts.style_guidance}`;
+    if (opts.text_guidance) params += `,text_guidance=${opts.text_guidance}`;
+    params += '}';
+
+    return {
+      filePath: `say:${params}${text.replace(/\n/g, ' ').replace(/\r/g, ' ')}`,
+      servedFromCache: false,
+      rtt: 0
+    };
+  }
+
+  try {
+    const post = bent('https://api.play.ht', 'POST', 'buffer', {
+      'AUTHORIZATION': api_key,
+      'X-USER-ID': user_id,
+      'Accept': 'audio/mpeg',
+      'Content-Type': 'application/json'
+    });
+    const mp3 = await post('/api/v2/tts/stream', {
+      text,
+      voice,
+      voice_engine,
+      output_format: 'mp3',
+      sample_rate: 8000,
+      ...opts
+    });
+    return mp3;
+  } catch (err) {
+    logger.info({err}, 'synth PlayHT returned error');
+    stats.increment('tts.count', ['vendor:playht', 'accepted:no']);
+    throw err;
+  }
+};
+
+const synthRimelabs = async(logger, {
+  credentials, options, stats, voice, text, renderForCaching, disableTtsStreaming
+}) => {
+  const {api_key, model_id, options: credOpts} = credentials;
+  const opts = !!options && Object.keys(options).length !== 0 ? options : JSON.parse(credOpts || '{}');
+
+  /* default to using the streaming interface, unless disabled by env var OR we want just a cache file */
+  if (!JAMBONES_DISABLE_TTS_STREAMING && !renderForCaching && !disableTtsStreaming) {
+    let params = '';
+    params += `{api_key=${api_key}`;
+    params += `,model_id=${model_id}`;
+    params += ',vendor=rimelabs';
+    params += `,voice=${voice}`;
+    params += ',write_cache_file=1';
+    if (opts.speedAlpha) params += `,speed_alpha=${opts.speedAlpha}`;
+    if (opts.reduceLatency) params += `,reduce_latency=${opts.reduceLatency}`;
+    params += '}';
+
+    return {
+      filePath: `say:${params}${text.replace(/\n/g, ' ').replace(/\r/g, ' ')}`,
+      servedFromCache: false,
+      rtt: 0
+    };
+  }
+
+  try {
+    const post = bent('https://users.rime.ai', 'POST', 'buffer', {
+      'Authorization': `Bearer ${api_key}`,
+      'Accept': 'audio/mp3',
+      'Content-Type': 'application/json'
+    });
+    const mp3 = await post('/v1/rime-tts', {
+      speaker: voice,
+      text,
+      modelId: model_id,
+      samplingRate: 8000,
+      ...opts
+    });
+    return mp3;
+  } catch (err) {
+    logger.info({err}, 'synth rimelabs returned error');
+    stats.increment('tts.count', ['vendor:rimelabs', 'accepted:no']);
+    throw err;
+  }
+};
+const synthVerbio = async(client, logger, {credentials, stats, voice, text, renderForCaching, disableTtsStreaming}) => {
+  //https://doc.speechcenter.verbio.com/#tag/Text-To-Speech-REST-API
+  if (text.length > 2000) {
+    throw new Error('Verbio cannot synthesize for the text length larger than 2000 characters');
+  }
+  const token = await getVerbioAccessToken(client, logger, credentials);
+  if (!process.env.JAMBONES_DISABLE_TTS_STREAMING && !renderForCaching && !disableTtsStreaming) {
+    let params = '';
+    params += `{access_token=${token.access_token}`;
+    params += ',vendor=verbio';
+    params += `,voice=${voice}`;
+    params += ',write_cache_file=1';
+    params += '}';
+
+    return {
+      filePath: `say:${params}${text.replace(/\n/g, ' ')}`,
+      servedFromCache: false,
+      rtt: 0
+    };
+  }
+
+  try {
+    const post = bent('https://us.rest.speechcenter.verbio.com', 'POST', 'buffer', {
+      'Authorization': `Bearer ${token.access_token}`,
+      'User-Agent': 'jambonz',
+      'Content-Type': 'application/json'
+    });
+    const r8 = await post('/api/v1/synthesize', {
+      voice_id: voice,
+      output_sample_rate: '8k',
+      output_encoding: 'pcm16',
+      text
+    });
+    return r8;
+  } catch (err) {
+    logger.info({err}, 'synth Verbio returned error');
+    stats.increment('tts.count', ['vendor:verbio', 'accepted:no']);
+    throw err;
+  }
+};
+
 const synthWhisper = async(logger, {credentials, stats, voice, text, renderForCaching, disableTtsStreaming}) => {
   const {api_key, model_id, baseURL, timeout, speed} = credentials;
   /* if the env is set to stream then bag out, unless we are specifically rendering to generate a cache file */
-  if (!process.env.JAMBONES_DISABLE_TTS_STREAMING && !renderForCaching && !disableTtsStreaming) {
+  if (!JAMBONES_DISABLE_TTS_STREAMING && !renderForCaching && !disableTtsStreaming) {
     let params = '';
     params += `{api_key=${api_key}`;
     params += `,model_id=${model_id}`;
@@ -697,8 +899,22 @@ const synthWhisper = async(logger, {credentials, stats, voice, text, renderForCa
   }
 };
 
