0.0.50

support mod_playht_tts

index.js

@@ -12,6 +12,7 @@ module.exports = (opts, logger) => {
     client,
     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),
     getNuanceAccessToken: require('./lib/get-nuance-access-token').bind(null, client, logger),
     getIbmAccessToken: require('./lib/get-ibm-access-token').bind(null, client, logger),

lib/add-file-to-cache.js

@@ -0,0 +1,30 @@
+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
+
+async function addFileToCache(client, logger, path,
+  {account_sid, vendor, language, voice, deploymentId, engine, text}) {
+  let key;
+  logger = logger || noopLogger;
+
+  try {
+    key = makeSynthKey({
+      account_sid,
+      vendor,
+      language: language || '',
+      voice: voice || deploymentId,
+      engine,
+      text,
+    });
+    const audioBuffer = await fs.readFile(path);
+    await client.setex(key, EXPIRES, audioBuffer.toString('base64'));
+  } catch (err) {
+    logger.error(err, 'addFileToCache: Error');
+    return;
+  }
+
+  logger.debug(`addFileToCache: added ${path} to cache with key ${key}`);
+  return key;
+}
+
+module.exports = addFileToCache;

lib/get-tts-voices.js

@@ -97,7 +97,7 @@ const getAwsVoices = async(_client, logger, credentials) => {
         secretAccessKey
       }
     });
-    const command = new DescribeVoicesCommand({LanguageCode: 'en-US'});
+    const command = new DescribeVoicesCommand({});
     const response = await client.send(command);
     return response;
   } catch (err) {

lib/synth-audio.js

@@ -77,7 +77,7 @@ const trimTrailingSilence = (buffer) => {
  */
 async function synthAudio(client, logger, stats, { account_sid,
   vendor, language, voice, gender, text, engine, salt, model, credentials, deploymentId,
-  disableTtsCache, renderForCaching, options
+  disableTtsCache, renderForCaching, disableTtsStreaming, options
 }) {
   let audioBuffer;
   let servedFromCache = false;
@@ -85,7 +85,7 @@ async function synthAudio(client, logger, stats, { account_sid,
   logger = logger || noopLogger;
 
   assert.ok(['google', 'aws', 'polly', 'microsoft',
-    'wellsaid', 'nuance', 'nvidia', 'ibm', 'elevenlabs', 'whisper', 'deepgram'].includes(vendor) ||
+    'wellsaid', 'nuance', 'nvidia', 'ibm', 'elevenlabs', 'whisper', 'deepgram', 'playht'].includes(vendor) ||
   vendor.startsWith('custom'),
   `synthAudio supported vendors are google, aws, microsoft, nuance, nvidia and wellsaid, not ${vendor}`);
   if ('google' === vendor) {
@@ -123,6 +123,11 @@ async function synthAudio(client, logger, stats, { account_sid,
     assert.ok(voice, 'synthAudio requires voice when elevenlabs is used');
     assert.ok(credentials.api_key, 'synthAudio requires api_key when elevenlabs is used');
     assert.ok(credentials.model_id, 'synthAudio requires model_id when elevenlabs is used');
+  }  else if ('playht' === vendor) {
+    assert.ok(voice, 'synthAudio requires voice when playht is used');
+    assert.ok(credentials.api_key, 'synthAudio requires api_key when playht is used');
+    assert.ok(credentials.user_id, 'synthAudio requires user_id when playht is used');
+    assert.ok(credentials.voice_engine, 'synthAudio requires voice_engine when playht is used');
   } else if ('whisper' === vendor) {
     assert.ok(voice, 'synthAudio requires voice when whisper is used');
     assert.ok(credentials.model_id, 'synthAudio requires model when whisper is used');
@@ -141,8 +146,12 @@ async function synthAudio(client, logger, stats, { account_sid,
   let filePath;
   if (['nuance', 'nvidia'].includes(vendor) ||
     (
-      process.env.JAMBONES_TTS_TRIM_SILENCE &&
+      (process.env.JAMBONES_TTS_TRIM_SILENCE || !process.env.JAMBONES_DISABLE_TTS_STREAMING) &&
       ['microsoft', 'azure'].includes(vendor)
+    ) ||
+    (
+      !process.env.JAMBONES_DISABLE_TTS_STREAMING &&
+      ['elevenlabs', 'deepgram'].includes(vendor)
     )
   ) {
     filePath = `${TMP_FOLDER}/${key.replace('tts:', `tts-${salt || ''}`)}.r8`;
@@ -179,7 +188,9 @@ async function synthAudio(client, logger, stats, { account_sid,
       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});
+        if (audioBuffer?.filePath) return audioBuffer;
         break;
       case 'nuance':
         model = model || 'enhanced';
@@ -196,20 +207,25 @@ async function synthAudio(client, logger, stats, { account_sid,
         break;
       case 'elevenlabs':
         audioBuffer = await synthElevenlabs(logger, {
-          credentials, options, stats, language, voice, text, renderForCaching, filePath
+          credentials, options, stats, language, voice, text, renderForCaching, disableTtsStreaming, filePath
         });
-        if (typeof audioBuffer === 'object' && audioBuffer.filePath) {
-          return audioBuffer;
-        }
-        else {
-          audioBuffer = await synthElevenlabs(logger, {credentials, options, stats, language, voice, text, filePath});
-        }
+        if (audioBuffer?.filePath) return audioBuffer;
+        break;
+      case 'playht':
+        audioBuffer = await synthPlayHT(logger, {
+          credentials, options, stats, language, voice, text, renderForCaching, disableTtsStreaming, filePath
+        });
+        if (audioBuffer?.filePath) return audioBuffer;
         break;
       case 'whisper':
-        audioBuffer = await synthWhisper(logger, {credentials, stats, voice, text});
+        audioBuffer = await synthWhisper(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});
+        if (audioBuffer?.filePath) return audioBuffer;
         break;
       case vendor.startsWith('custom') ? vendor : 'cant_match_value':
         ({ audioBuffer, filePath } = await synthCustomVendor(logger,
@@ -380,10 +396,47 @@ 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 (!process.env.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=${custom_tts_endpoint}`;
+      if (process.env.JAMBONES_HTTP_PROXY_IP) params += `,http_proxy_ip=${process.env.JAMBONES_HTTP_PROXY_IP}`;
+      if (process.env.JAMBONES_HTTP_PROXY_PORT) params += `,http_proxy_port=${process.env.JAMBONES_HTTP_PROXY_PORT}`;
+      params += '}';
+      return {
+        filePath: `say:${params}${content.replace(/\n/g, ' ')}`,
+        servedFromCache: false,
+        rtt: 0
+      };
+    }
     if (use_custom_tts && custom_tts_endpoint_url) {
       return await _synthOnPremMicrosoft(logger, {
         credentials,
@@ -395,20 +448,12 @@ const synthMicrosoft = async(logger, {
       });
     }
     const trimSilence = filePath.endsWith('.r8');
-    let content = text;
     const speechConfig = SpeechConfig.fromSubscription(apiKey, region);
     speechConfig.speechSynthesisLanguage = language;
     speechConfig.speechSynthesisVoiceName = voice;
     if (use_custom_tts && custom_tts_endpoint) {
       speechConfig.endpointId = custom_tts_endpoint;
     }
-    if (use_custom_tts && !content.startsWith('<speak')) {
-      /**
-       * Note: it seems that to use custom voice ssml is required with the voice attribute
-       * Otherwise sending plain text we get "Voice does not match"
-       */
-      content = `<speak>${text}</speak>`;
-    }
     speechConfig.speechSynthesisOutputFormat = trimSilence ?
       SpeechSynthesisOutputFormat.Raw8Khz16BitMonoPcm :
       SpeechSynthesisOutputFormat.Audio16Khz32KBitRateMonoMp3;
@@ -420,14 +465,6 @@ const synthMicrosoft = async(logger, {
     }
     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) :
@@ -603,16 +640,21 @@ const synthCustomVendor = async(logger, {credentials, stats, language, voice, te
   }
 };
 
