0.2.27

fixed cartesia collect audio from stream

.github/workflows/ci.yml

@@ -11,7 +11,7 @@ jobs:
       - uses: actions/checkout@v4
       - uses: actions/setup-node@v4
         with:
-          node-version: lts/*
+          node-version: '20'
       - run: npm install
       - name: Install Docker Compose
         run: |

lib/synth-audio.js

@@ -53,7 +53,6 @@ const EXPIRES = JAMBONES_TTS_CACHE_DURATION_MINS;
 const OpenAI = require('openai');
 const getAwsAuthToken = require('./get-aws-sts-token');
 
-
 const trimTrailingSilence = (buffer) => {
   assert.ok(buffer instanceof Buffer, 'trimTrailingSilence - argument is not a Buffer');
 
@@ -205,74 +204,81 @@ async function synthAudio(client, createHash, retrieveHash, logger, stats, { acc
     const startAt = process.hrtime();
     switch (vendor) {
       case 'google':
-        audioData = await synthGoogle(logger, {credentials, stats, language, voice, gender, text});
+        audioData = await synthGoogle(logger, {credentials, stats, language, voice, gender, key, text});
         break;
       case 'aws':
       case 'polly':
         vendorLabel = 'aws';
         audioData = await synthPolly(createHash, retrieveHash, logger,
-          {credentials, stats, language, voice, text, engine});
+          {credentials, stats, language, voice, key, text, engine, renderForCaching, disableTtsStreaming,
+            disableTtsCache});
         break;
       case 'azure':
       case 'microsoft':
         vendorLabel = 'microsoft';
-        audioData = await synthMicrosoft(logger, {credentials, stats, language, voice, text, deploymentId,
+        audioData = await synthMicrosoft(logger, {credentials, stats, language, voice, key, text, deploymentId,
-          renderForCaching, disableTtsStreaming});
+          renderForCaching, disableTtsStreaming, disableTtsCache});
         break;
       case 'nuance':
         model = model || 'enhanced';
-        audioData = await synthNuance(client, logger, {credentials, stats, voice, model, text});
+        audioData = await synthNuance(client, logger, {credentials, stats, voice, model, key, text});
         break;
       case 'nvidia':
-        audioData = await synthNvidia(client, logger, {credentials, stats, language, voice, model, text,
+        audioData = await synthNvidia(client, logger, {credentials, stats, language, voice, model, key, text,
-          renderForCaching, disableTtsStreaming});
+          renderForCaching, disableTtsStreaming, disableTtsCache});
         break;
       case 'ibm':
-        audioData = await synthIbm(logger, {credentials, stats, voice, text});
+        audioData = await synthIbm(logger, {credentials, stats, voice, key, text});
         break;
       case 'wellsaid':
-        audioData = await synthWellSaid(logger, {credentials, stats, language, voice, text});
+        audioData = await synthWellSaid(logger, {credentials, stats, language, voice, key, text});
         break;
       case 'elevenlabs':
         audioData = await synthElevenlabs(logger, {
-          credentials, options, stats, language, voice, text, renderForCaching, disableTtsStreaming});
+          credentials, options, stats, language, voice, key, text, renderForCaching, disableTtsStreaming,
+          disableTtsCache});
         break;
       case 'playht':
         audioData = await synthPlayHT(client, logger, {
-          credentials, options, stats, language, voice, text, renderForCaching, disableTtsStreaming});
+          credentials, options, stats, language, voice, key, text, renderForCaching, disableTtsStreaming,
+          disableTtsCache});
         break;
       case 'cartesia':
         audioData = await synthCartesia(logger, {
-          credentials, options, stats, language, voice, text, renderForCaching, disableTtsStreaming});
+          credentials, options, stats, language, voice, key, text, renderForCaching, disableTtsStreaming,
+          disableTtsCache});
         break;
       case 'inworld':
         audioData = await synthInworld(logger, {
-          credentials, options, stats, language, voice, text, renderForCaching, disableTtsStreaming});
+          credentials, options, stats, language, voice, key, text, renderForCaching, disableTtsStreaming,
+          disableTtsCache});
         break;
       case 'rimelabs':
         audioData = await synthRimelabs(logger, {
-          credentials, options, stats, language, voice, text, renderForCaching, disableTtsStreaming});
+          credentials, options, stats, language, voice, key, text, renderForCaching, disableTtsStreaming,
+          disableTtsCache});
         break;
       case 'whisper':
         audioData = await synthWhisper(logger, {
-          credentials, stats, voice, text, instructions, renderForCaching, disableTtsStreaming});
+          credentials, stats, voice, key, text, instructions, renderForCaching, disableTtsStreaming,
+          disableTtsCache});
         break;
       case 'verbio':
         audioData = await synthVerbio(client, logger, {
-          credentials, stats, voice, text, renderForCaching, disableTtsStreaming});
+          credentials, stats, voice, key, text, renderForCaching, disableTtsStreaming, disableTtsCache});
         if (audioData?.filePath) return audioData;
         break;
       case 'deepgram':
-        audioData = await synthDeepgram(logger, {credentials, stats, model, text,
+        audioData = await synthDeepgram(logger, {credentials, stats, model, key, text,
-          renderForCaching, disableTtsStreaming});
+          renderForCaching, disableTtsStreaming, disableTtsCache});
         break;
       case 'resemble':
         audioData = await synthResemble(logger, {
-          credentials, stats, voice, text, options, renderForCaching, disableTtsStreaming});
+          credentials, stats, voice, key, text, options, renderForCaching, disableTtsStreaming, disableTtsCache});
         break;
       case vendor.startsWith('custom') ? vendor : 'cant_match_value':
         audioData = await synthCustomVendor(logger,
-          {credentials, stats, language, voice, text});
+          {credentials, stats, language, voice, key, text, renderForCaching, disableTtsStreaming, disableTtsCache});
         break;
       default:
         assert(`synthAudio: unsupported speech vendor ${vendor}`);
@@ -307,9 +313,43 @@ async function synthAudio(client, createHash, retrieveHash, logger, stats, { acc
 }
 