-const synthDeepgram = async(logger, {credentials, stats, model, text}) => {
+const synthDeepgram = async(logger, {credentials, stats, model, text, renderForCaching, disableTtsStreaming}) => {
   const {api_key} = credentials;
+  if (!JAMBONES_DISABLE_TTS_STREAMING && !renderForCaching && !disableTtsStreaming) {
+    let params = '';
+    params += `{api_key=${api_key}`;
+    params += ',vendor=deepgram';
+    params += `,voice=${model}`;
+    params += ',write_cache_file=1';
+    params += '}';
+
+    return {
+      filePath: `say:${params}${text.replace(/\n/g, ' ')}`,
+      servedFromCache: false,
+      rtt: 0
+    };
+  }
   try {
     const post = bent('https://api.deepgram.com', 'POST', 'buffer', {
       'Authorization': `Token ${api_key}`,

lib/utils.js

@@ -3,10 +3,10 @@ const {SynthesizerClient} = require('../stubs/nuance/synthesizer_grpc_pb');
 const {RivaSpeechSynthesisClient} = require('../stubs/riva/proto/riva_tts_grpc_pb');
 const {Pool} = require('undici');
 const pool = new Pool('https://auth.crt.nuance.com');
-const HTTP_TIMEOUT = 5000;
 const NUANCE_AUTH_ENDPOINT = 'tts.api.nuance.com:443';
 const grpc = require('@grpc/grpc-js');
 const formurlencoded = require('form-urlencoded');
+const { JAMBONES_DISABLE_TTS_STREAMING, JAMBONES_TTS_TRIM_SILENCE, TMP_FOLDER, HTTP_TIMEOUT } = require('./config');
 
 const debug = require('debug')('jambonz:realtimedb-helpers');
 /**
@@ -19,7 +19,44 @@ const debug = require('debug')('jambonz:realtimedb-helpers');
 function makeSynthKey({account_sid = '', vendor, language, voice, engine = '', text}) {
   const hash = crypto.createHash('sha1');
   hash.update(`${language}:${vendor}:${voice}:${engine}:${text}`);
-  return `tts${account_sid ? (':' + account_sid) : ''}:${hash.digest('hex')}`;
+  const hexHashKey = hash.digest('hex');
+  const accountKey = account_sid ? `:${account_sid}` : '';
+  const namespace = vendor.startsWith('custom') ? vendor : getFileExtension(vendor);
+  const key = `tts${accountKey}:${namespace}:${hexHashKey}`;
+  return key;
+}
+
+function makeFilePath(vendor, key, salt = '') {
+  const extension = getFileExtension(vendor);
+  return `${TMP_FOLDER}/${key.replace('tts:', `tts-${salt}`)}.${extension}`;
+}
+
+function getFileExtension(vendor) {
+  const mp3Extension = 'mp3';
+  const r8Extension = 'r8';
+
+  switch (vendor) {
+    case 'azure':
+    case 'microsoft':
+      if (!JAMBONES_DISABLE_TTS_STREAMING || JAMBONES_TTS_TRIM_SILENCE) {
+        return r8Extension;
+      } else {
+        return mp3Extension;
+      }
+    case 'deepgram':
+    case 'elevenlabs':
+    case 'rimlabs':
+      if (!JAMBONES_DISABLE_TTS_STREAMING) {
+        return r8Extension;
+      } else {
+        return mp3Extension;
+      }
+    case 'nuance':
+    case 'nvidia':
+      return r8Extension;
+    default:
+      return mp3Extension;
+  }
 }
 
 const noopLogger = {
@@ -49,6 +86,12 @@ function makeAwsKey(awsAccessKeyId) {
   return `aws:${hash.digest('hex')}`;
 }
 
+function makeVerbioKey(client_id) {
+  const hash = crypto.createHash('sha1');
+  hash.update(client_id);
+  return `verbio:${hash.digest('hex')}`;
+}
+
 function makeNuanceKey(clientId, secret, scope) {
   const hash = crypto.createHash('sha1');
   hash.update(`${clientId}:${secret}:${scope}`);
@@ -117,11 +160,13 @@ module.exports = {
   makeNuanceKey,
   makeIbmKey,
   makeAwsKey,
+  makeVerbioKey,
   getNuanceAccessToken,
   createNuanceClient,
   createKryptonClient,
   createRivaClient,
   makeBasicAuthHeader,
   NUANCE_AUTH_ENDPOINT,
-  noopLogger
+  noopLogger,
+  makeFilePath
 };