-const synthElevenlabs = async(logger, {credentials, options, stats, language, voice, text, renderForCaching}) => {
+const synthElevenlabs = 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 || '{}');
 
-  /* if the env is set to stream then bag out, unless we are specifically rendering to generate a cache file */
-  if (process.env.JAMBONES_ELEVENLABS_STREAMING && !renderForCaching) {
+  /* 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) {
     let params = '';
     params += `{api_key=${api_key}`;
+    params += ',vendor=elevenlabs';
+    params += `,voice=${voice}`;
     params += `,model_id=${model_id}`;
     params += `,optimize_streaming_latency=${opts.optimize_streaming_latency || 2}`;
+    params += ',write_cache_file=1';
     if (opts.voice_settings?.similarity_boost) params += `,similarity_boost=${opts.voice_settings.similarity_boost}`;
     if (opts.voice_settings?.stability) params += `,stability=${opts.voice_settings.stability}`;
     if (opts.voice_settings?.style) params += `,style=${opts.voice_settings.style}`;
@@ -620,7 +662,7 @@ const synthElevenlabs = async(logger, {credentials, options, stats, language, vo
     params += '}';
 
     return {
-      filePath: `say:${params}${text.replace(/\n/g, ' ')}`,
+      filePath: `say:${params}${text.replace(/\n/g, ' ').replace(/\r/g, ' ')}`,
       servedFromCache: false,
       rtt: 0
     };
@@ -651,8 +693,80 @@ const synthElevenlabs = async(logger, {credentials, options, stats, language, vo
   }
 };
 
-const synthWhisper = async(logger, {credentials, stats, voice, text}) => {
-  const {api_key, model_id, baseURL, timeout} = credentials;
+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 (!process.env.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 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) {
+    let params = '';
+    params += `{api_key=${api_key}`;
+    params += `,model_id=${model_id}`;
+    params += ',vendor=whisper';
+    params += `,voice=${voice}`;
+    params += ',write_cache_file=1';
+    if (speed) params += `,speed=${speed}`;
+    params += '}';
+
+    return {
+      filePath: `say:${params}${text.replace(/\n/g, ' ')}`,
+      servedFromCache: false,
+      rtt: 0
+    };
+  }
   try {
     const openai = new OpenAI.OpenAI({
       apiKey: api_key,
@@ -674,10 +788,24 @@ const synthWhisper = async(logger, {credentials, stats, voice, text}) => {
   }
 };
 