 const synthPolly = async(createHash, retrieveHash, logger,
-  {credentials, stats, language, voice, engine, text}) => {
+  {credentials, stats, language, voice, engine, key, text, renderForCaching, disableTtsStreaming, disableTtsCache}) => {
+  const {region, accessKeyId, secretAccessKey, roleArn} = credentials;
+  if (!JAMBONES_DISABLE_TTS_STREAMING && !renderForCaching && !disableTtsStreaming) {
+
+    let params = '{';
+    params += `language=${language}`;
+    params += `,playback_id=${key}`;
+    params += `,write_cache_file=${disableTtsCache ? 0 : 1}`;
+    params += ',vendor=aws';
+    if (accessKeyId && secretAccessKey) {
+      if (accessKeyId) params += `,accessKeyId=${accessKeyId}`;
+      if (secretAccessKey) params += `,secretAccessKey=${secretAccessKey}`;
+    } else if (roleArn) {
+      const cred = await getAwsAuthToken(
+        logger, createHash, retrieveHash,
+        {
+          region,
+          roleArn
+        });
+
+      if (cred) {
+        params += `,accessKeyId=${cred.accessKeyId}`;
+        params += `,secretAccessKey=${cred.secretAccessKey}`;
+        params += `,sessionToken=${cred.sessionToken}`;
+      }
+    }
+    if (region) params += `,region=${region}`;
+    if (engine) params += `,engine=${engine}`;
+    params += '}';
+
+    return {
+      filePath: `say:${params}${text.replace(/\n/g, ' ').replace(/\r/g, ' ')}`,
+      servedFromCache: false,
+      rtt: 0
+    };
+  }
   try {
-    const {region, accessKeyId, secretAccessKey, roleArn} = credentials;
     let polly;
     if (accessKeyId && secretAccessKey) {
       polly = new PollyClient({
@@ -527,9 +567,11 @@ const synthMicrosoft = async(logger, {
   stats,
   language,
   voice,
+  key,
   text,
   renderForCaching,
-  disableTtsStreaming
+  disableTtsStreaming,
+  disableTtsCache
 }) => {
   try {
     const {api_key: apiKey, region, use_custom_tts, custom_tts_endpoint, custom_tts_endpoint_url} = credentials;
@@ -556,12 +598,13 @@ const synthMicrosoft = async(logger, {
     }
     if (!JAMBONES_DISABLE_TTS_STREAMING && !JAMBONES_DISABLE_AZURE_TTS_STREAMING &&
       !renderForCaching && !disableTtsStreaming) {
-      let params = '';
+      let params = '{';
-      params += `{api_key=${apiKey}`;
+      params += `api_key=${apiKey}`;
+      params += `,playback_id=${key}`;
       params += `,language=${language}`;
       params += ',vendor=microsoft';
       params += `,voice=${voice}`;
-      params += ',write_cache_file=1';
+      params += `,write_cache_file=${disableTtsCache ? 0 : 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}`;
@@ -734,15 +777,16 @@ const synthNuance = async(client, logger, {credentials, stats, voice, model, tex
 };
 