package-lock.json

@@ -1,12 +1,12 @@
 {
   "name": "@jambonz/speech-utils",
-  "version": "0.0.44",
+  "version": "0.1.2",
   "lockfileVersion": 2,
   "requires": true,
   "packages": {
     "": {
       "name": "@jambonz/speech-utils",
-      "version": "0.0.44",
+      "version": "0.1.2",
       "license": "MIT",
       "dependencies": {
         "@aws-sdk/client-polly": "^3.496.0",
@@ -19,7 +19,7 @@
         "form-urlencoded": "^6.1.4",
         "google-protobuf": "^3.21.2",
         "ibm-watson": "^8.0.0",
-        "microsoft-cognitiveservices-speech-sdk": "1.34.0",
+        "microsoft-cognitiveservices-speech-sdk": "1.36.0",
         "openai": "^4.25.0",
         "undici": "^6.4.0"
       },
@@ -1124,14 +1124,6 @@
         "node": "^12.22.0 || ^14.17.0 || >=16.0.0"
       }
     },
-    "node_modules/@fastify/busboy": {
-      "version": "2.1.0",
-      "resolved": "https://registry.npmjs.org/@fastify/busboy/-/busboy-2.1.0.tgz",
-      "integrity": "sha512-+KpH+QxZU7O4675t3mnkQKcZZg56u+K/Ct2K+N2AZYNVK8kyeo/bI18tI8aPm3tvNNRyTWfj6s5tnGNlcbQRsA==",
-      "engines": {
-        "node": ">=14"
-      }
-    },
     "node_modules/@google-cloud/text-to-speech": {
       "version": "5.0.2",
       "resolved": "https://registry.npmjs.org/@google-cloud/text-to-speech/-/text-to-speech-5.0.2.tgz",
@@ -3125,13 +3117,14 @@
       }
     },
     "node_modules/es5-ext": {
-      "version": "0.10.62",
-      "resolved": "https://registry.npmjs.org/es5-ext/-/es5-ext-0.10.62.tgz",
-      "integrity": "sha512-BHLqn0klhEpnOKSrzn/Xsz2UIW8j+cGmo9JLzr8BiUapV8hPL9+FliFqjwr9ngW7jWdnxv6eO+/LqyhJVqgrjA==",
+      "version": "0.10.64",
+      "resolved": "https://registry.npmjs.org/es5-ext/-/es5-ext-0.10.64.tgz",
+      "integrity": "sha512-p2snDhiLaXe6dahss1LddxqEm+SkuDvV8dnIQG0MWjyHpcMNfXKPE+/Cc0y+PhxJX3A4xGNeFCj5oc0BUh6deg==",
       "hasInstallScript": true,
       "dependencies": {
         "es6-iterator": "^2.0.3",
         "es6-symbol": "^3.1.3",
+        "esniff": "^2.0.1",
         "next-tick": "^1.1.0"
       },
       "engines": {
@@ -3278,6 +3271,25 @@
         "url": "https://opencollective.com/eslint"
       }
     },
+    "node_modules/esniff": {
+      "version": "2.0.1",
+      "resolved": "https://registry.npmjs.org/esniff/-/esniff-2.0.1.tgz",
+      "integrity": "sha512-kTUIGKQ/mDPFoJ0oVfcmyJn4iBDRptjNVIzwIFR7tqWXdVI9xfA2RMwY/gbSpJG3lkdWNEjLap/NqVHZiJsdfg==",
+      "dependencies": {
+        "d": "^1.0.1",
+        "es5-ext": "^0.10.62",
+        "event-emitter": "^0.3.5",
+        "type": "^2.7.2"
+      },
+      "engines": {
+        "node": ">=0.10"
+      }
+    },
+    "node_modules/esniff/node_modules/type": {
+      "version": "2.7.2",
+      "resolved": "https://registry.npmjs.org/type/-/type-2.7.2.tgz",
+      "integrity": "sha512-dzlvlNlt6AXU7EBSfpAscydQ7gXB+pPGsPnfJnZpiNJBDj7IaJzQlBZYGdEi4R9HmPdBv2XmWJ6YUtoTa7lmCw=="
+    },
     "node_modules/espree": {
       "version": "9.6.1",
       "resolved": "https://registry.npmjs.org/espree/-/espree-9.6.1.tgz",
@@ -3350,6 +3362,15 @@
         "node": ">=0.10.0"
       }
     },
+    "node_modules/event-emitter": {
+      "version": "0.3.5",
+      "resolved": "https://registry.npmjs.org/event-emitter/-/event-emitter-0.3.5.tgz",
+      "integrity": "sha512-D9rRn9y7kLPnJ+hMq7S/nhvoKwwvVJahBi2BPmx3bvbsEdK3W9ii8cBSGjP+72/LnM4n6fo3+dkCX5FeTQruXA==",
+      "dependencies": {
+        "d": "1",
+        "es5-ext": "~0.10.14"
+      }
+    },
     "node_modules/event-target-shim": {
       "version": "5.0.1",