-const synthDeepgram = async(logger, {credentials, stats, model, text}) => {
+const synthDeepgram = async(logger, {credentials, stats, model, text, renderForCaching, disableTtsStreaming}) => {
   const {api_key} = credentials;
+  if (!process.env.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.beta.deepgram.com', 'POST', 'buffer', {
+    const post = bent('https://api.deepgram.com', 'POST', 'buffer', {
       'Authorization': `Token ${api_key}`,
       'Accept': 'audio/mpeg',
       'Content-Type': 'application/json'

package-lock.json

@@ -1,12 +1,12 @@
 {
   "name": "@jambonz/speech-utils",
-  "version": "0.0.34",
+  "version": "0.0.50",
   "lockfileVersion": 2,
   "requires": true,
   "packages": {
     "": {
       "name": "@jambonz/speech-utils",
-      "version": "0.0.34",
+      "version": "0.0.50",
       "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": "0.0.34",
+  "version": "0.0.50",
   "description": "TTS-related speech utilities for jambonz",
   "main": "index.js",
   "author": "Dave Horton",
@@ -34,7 +34,7 @@
     "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"
   },

test/synth.js

@@ -20,7 +20,7 @@ const stats = {
 
 test('Google speech synth tests', async(t) => {
   const fn = require('..');
-  const {synthAudio, client} = fn(opts, logger);
+  const {synthAudio, addFileToCache, client} = fn(opts, logger);
 
   if (!process.env.GCP_FILE && !process.env.GCP_JSON_KEY) {
     t.pass('skipping google speech synth tests since neither GCP_FILE nor GCP_JSON_KEY provided');
@@ -58,6 +58,14 @@ test('Google speech synth tests', async(t) => {
     });
     t.ok(opts.servedFromCache, `successfully retrieved cached google audio from ${opts.filePath}`);
 
+    const success = await addFileToCache(opts.filePath, {
+      vendor: 'google',
+      language: 'en-GB',
+      gender: 'FEMALE',
+      text: 'This is a test.  This is only a test'
+    });
+    t.ok(success, `successfully added ${opts.filePath} to cache`);
+
     opts = await synthAudio(stats, {
       vendor: 'google',
       credentials: {
@@ -180,6 +188,7 @@ 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) {
@@ -195,6 +204,7 @@ 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) {
@@ -204,6 +214,58 @@ test('Azure speech synth tests', async(t) => {
   client.quit();
 });
 
+test('Azure SSML tests', async(t) => {
+  const fn = require('..');
+  const {synthAudio, client} = fn(opts, logger);
+
+  if (!process.env.MICROSOFT_API_KEY || !process.env.MICROSOFT_REGION) {
+    t.pass('skipping Microsoft speech synth tests since MICROSOFT_API_KEY or MICROSOFT_REGION not provided');
+    return t.end();
+  }
+  try {
+    const text = `<speak version="1.0" xmlns="http://www.w3.org/2001/10/synthesis" xmlns:mstts="https://www.w3.org/2001/mstts" xml:lang="en-US">
+    <voice name="en-US-JennyMultilingualNeural">
+    <mstts:express-as style="cheerful" styledegree="2">That'd be just amazing!
+    </mstts:express-as>
+    </voice>
+    </speak>`;
+
+    let opts = await synthAudio(stats, {
+      vendor: 'microsoft',
+      credentials: {
+        api_key: process.env.MICROSOFT_API_KEY,
+        region: process.env.MICROSOFT_REGION,
+      },
+      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) {
+      t.pass('successfully used proxy to reach microsoft tts service');
+    }
+
+    opts = await synthAudio(stats, {
+      vendor: 'microsoft',
+      credentials: {
+        api_key: process.env.MICROSOFT_API_KEY,
+        region: process.env.MICROSOFT_REGION,
+      },
+      language: 'en-US',
+      voice: 'en-US-ChristopherNeural',
+      text,
+      renderForCaching: true
+    });
+    t.ok(opts.servedFromCache, `successfully retrieved microsoft audio from cache ${opts.filePath}`);
+  } catch (err) {
+    console.error(err);
+    t.end(err);
+  }
+  client.quit();
+});
+
+
 test('Azure custom voice speech synth tests', async(t) => {
   const fn = require('..');
   const {synthAudio, client} = fn(opts, logger);
@@ -225,6 +287,7 @@ 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}`);
 
@@ -239,6 +302,7 @@ 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) {
@@ -485,6 +549,47 @@ 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 synthesized eleven 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);
@@ -529,8 +634,9 @@ test('Deepgram speech synth tests', async(t) => {
       credentials: {
         api_key: process.env.DEEPGRAM_API_KEY
       },
-      model: 'alpha-asteria-en-v2',
+      model: 'aura-asteria-en',
       text,
+      renderForCaching: true
     });
     t.ok(!opts.servedFromCache, `successfully synthesized deepgram audio to ${opts.filePath}`);