 const synthNvidia = async(client, logger, {
-  credentials, stats, language,  voice, model, text, renderForCaching, disableTtsStreaming
+  credentials, stats, language,  voice, model, key, text, renderForCaching, disableTtsStreaming, disableTtsCache
 }) => {
   const {riva_server_uri} = credentials;
   if (!JAMBONES_DISABLE_TTS_STREAMING && !renderForCaching && !disableTtsStreaming) {
     let params = '';
     params += `{riva_server_uri=${riva_server_uri}`;
+    params += `,playback_id=${key}`;
     params += `,voice=${voice}`;
     params += `,language=${language}`;
-    params += ',write_cache_file=1';
+    params += `,write_cache_file=${disableTtsCache ? 0 : 1}`;
     params += '}';
 
     return {
@@ -782,9 +826,27 @@ const synthNvidia = async(client, logger, {
 };
 
 
-const synthCustomVendor = async(logger, {credentials, stats, language, voice, text, filePath}) => {
+const synthCustomVendor = async(logger, {credentials, stats, language, voice,
+  text, filePath, renderForCaching, disableTtsStreaming, key, disableTtsCache}) => {
   const {vendor, auth_token, custom_tts_url} = credentials;
 
+  if (!JAMBONES_DISABLE_TTS_STREAMING && !renderForCaching && !disableTtsStreaming) {
+    let params = '{';
+    params += `auth_token=${auth_token}`;
+    params += `,playback_id=${key}`;
+    params += `,custom_tts_url=${custom_tts_url}`;
+    params += ',vendor=custom';
+    params += `,voice=${voice}`;
+    params += `,write_cache_file=${disableTtsCache ? 0 : 1}`;
+    params += '}';
+
+    return {
+      filePath: `say:${params}${text.replace(/\n/g, ' ').replace(/\r/g, ' ')}`,
+      servedFromCache: false,
+      rtt: 0
+    };
+  }
+
   try {
     const post = bent('POST', {
       'Authorization': `Bearer ${auth_token}`,
@@ -813,20 +875,24 @@ const synthCustomVendor = async(logger, {credentials, stats, language, voice, te
 };
 
 const synthElevenlabs = async(logger, {
-  credentials, options, stats, voice, text, renderForCaching, disableTtsStreaming
+  credentials, options, stats, voice, key, text, renderForCaching, disableTtsStreaming, disableTtsCache
 }) => {
-  const {api_key, model_id, options: credOpts} = credentials;
+  const {api_key, model_id, api_uri, 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 = '';
+    let params = '{';
-    params += `{api_key=${api_key}`;
+    params += `api_key=${api_key}`;
+    params += `,playback_id=${key}`;
     params += ',vendor=elevenlabs';
     params += `,voice=${voice}`;
     params += `,model_id=${model_id}`;
-    params += `,optimize_streaming_latency=${opts.optimize_streaming_latency || 2}`;
+    params += `,write_cache_file=${disableTtsCache ? 0 : 1}`;
-    params += ',write_cache_file=1';
+    if (api_uri) params += `,api_uri=${api_uri}`;
+    if (opts.optimize_streaming_latency !== null && opts.optimize_streaming_latency !== undefined) {
+      params += `,optimize_streaming_latency=${opts.optimize_streaming_latency}`;
+    }
     if (opts.voice_settings?.similarity_boost) params += `,similarity_boost=${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}`;
@@ -849,7 +915,7 @@ const synthElevenlabs = async(logger, {
   const optimize_streaming_latency = opts.optimize_streaming_latency ?
     `?optimize_streaming_latency=${opts.optimize_streaming_latency}` : '';
   try {
-    const post = bent('https://api.elevenlabs.io', 'POST', 'buffer', {
+    const post = bent(`https://${api_uri}`, 'POST', 'buffer', {
       'xi-api-key': api_key,
       'Accept': 'audio/mpeg',
       'Content-Type': 'application/json'
@@ -876,7 +942,7 @@ const synthElevenlabs = async(logger, {
 };
 
 const synthPlayHT = async(client, logger, {
-  credentials, options, stats, voice, language, text, renderForCaching, disableTtsStreaming
+  credentials, options, stats, voice, language, key, text, renderForCaching, disableTtsStreaming, disableTtsCache
 }) => {
   const {api_key, user_id, voice_engine, playht_tts_uri, options: credOpts} = credentials;
   const opts = !!options && Object.keys(options).length !== 0 ? options : JSON.parse(credOpts || '{}');
@@ -910,14 +976,15 @@ const synthPlayHT = async(client, logger, {
 