       "resolved": "https://registry.npmjs.org/event-target-shim/-/event-target-shim-5.0.1.tgz",
@@ -3551,9 +3572,9 @@
       "dev": true
     },
     "node_modules/follow-redirects": {
-      "version": "1.15.5",
-      "resolved": "https://registry.npmjs.org/follow-redirects/-/follow-redirects-1.15.5.tgz",
-      "integrity": "sha512-vSFWUON1B+yAw1VN4xMfxgn5fTUiaOzAJCKBwIIgT/+7CuGy9+r+5gITvP62j3RmaD5Ph65UaERdOSRGUzZtgw==",
+      "version": "1.15.6",
+      "resolved": "https://registry.npmjs.org/follow-redirects/-/follow-redirects-1.15.6.tgz",
+      "integrity": "sha512-wWN62YITEaOpSK584EZXJafH1AGpO8RVgElfkuXbTOrPX4fIfOyEpW/CsiNd8JdYrAoOvafRTOEnvsO++qCqFA==",
       "funding": [
         {
           "type": "individual",
@@ -5084,9 +5105,9 @@
       }
     },
     "node_modules/microsoft-cognitiveservices-speech-sdk": {
-      "version": "1.34.0",
-      "resolved": "https://registry.npmjs.org/microsoft-cognitiveservices-speech-sdk/-/microsoft-cognitiveservices-speech-sdk-1.34.0.tgz",
-      "integrity": "sha512-WAR0YqouRzVux2kI+f5wTPC6NyJgIVC1g65d79dJ9I32WPJs2kK+eb/BMB6mhSdCjackO5FsrW7JLaQ/vB1heQ==",
+      "version": "1.36.0",
+      "resolved": "https://registry.npmjs.org/microsoft-cognitiveservices-speech-sdk/-/microsoft-cognitiveservices-speech-sdk-1.36.0.tgz",
+      "integrity": "sha512-wPxuEXgjLdqMMIrdBtl8jquGahLV19LQE0ie8MI/PcBcNLG5buVzwS2rQEyHMsRGx+C/4OdBo1ROdNIUzCm4Lg==",
       "dependencies": {
         "@types/webrtc": "^0.0.37",
         "agent-base": "^6.0.1",
@@ -6851,12 +6872,9 @@
       }
     },
     "node_modules/undici": {
-      "version": "6.4.0",
-      "resolved": "https://registry.npmjs.org/undici/-/undici-6.4.0.tgz",
-      "integrity": "sha512-wYaKgftNqf6Je7JQ51YzkEkEevzOgM7at5JytKO7BjaURQpERW8edQSMrr2xb+Yv4U8Yg47J24+lc9+NbeXMFA==",
-      "dependencies": {
-        "@fastify/busboy": "^2.0.0"
-      },
+      "version": "6.11.1",
+      "resolved": "https://registry.npmjs.org/undici/-/undici-6.11.1.tgz",
+      "integrity": "sha512-KyhzaLJnV1qa3BSHdj4AZ2ndqI0QWPxYzaIOio0WzcEJB9gvuysprJSLtpvc2D9mhR9jPDUk7xlJlZbH2KR5iw==",
       "engines": {
         "node": ">=18.0"
       }
@@ -8092,11 +8110,6 @@
       "integrity": "sha512-gMsVel9D7f2HLkBma9VbtzZRehRogVRfbr++f06nL2vnCGCNlzOD+/MUov/F4p8myyAHspEhVobgjpX64q5m6A==",
       "dev": true
     },
-    "@fastify/busboy": {
-      "version": "2.1.0",
-      "resolved": "https://registry.npmjs.org/@fastify/busboy/-/busboy-2.1.0.tgz",
-      "integrity": "sha512-+KpH+QxZU7O4675t3mnkQKcZZg56u+K/Ct2K+N2AZYNVK8kyeo/bI18tI8aPm3tvNNRyTWfj6s5tnGNlcbQRsA=="
-    },
     "@google-cloud/text-to-speech": {
       "version": "5.0.2",
       "resolved": "https://registry.npmjs.org/@google-cloud/text-to-speech/-/text-to-speech-5.0.2.tgz",
@@ -9652,12 +9665,13 @@
       }
     },
     "es5-ext": {
-      "version": "0.10.62",
-      "resolved": "https://registry.npmjs.org/es5-ext/-/es5-ext-0.10.62.tgz",
-      "integrity": "sha512-BHLqn0klhEpnOKSrzn/Xsz2UIW8j+cGmo9JLzr8BiUapV8hPL9+FliFqjwr9ngW7jWdnxv6eO+/LqyhJVqgrjA==",
+      "version": "0.10.64",
+      "resolved": "https://registry.npmjs.org/es5-ext/-/es5-ext-0.10.64.tgz",
+      "integrity": "sha512-p2snDhiLaXe6dahss1LddxqEm+SkuDvV8dnIQG0MWjyHpcMNfXKPE+/Cc0y+PhxJX3A4xGNeFCj5oc0BUh6deg==",
       "requires": {
         "es6-iterator": "^2.0.3",
         "es6-symbol": "^3.1.3",
+        "esniff": "^2.0.1",
         "next-tick": "^1.1.0"
       }
     },
@@ -9766,6 +9780,24 @@
       "integrity": "sha512-wpc+LXeiyiisxPlEkUzU6svyS1frIO3Mgxj1fdy7Pm8Ygzguax2N3Fa/D/ag1WqbOprdI+uY6wMUl8/a2G+iag==",
       "dev": true
     },
+    "esniff": {
+      "version": "2.0.1",
+      "resolved": "https://registry.npmjs.org/esniff/-/esniff-2.0.1.tgz",
+      "integrity": "sha512-kTUIGKQ/mDPFoJ0oVfcmyJn4iBDRptjNVIzwIFR7tqWXdVI9xfA2RMwY/gbSpJG3lkdWNEjLap/NqVHZiJsdfg==",
+      "requires": {
+        "d": "^1.0.1",
+        "es5-ext": "^0.10.62",
+        "event-emitter": "^0.3.5",
+        "type": "^2.7.2"
+      },
+      "dependencies": {
+        "type": {
+          "version": "2.7.2",
+          "resolved": "https://registry.npmjs.org/type/-/type-2.7.2.tgz",
+          "integrity": "sha512-dzlvlNlt6AXU7EBSfpAscydQ7gXB+pPGsPnfJnZpiNJBDj7IaJzQlBZYGdEi4R9HmPdBv2XmWJ6YUtoTa7lmCw=="
+        }
+      }
+    },
     "espree": {
       "version": "9.6.1",
       "resolved": "https://registry.npmjs.org/espree/-/espree-9.6.1.tgz",
@@ -9813,6 +9845,15 @@
       "integrity": "sha512-kVscqXk4OCp68SZ0dkgEKVi6/8ij300KBWTJq32P/dYeWTSwK41WyTxalN1eRmA5Z9UU/LX9D7FWSmV9SAYx6g==",
       "dev": true
     },
+    "event-emitter": {
+      "version": "0.3.5",
+      "resolved": "https://registry.npmjs.org/event-emitter/-/event-emitter-0.3.5.tgz",
+      "integrity": "sha512-D9rRn9y7kLPnJ+hMq7S/nhvoKwwvVJahBi2BPmx3bvbsEdK3W9ii8cBSGjP+72/LnM4n6fo3+dkCX5FeTQruXA==",
+      "requires": {
+        "d": "1",
+        "es5-ext": "~0.10.14"
+      }
+    },
     "event-target-shim": {
       "version": "5.0.1",
       "resolved": "https://registry.npmjs.org/event-target-shim/-/event-target-shim-5.0.1.tgz",
@@ -9964,9 +10005,9 @@
       "dev": true
     },
     "follow-redirects": {
-      "version": "1.15.5",
-      "resolved": "https://registry.npmjs.org/follow-redirects/-/follow-redirects-1.15.5.tgz",
-      "integrity": "sha512-vSFWUON1B+yAw1VN4xMfxgn5fTUiaOzAJCKBwIIgT/+7CuGy9+r+5gITvP62j3RmaD5Ph65UaERdOSRGUzZtgw=="
+      "version": "1.15.6",
+      "resolved": "https://registry.npmjs.org/follow-redirects/-/follow-redirects-1.15.6.tgz",
+      "integrity": "sha512-wWN62YITEaOpSK584EZXJafH1AGpO8RVgElfkuXbTOrPX4fIfOyEpW/CsiNd8JdYrAoOvafRTOEnvsO++qCqFA=="
     },
     "for-each": {
       "version": "0.3.3",
@@ -11108,9 +11149,9 @@
       }
     },
     "microsoft-cognitiveservices-speech-sdk": {
-      "version": "1.34.0",
-      "resolved": "https://registry.npmjs.org/microsoft-cognitiveservices-speech-sdk/-/microsoft-cognitiveservices-speech-sdk-1.34.0.tgz",
-      "integrity": "sha512-WAR0YqouRzVux2kI+f5wTPC6NyJgIVC1g65d79dJ9I32WPJs2kK+eb/BMB6mhSdCjackO5FsrW7JLaQ/vB1heQ==",
+      "version": "1.36.0",
+      "resolved": "https://registry.npmjs.org/microsoft-cognitiveservices-speech-sdk/-/microsoft-cognitiveservices-speech-sdk-1.36.0.tgz",
+      "integrity": "sha512-wPxuEXgjLdqMMIrdBtl8jquGahLV19LQE0ie8MI/PcBcNLG5buVzwS2rQEyHMsRGx+C/4OdBo1ROdNIUzCm4Lg==",
       "requires": {
         "@types/webrtc": "^0.0.37",
         "agent-base": "^6.0.1",
@@ -12405,12 +12446,9 @@
       }
     },
     "undici": {
-      "version": "6.4.0",
-      "resolved": "https://registry.npmjs.org/undici/-/undici-6.4.0.tgz",
-      "integrity": "sha512-wYaKgftNqf6Je7JQ51YzkEkEevzOgM7at5JytKO7BjaURQpERW8edQSMrr2xb+Yv4U8Yg47J24+lc9+NbeXMFA==",
-      "requires": {
-        "@fastify/busboy": "^2.0.0"
-      }
+      "version": "6.11.1",
+      "resolved": "https://registry.npmjs.org/undici/-/undici-6.11.1.tgz",
+      "integrity": "sha512-KyhzaLJnV1qa3BSHdj4AZ2ndqI0QWPxYzaIOio0WzcEJB9gvuysprJSLtpvc2D9mhR9jPDUk7xlJlZbH2KR5iw=="
     },
     "undici-types": {
       "version": "5.26.5",