   /* 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 = '';
+    let params = '{';
-    params += `{api_key=${api_key}`;
+    params += `api_key=${api_key}`;
+    params += `,playback_id=${key}`;
     params += `,user_id=${user_id}`;
     params += ',vendor=playht';
     params += `,voice=${voice}`;
     params += `,voice_engine=${voice_engine}`;
     params += `,synthesize_url=${synthesizeUrl}`;
-    params += ',write_cache_file=1';
+    params += `,write_cache_file=${disableTtsCache ? 0 : 1}`;
     params += `,language=${language}`;
     if (opts.quality) params += `,quality=${opts.quality}`;
     if (opts.speed) params += `,speed=${opts.speed}`;
@@ -970,19 +1037,20 @@ const synthPlayHT = async(client, logger, {
 };
 
 const synthInworld = async(logger, {
-  credentials, options, stats, voice, text, renderForCaching, disableTtsStreaming
+  credentials, options, stats, voice, key, text, renderForCaching, disableTtsStreaming, disableTtsCache
 }) => {
   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 = '';
+    let params = '{';
-    params += `{api_key=${api_key}`;
+    params += `api_key=${api_key}`;
+    params += `,playback_id=${key}`;
     params += `,model_id=${model_id}`;
     params += ',vendor=inworld';
     params += `,voice=${voice}`;
-    params += ',write_cache_file=1';
+    params += `,write_cache_file=${disableTtsCache ? 0 : 1}`;
     if (opts.temperature) params += `,temperature=${opts.temperature}`;
     if (opts.audioConfig?.pitch) params += `,pitch=${opts.pitch}`;
     if (opts.audioConfig?.speakingRate) params += `,speakingRate=${opts.speakingRate}`;
@@ -1034,20 +1102,21 @@ const synthInworld = async(logger, {
 };
 
 const synthRimelabs = async(logger, {
-  credentials, options, stats, language, voice, text, renderForCaching, disableTtsStreaming
+  credentials, options, stats, language, voice, key, text, renderForCaching, disableTtsStreaming, disableTtsCache
 }) => {
   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 = '';
+    let params = '{';
-    params += `{api_key=${api_key}`;
+    params += `api_key=${api_key}`;
+    params += `,playback_id=${key}`;
     params += `,model_id=${model_id}`;
     params += ',vendor=rimelabs';
     params += `,language=${language}`;
     params += `,voice=${voice}`;
-    params += ',write_cache_file=1';
+    params += `,write_cache_file=${disableTtsCache ? 0 : 1}`;
     if (opts.speedAlpha) params += `,speed_alpha=${opts.speedAlpha}`;
     if (opts.reduceLatency) params += `,reduce_latency=${opts.reduceLatency}`;
     // Arcana model parameters
@@ -1090,18 +1159,21 @@ const synthRimelabs = async(logger, {
     throw err;
   }
 };
-const synthVerbio = async(client, logger, {credentials, stats, voice, text, renderForCaching, disableTtsStreaming}) => {
+const synthVerbio = async(client, logger, {
+  credentials, stats, voice, key, text, renderForCaching, disableTtsStreaming, disableTtsCache
+}) => {
   //https://doc.speechcenter.verbio.com/#tag/Text-To-Speech-REST-API
   if (text.length > 2000) {
     throw new Error('Verbio cannot synthesize for the text length larger than 2000 characters');
   }
   const token = await getVerbioAccessToken(client, logger, credentials);
   if (!process.env.JAMBONES_DISABLE_TTS_STREAMING && !renderForCaching && !disableTtsStreaming) {
-    let params = '';
+    let params = '{';
-    params += `{access_token=${token.access_token}`;
+    params += `access_token=${token.access_token}`;
+    params += `,playback_id=${key}`;
     params += ',vendor=verbio';
     params += `,voice=${voice}`;
-    params += ',write_cache_file=1';
+    params += `,write_cache_file=${disableTtsCache ? 0 : 1}`;
     params += '}';
 
     return {
@@ -1135,17 +1207,18 @@ const synthVerbio = async(client, logger, {credentials, stats, voice, text, rend
   }
 };
 