package.json

@@ -1,6 +1,6 @@
 {
   "name": "@jambonz/speech-utils",
-  "version": "0.0.44",
+  "version": "0.1.2",
   "description": "TTS-related speech utilities for jambonz",
   "main": "index.js",
   "author": "Dave Horton",
@@ -12,6 +12,7 @@
     "test": "NODE_ENV=test node test/ ",
     "coverage": "nyc --reporter html --report-dir ./coverage npm run test",
     "jslint": "eslint index.js lib",
+    "jslint:fix": "eslint --fix '**/*.js'",
     "build": "./build_stubs.sh"
   },
   "repository": {
@@ -34,7 +35,7 @@
     "form-urlencoded": "^6.1.4",
     "google-protobuf": "^3.21.2",
     "ibm-watson": "^8.0.0",
-    "microsoft-cognitiveservices-speech-sdk": "1.34.0",
+    "microsoft-cognitiveservices-speech-sdk": "1.36.0",
     "openai": "^4.25.0",
     "undici": "^6.4.0"
   },

stubs/nuance/synthesizer_grpc_pb.js

@@ -62,9 +62,9 @@ function deserialize_nuance_tts_v1_UnarySynthesisResponse(buffer_arg) {
 
 //
 // The Synthesizer service offers these functionalities:
-// - GetVoices: Queries the list of available voices, with filters to reduce the search space.  
-// - Synthesize: Synthesizes audio from input text and parameters, and returns an audio stream. 
-// - UnarySynthesize: Synthesizes audio from input text and parameters, and returns a single audio response. 
+// - GetVoices: Queries the list of available voices, with filters to reduce the search space.
+// - Synthesize: Synthesizes audio from input text and parameters, and returns an audio stream.
+// - UnarySynthesize: Synthesizes audio from input text and parameters, and returns a single audio response.
 var SynthesizerService = exports.SynthesizerService = {
   getVoices: {
     path: '/nuance.tts.v1.Synthesizer/GetVoices',

stubs/riva/proto/riva_audio_grpc_pb.js

@@ -1 +1 @@
-// GENERATED CODE -- NO SERVICES IN PROTO
+// GENERATED CODE -- NO SERVICES IN PROTO

stubs/riva/proto/riva_tts_grpc_pb.js

@@ -57,7 +57,7 @@ function deserialize_nvidia_riva_tts_SynthesizeSpeechResponse(buffer_arg) {
 var RivaSpeechSynthesisService = exports.RivaSpeechSynthesisService = {
   // Used to request text-to-speech from the service. Submit a request containing the
 // desired text and configuration, and receive audio bytes in the requested format.
-synthesize: {
+  synthesize: {
     path: '/nvidia.riva.tts.RivaSpeechSynthesis/Synthesize',
     requestStream: false,
     responseStream: false,
@@ -69,9 +69,9 @@ synthesize: {
     responseDeserialize: deserialize_nvidia_riva_tts_SynthesizeSpeechResponse,
   },
   // Used to request text-to-speech returned via stream as it becomes available.
-// Submit a SynthesizeSpeechRequest with desired text and configuration,
-// and receive stream of bytes in the requested format.
-synthesizeOnline: {
+  // Submit a SynthesizeSpeechRequest with desired text and configuration,
+  // and receive stream of bytes in the requested format.
+  synthesizeOnline: {
     path: '/nvidia.riva.tts.RivaSpeechSynthesis/SynthesizeOnline',
     requestStream: false,
     responseStream: true,
@@ -83,7 +83,7 @@ synthesizeOnline: {
     responseDeserialize: deserialize_nvidia_riva_tts_SynthesizeSpeechResponse,
   },
   // Enables clients to request the configuration of the current Synthesize service, or a specific model within the service.
-getRivaSynthesisConfig: {
+  getRivaSynthesisConfig: {
     path: '/nvidia.riva.tts.RivaSpeechSynthesis/GetRivaSynthesisConfig',
     requestStream: false,
     responseStream: false,

test/list-voices.js

@@ -12,6 +12,32 @@ const stats = {
   histogram: () => {}
 };
 
+test('Verbio - get Access key and voices', async(t) => {
+  const fn = require('..');
+  const {client, getTtsVoices, getVerbioAccessToken} = fn(opts, logger);
+  if (!process.env.VERBIO_CLIENT_ID || !process.env.VERBIO_CLIENT_SECRET) {
+    t.pass('skipping Verbio test since no Verbio Keys provided');
+    t.end();
+    client.quit();
+    return;
+  }
+
+  try {
+    const credentials = {
+      client_id: process.env.VERBIO_CLIENT_ID,
+      client_secret: process.env.VERBIO_CLIENT_SECRET
+    };
+    let obj = await getVerbioAccessToken(credentials);
+    t.ok(obj.access_token , 'successfully received access token not from cache');
+    const voices = await getTtsVoices({vendor: 'verbio', credentials});
+    t.ok(voices && voices.length != 0, 'successfully received verbio voices');
+  } catch (err) {
+    console.error(err);
+    t.end(err);
+  }
+  client.quit();
+});
+
 test('IBM - create access key', async(t) => {
   const fn = require('..');
   const {client, getIbmAccessToken} = fn(opts, logger);

test/synth.js

@@ -162,6 +162,33 @@ test('AWS speech synth tests', async(t) => {
   client.quit();
 });
 
+test('AWS speech synth tests by RoleArn', async(t) => {
+  const fn = require('..');
+  const {synthAudio, client} = fn(opts, logger);
+
+  if (!process.env.AWS_ROLE_ARN || !process.env.AWS_REGION) {
+    t.pass('skipping AWS speech synth tests by RoleArn since AWS_ROLE_ARN or AWS_REGION not provided');
+    return t.end();
+  }
+  try {
+    let opts = await synthAudio(stats, {
+      vendor: 'aws',
+      credentials: {
+        roleArn: process.env.AWS_ROLE_ARN,
+        region: process.env.AWS_REGION,
+      },
+      language: 'en-US',
+      voice: 'Joey',
+      text: 'This is a test.  This is only a test',
+    });
+    t.ok(!opts.servedFromCache, `successfully synthesized aws by roleArn audio to ${opts.filePath}`);
+  } catch (err) {
+    console.error(err);
+    t.end(err);
+  }
+  client.quit();
+});
+
 test('Azure speech synth tests', async(t) => {
   const fn = require('..');
   const {synthAudio, client} = fn(opts, logger);
@@ -188,6 +215,7 @@ test('Azure speech synth tests', async(t) => {
       language: 'en-US',
       voice: 'en-US-ChristopherNeural',
       text: longText,
+      renderForCaching: true
     });
     t.ok(!opts.servedFromCache, `successfully synthesized microsoft audio to ${opts.filePath}`);
     if (process.env.JAMBONES_HTTP_PROXY_IP && process.env.JAMBONES_HTTP_PROXY_PORT) {
@@ -203,6 +231,7 @@ test('Azure speech synth tests', async(t) => {
       language: 'en-US',
       voice: 'en-US-ChristopherNeural',
       text: longText,
+      renderForCaching: true
     });
     t.ok(opts.servedFromCache, `successfully retrieved microsoft audio from cache ${opts.filePath}`);
   } catch (err) {
@@ -237,6 +266,7 @@ test('Azure SSML tests', async(t) => {
       language: 'en-US',
       voice: 'en-US-ChristopherNeural',
       text,
+      renderForCaching: true
     });
     t.ok(!opts.servedFromCache, `successfully synthesized microsoft audio to ${opts.filePath}`);
     if (process.env.JAMBONES_HTTP_PROXY_IP && process.env.JAMBONES_HTTP_PROXY_PORT) {
@@ -252,6 +282,7 @@ test('Azure SSML tests', async(t) => {
       language: 'en-US',
       voice: 'en-US-ChristopherNeural',
       text,
+      renderForCaching: true
     });
     t.ok(opts.servedFromCache, `successfully retrieved microsoft audio from cache ${opts.filePath}`);
   } catch (err) {
@@ -283,6 +314,7 @@ test('Azure custom voice speech synth tests', async(t) => {
       language: 'en-US',
       voice: process.env.MICROSOFT_CUSTOM_VOICE,
       text,
+      renderForCaching: true
     });
     t.ok(!opts.servedFromCache, `successfully synthesized microsoft audio to ${opts.filePath}`);
 
@@ -297,6 +329,7 @@ test('Azure custom voice speech synth tests', async(t) => {
       language: 'en-US',
       voice: process.env.MICROSOFT_CUSTOM_VOICE,
       text,
+      renderForCaching: true
     });
     t.ok(opts.servedFromCache, `successfully retrieved microsoft custom voice audio from cache ${opts.filePath}`);
   } catch (err) {
@@ -543,6 +576,81 @@ test('Elevenlabs speech synth tests', async(t) => {
   client.quit();
 })
 