-const synthWhisper = async(logger, {credentials, stats, voice, text, instructions,
+const synthWhisper = async(logger, {credentials, stats, voice, key, text, instructions,
-  renderForCaching, disableTtsStreaming}) => {
+  renderForCaching, disableTtsStreaming, disableTtsCache}) => {
   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 (!JAMBONES_DISABLE_TTS_STREAMING && !renderForCaching && !disableTtsStreaming) {
-    let params = '';
+    let params = '{';
-    params += `{api_key=${api_key}`;
+    params += `api_key=${api_key}`;
+    params += `,playback_id=${key}`;
     params += `,model_id=${model_id}`;
     params += ',vendor=whisper';
     params += `,voice=${voice}`;
-    params += ',write_cache_file=1';
+    params += `,write_cache_file=${disableTtsCache ? 0 : 1}`;
     if (speed) params += `,speed=${speed}`;
     // comma is used to separated parameters in freeswitch tts module
     if (instructions) params += `,instructions=${instructions.replace(/\n/g, ' ').replace(/,/g, ';')}`;
@@ -1183,14 +1256,16 @@ const synthWhisper = async(logger, {credentials, stats, voice, text, instruction
   }
 };
 
-const synthDeepgram = async(logger, {credentials, stats, model, text, renderForCaching, disableTtsStreaming}) => {
+const synthDeepgram = async(logger, {credentials, stats, model, key, text, renderForCaching,
+  disableTtsStreaming, disableTtsCache}) => {
   const {api_key, deepgram_tts_uri} = credentials;
   if (!JAMBONES_DISABLE_TTS_STREAMING && !renderForCaching && !disableTtsStreaming) {
-    let params = '';
+    let params = '{';
-    params += `{api_key=${api_key}`;
+    params += `api_key=${api_key}`;
+    params += `,playback_id=${key}`;
     params += ',vendor=deepgram';
     params += `,voice=${model}`;
-    params += ',write_cache_file=1';
+    params += `,write_cache_file=${disableTtsCache ? 0 : 1}`;
     if (deepgram_tts_uri) params += `,endpoint=${deepgram_tts_uri}`;
     params += '}';
 
@@ -1223,23 +1298,25 @@ const synthDeepgram = async(logger, {credentials, stats, model, text, renderForC
 };
 
 const synthCartesia = async(logger, {
-  credentials, options, stats, voice, language, text, renderForCaching, disableTtsStreaming
+  credentials, options, stats, voice, language, key, text, renderForCaching, disableTtsStreaming, disableTtsCache
 }) => {
   const {api_key, model_id, embedding, options: credOpts} = credentials;
   const opts = !!options && Object.keys(options).length !== 0 ? options : JSON.parse(credOpts || '{}');
 
   if (!JAMBONES_DISABLE_TTS_STREAMING && !renderForCaching && !disableTtsStreaming) {
-    let params = '';
+    let params = '{';
-    params += `{api_key=${api_key}`;
+    params += `api_key=${api_key}`;
+    params += `,playback_id=${key}`;
     params += `,model_id=${model_id}`;
     params += ',vendor=cartesia';
     params += `,voice=${voice}`;
-    params += ',write_cache_file=1';
+    params += `,write_cache_file=${disableTtsCache ? 0 : 1}`;
     params += `,language=${language}`;
     params += `,voice_mode=${embedding ? 'embedding' : 'id'}`;
     if (embedding) params += `,embedding=${embedding}`;
     if (opts.speed) params += `,speed=${opts.speed}`;
     if (opts.emotion) params += `,emotion=${opts.emotion}`;
+    if (opts.volume) params += `,volume=${opts.volume}`;
     params += '}';
 