+test('PlayHT speech synth tests', async(t) => {
+  const fn = require('..');
+  const {synthAudio, client} = fn(opts, logger);
+
+  if (!process.env.PLAYHT_API_KEY || !process.env.PLAYHT_USER_ID) {
+    t.pass('skipping PlayHT speech synth tests since PLAYHT_API_KEY or PLAYHT_USER_ID is/are not provided');
+    return t.end();
+  }
+  const text = 'Hi there and welcome to jambones!';
+  try {
+    let opts = await synthAudio(stats, {
+      vendor: 'playht',
+      credentials: {
+        api_key: process.env.PLAYHT_API_KEY,
+        user_id: process.env.PLAYHT_USER_ID,
+        voice_engine: 'PlayHT2.0-turbo',
+        options: JSON.stringify({
+          quality: "medium",
+          speed: 1,
+          seed: 1,
+          temperature: 1,
+          emotion: "female_happy",
+          voice_guidance: 3,
+          style_guidance: 20,
+          text_guidance: 1,
+        })
+      },
+      language: 'en-US',
+      voice: 's3://voice-cloning-zero-shot/d9ff78ba-d016-47f6-b0ef-dd630f59414e/female-cs/manifest.json',
+      text,
+      renderForCaching: true
+    });
+    t.ok(!opts.servedFromCache, `successfully playht eleven audio to ${opts.filePath}`);
+
+  } catch (err) {
+    console.error(JSON.stringify(err));
+    t.end(err);
+  }
+  client.quit();
+});
+
+test('rimelabs speech synth tests', async(t) => {
+  const fn = require('..');
+  const {synthAudio, client} = fn(opts, logger);
+
+  if (!process.env.RIMELABS_API_KEY) {
+    t.pass('skipping rimelabs speech synth tests since RIMELABS_API_KEY is not provided');
+    return t.end();
+  }
+  const text = 'Hi there and welcome to jambones!';
+  try {
+    let opts = await synthAudio(stats, {
+      vendor: 'rimelabs',
+      credentials: {
+        api_key: process.env.RIMELABS_API_KEY,
+        model_id: 'mist',
+        options: JSON.stringify({
+          speedAlpha: 1.0,
+          reduceLatency: false
+        })
+      },
+      language: 'en-US',
+      voice: 'amber',
+      text,
+      renderForCaching: true
+    });
+    t.ok(!opts.servedFromCache, `successfully synthesized rimelabs audio to ${opts.filePath}`);
+
+  } catch (err) {
+    console.error(JSON.stringify(err));
+    t.end(err);
+  }
+  client.quit();
+});
+
 test('whisper speech synth tests', async(t) => {
   const fn = require('..');
   const {synthAudio, client} = fn(opts, logger);
@@ -562,6 +670,40 @@ test('whisper speech synth tests', async(t) => {
       language: 'en-US',
       voice: 'alloy',
       text,
+      renderForCaching: true
+    });
+    t.ok(!opts.servedFromCache, `successfully synthesized whisper audio to ${opts.filePath}`);
+
+  } catch (err) {
+    console.error(JSON.stringify(err));
+    t.end(err);
+  }
+  client.quit();
+});
+
+test('Verbio speech synth tests', async(t) => {
+  const fn = require('..');
+  const {synthAudio, client} = fn(opts, logger);
+
+  if (!process.env.VERBIO_CLIENT_ID || !process.env.VERBIO_CLIENT_SECRET) {
+    t.pass('skipping Verbio Synthesize test since no Verbio Keys provided');
+    t.end();
+    client.quit();
+    return;
+  }
+
+  const text = 'Hi there and welcome to jambones!';
+  try {
+    let opts = await synthAudio(stats, {
+      vendor: 'verbio',
+      credentials: {
+        client_id: process.env.VERBIO_CLIENT_ID,
+        client_secret: process.env.VERBIO_CLIENT_SECRET
+      },
+      language: 'en-US',
+      voice: 'tommy_en-us',
+      text,
+      renderForCaching: true
     });
     t.ok(!opts.servedFromCache, `successfully synthesized whisper audio to ${opts.filePath}`);
 
@@ -589,6 +731,7 @@ test('Deepgram speech synth tests', async(t) => {
       },
       model: 'aura-asteria-en',
       text,
+      renderForCaching: true
     });
     t.ok(!opts.servedFromCache, `successfully synthesized deepgram audio to ${opts.filePath}`);
 
@@ -646,8 +789,19 @@ test('TTS Cache tests', async(t) => {
     t.ok(error, `error returned when specified key was not found`);
 
     // make sure other tts keys are still there
-    const cached = (await client.keys('tts:*')).length;
-    t.ok(cached >= 1, `successfully kept all non-specified tts records in cache`);
+    const cached = await client.keys('tts:*')
+    t.ok(cached.length >= 1, `successfully kept all non-specified tts records in cache`);
+
+    // retrieve keys from cache and check the key contains the file extension    
+    let key = cached[0];
+    t.ok(key.includes('mp3'), `tts cache extension shoult be part of the key and equal mp3`);
+
+    process.env.VG_TRIM_TTS_SILENCE = 'true';    
+    await client.set(makeSynthKey({ vendor: 'azure' }), 'value');
+
+    const r8Keys = await client.keys('tts:r8*');
+    key = r8Keys[0];
+    t.ok(key.includes('r8'), `tts cache extension shoult be part of the key and equal r8`);
 
   } catch (err) {
     console.error(JSON.stringify(err));