     return {
@@ -1252,7 +1329,7 @@ const synthCartesia = async(logger, {
   try {
     const client = new CartesiaClient({ apiKey: api_key });
     const sampleRate = 48000;
-    const mp3 = await client.tts.bytes({
+    const mp3Stream = await client.tts.bytes({
       modelId: model_id,
       transcript: text,
       voice: {
@@ -1265,13 +1342,20 @@ const synthCartesia = async(logger, {
             id: voice
           }
         ),
-        ...(opts.speed || opts.emotion && {
+        ...(model_id === 'sonic-2' && (opts.speed || opts.emotion) && {
           experimentalControls: {
             ...(opts.speed !== null && opts.speed !== undefined && {speed: opts.speed}),
-            ...(opts.emotion && {emotion: opts.emotion}),
+            ...(opts.emotion && {emotion: [opts.emotion]}),
           }
-        })
+        }),
       },
+      ...(model_id === 'sonic-3' && (opts.speed || opts.emotion || opts.volume) && {
+        generationConfig: {
+          ...(opts.volume !== null && opts.volume !== undefined && {volume: opts.volume}),
+          ...(opts.speed !== null && opts.speed !== undefined && {speed: opts.speed}),
+          ...(opts.emotion !== null && opts.emotion !== undefined && {emotion: opts.emotion}),
+        }
+      }),
       language: language,
       outputFormat: {
         container: 'mp3',
@@ -1279,8 +1363,16 @@ const synthCartesia = async(logger, {
         sampleRate
       },
     });
+
+    // bytes() returns a ReadableStream - collect all chunks
+    const chunks = [];
+    for await (const chunk of mp3Stream) {
+      chunks.push(chunk);
+    }
+    const audioBuffer = Buffer.concat(chunks);
+
     return {
-      audioContent: Buffer.from(mp3),
+      audioContent: audioBuffer,
       extension: 'mp3',
       sampleRate
     };
@@ -1293,21 +1385,23 @@ const synthCartesia = async(logger, {
 };
 
 const synthResemble = async(logger, {
-  credentials, options, stats, voice, text, renderForCaching, disableTtsStreaming
+  credentials, options, stats, voice, key, text, renderForCaching, disableTtsStreaming, disableTtsCache
 }) => {
-  const {api_key, resemble_tts_uri} = credentials;
+  const {api_key, resemble_tts_uri, resemble_tts_use_tls} = credentials;
   const {project_uuid, use_hd} = options || {};
 
   /* 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 = '';
+    let params = '{';
-    params += `{api_key=${api_key}`;
+    params += `api_key=${api_key}`;
+    params += `,playback_id=${key}`;
     params += ',vendor=resemble';
     params += `,voice=${voice}`;
-    params += ',write_cache_file=1';
+    params += `,write_cache_file=${disableTtsCache ? 0 : 1}`;
     if (project_uuid) params += `,project_uuid=${project_uuid}`;
     if (use_hd) params += `,use_hd=${use_hd}`;
     if (resemble_tts_uri) params += `,endpoint=${resemble_tts_uri}`;
+    if (resemble_tts_use_tls) params += `,use_tls=${resemble_tts_use_tls}`;
 
     params += '}';
 

package-lock.json

@@ -1,18 +1,18 @@
 {
   "name": "@jambonz/speech-utils",
-  "version": "0.2.16",
+  "version": "0.2.27",
   "lockfileVersion": 2,
   "requires": true,
   "packages": {
     "": {
       "name": "@jambonz/speech-utils",
-      "version": "0.2.16",
+      "version": "0.2.27",
       "license": "MIT",
       "dependencies": {
         "23": "^0.0.0",
         "@aws-sdk/client-polly": "^3.496.0",
         "@aws-sdk/client-sts": "^3.496.0",
-        "@cartesia/cartesia-js": "^2.1.0",
+        "@cartesia/cartesia-js": "^2.2.7",
         "@google-cloud/text-to-speech": "^5.5.0",
         "@grpc/grpc-js": "^1.9.14",
         "@jambonz/realtimedb-helpers": "^0.8.7",
@@ -1046,17 +1046,17 @@
       }
     },
     "node_modules/@cartesia/cartesia-js": {
-      "version": "2.1.0",
+      "version": "2.2.7",
-      "resolved": "https://registry.npmjs.org/@cartesia/cartesia-js/-/cartesia-js-2.1.0.tgz",
+      "resolved": "https://registry.npmjs.org/@cartesia/cartesia-js/-/cartesia-js-2.2.7.tgz",
-      "integrity": "sha512-r4uqhFWxVmFbm0A6aDeY0y3RVJHAc6PQ8hTwHmb2ZO2kmyAP4WoZdEEQ9GIU/5c1XpV5n3Y/y6zKjMzUwgG64A==",
+      "integrity": "sha512-yXIK/rEQzY3LPdrFd+oWlW39dxuAb5mnP3NOemaeLiMaRKAYatMdh154NyB3lAf3BgqMXpGWUg46vSzBCDiEnQ==",
       "dependencies": {
         "emittery": "^0.13.1",
         "form-data": "^4.0.0",
         "form-data-encoder": "^4.0.2",
         "formdata-node": "^6.0.3",
         "human-id": "^4.1.1",
-        "node-fetch": "2.7.0",
+        "node-fetch": "^2.7.0",
-        "qs": "6.11.2",
+        "qs": "^6.13.1",
         "readable-stream": "^4.5.2",
         "url-join": "4.0.1",
         "ws": "^8.15.13"
@@ -2619,6 +2619,7 @@
       "version": "1.0.7",
       "resolved": "https://registry.npmjs.org/call-bind/-/call-bind-1.0.7.tgz",
       "integrity": "sha512-GHTSNSYICQ7scH7sZ+M2rFopRoLh8t2bLSW6BbgrtLsahOIB5iyAVJf9GjWK3cYTDaMj4XdBpM1cA6pIS0Kv2w==",
+      "dev": true,
       "dependencies": {
         "es-define-property": "^1.0.0",
         "es-errors": "^1.3.0",
@@ -2645,6 +2646,22 @@
         "node": ">= 0.4"
       }
     },
+    "node_modules/call-bound": {
+      "version": "1.0.4",
+      "resolved": "https://registry.npmjs.org/call-bound/-/call-bound-1.0.4.tgz",
+      "integrity": "sha512-+ys997U96po4Kx/ABpBCqhA9EuxJaQWDQg7295H4hBphv3IZg0boBKuwYpt4YXp6MZ5AmZQnU/tyMTlRpaSejg==",
+      "license": "MIT",
+      "dependencies": {
+        "call-bind-apply-helpers": "^1.0.2",
+        "get-intrinsic": "^1.3.0"
+      },
+      "engines": {
+        "node": ">= 0.4"
+      },
+      "funding": {
+        "url": "https://github.com/sponsors/ljharb"
+      }
+    },
     "node_modules/callsites": {
       "version": "3.1.0",
       "resolved": "https://registry.npmjs.org/callsites/-/callsites-3.1.0.tgz",
@@ -2899,6 +2916,7 @@
       "version": "1.1.4",
       "resolved": "https://registry.npmjs.org/define-data-property/-/define-data-property-1.1.4.tgz",
       "integrity": "sha512-rBMvIzlpA8v6E+SJZoo++HAYqsLrkg7MSfIinMPFhmkorw7X+dOXVJQs+QT69zGkzMyfDnIMN2Wid1+NbL3T+A==",
+      "dev": true,
       "dependencies": {
         "es-define-property": "^1.0.0",
         "es-errors": "^1.3.0",
@@ -4036,6 +4054,7 @@
       "version": "1.0.2",
       "resolved": "https://registry.npmjs.org/has-property-descriptors/-/has-property-descriptors-1.0.2.tgz",
       "integrity": "sha512-55JNKuIW+vq4Ke1BjOTjM2YctQIvCT7GFzHwmfZPGo5wnrgkid0YQtnAleFSqumZm4az3n2BS+erby5ipJdgrg==",
+      "dev": true,
       "dependencies": {
         "es-define-property": "^1.0.0"
       },
@@ -5492,9 +5511,13 @@
       }
     },
     "node_modules/object-inspect": {
-      "version": "1.13.1",
+      "version": "1.13.4",
-      "resolved": "https://registry.npmjs.org/object-inspect/-/object-inspect-1.13.1.tgz",
+      "resolved": "https://registry.npmjs.org/object-inspect/-/object-inspect-1.13.4.tgz",
-      "integrity": "sha512-5qoj1RUiKOMsCCNLV1CBiPYE10sziTsnmNxkAI/rZhiD63CF7IqdFGC/XzjWjpSgLf0LxXX3bDFIh0E18f6UhQ==",
+      "integrity": "sha512-W67iLl4J2EXEGTbfeHCffrjDfitvLANg0UlX3wFUUSTx92KXRFegMHUVgSqE+wvhAbi4WqjGg9czysTV2Epbew==",
+      "license": "MIT",
+      "engines": {
+        "node": ">= 0.4"
+      },
       "funding": {
         "url": "https://github.com/sponsors/ljharb"
       }
@@ -5955,12 +5978,12 @@
       }
     },
     "node_modules/qs": {
-      "version": "6.11.2",
+      "version": "6.14.0",
-      "resolved": "https://registry.npmjs.org/qs/-/qs-6.11.2.tgz",
+      "resolved": "https://registry.npmjs.org/qs/-/qs-6.14.0.tgz",
-      "integrity": "sha512-tDNIz22aBzCDxLtVH++VnTfzxlfeK5CbqohpSqpJgj1Wg/cQbStNAz3NuqCs5vV+pjBsK4x4pN9HlVh7rcYRiA==",
+      "integrity": "sha512-YWWTjgABSKcvs/nWBi9PycY/JiPJqOD4JA6o9Sej2AtvSGarXxKC3OQSk4pAarbdQlKAh5D4FCQkJNkW+GAn3w==",
       "license": "BSD-3-Clause",
       "dependencies": {
-        "side-channel": "^1.0.4"
+        "side-channel": "^1.1.0"
       },
       "engines": {
         "node": ">=0.6"
@@ -6301,6 +6324,7 @@
       "version": "1.2.2",
       "resolved": "https://registry.npmjs.org/set-function-length/-/set-function-length-1.2.2.tgz",
       "integrity": "sha512-pgRc4hJ4/sNjWCSS9AmnS40x3bNMDTknHgL5UaMBTMyJnU90EgWh1Rz+MC9eFu4BuN/UwZjKQuY/1v3rM7HMfg==",
+      "dev": true,
       "dependencies": {
         "define-data-property": "^1.1.4",
         "es-errors": "^1.3.0",
@@ -6349,13 +6373,72 @@
       }
     },
     "node_modules/side-channel": {
-      "version": "1.0.4",
+      "version": "1.1.0",
-      "resolved": "https://registry.npmjs.org/side-channel/-/side-channel-1.0.4.tgz",
+      "resolved": "https://registry.npmjs.org/side-channel/-/side-channel-1.1.0.tgz",
-      "integrity": "sha512-q5XPytqFEIKHkGdiMIrY10mvLRvnQh42/+GoBlFW3b2LXLE2xxJpZFdm94we0BaoV3RwJyGqg5wS7epxTv0Zvw==",
+      "integrity": "sha512-ZX99e6tRweoUXqR+VBrslhda51Nh5MTQwou5tnUDgbtyM0dBgmhEDtWGP/xbKn6hqfPRHujUNwz5fy/wbbhnpw==",
+      "license": "MIT",
       "dependencies": {
-        "call-bind": "^1.0.0",
+        "es-errors": "^1.3.0",
-        "get-intrinsic": "^1.0.2",
+        "object-inspect": "^1.13.3",
-        "object-inspect": "^1.9.0"
+        "side-channel-list": "^1.0.0",
+        "side-channel-map": "^1.0.1",
+        "side-channel-weakmap": "^1.0.2"
+      },
+      "engines": {
+        "node": ">= 0.4"
+      },
+      "funding": {
+        "url": "https://github.com/sponsors/ljharb"
+      }
+    },
+    "node_modules/side-channel-list": {
+      "version": "1.0.0",
+      "resolved": "https://registry.npmjs.org/side-channel-list/-/side-channel-list-1.0.0.tgz",
+      "integrity": "sha512-FCLHtRD/gnpCiCHEiJLOwdmFP+wzCmDEkc9y7NsYxeF4u7Btsn1ZuwgwJGxImImHicJArLP4R0yX4c2KCrMrTA==",
+      "license": "MIT",
+      "dependencies": {
+        "es-errors": "^1.3.0",
+        "object-inspect": "^1.13.3"
+      },
+      "engines": {
+        "node": ">= 0.4"
+      },
+      "funding": {
+        "url": "https://github.com/sponsors/ljharb"
+      }
+    },
+    "node_modules/side-channel-map": {
+      "version": "1.0.1",
+      "resolved": "https://registry.npmjs.org/side-channel-map/-/side-channel-map-1.0.1.tgz",
+      "integrity": "sha512-VCjCNfgMsby3tTdo02nbjtM/ewra6jPHmpThenkTYh8pG9ucZ/1P8So4u4FGBek/BjpOVsDCMoLA/iuBKIFXRA==",
+      "license": "MIT",
+      "dependencies": {
+        "call-bound": "^1.0.2",
+        "es-errors": "^1.3.0",
+        "get-intrinsic": "^1.2.5",
+        "object-inspect": "^1.13.3"
+      },
+      "engines": {
+        "node": ">= 0.4"
+      },
+      "funding": {
+        "url": "https://github.com/sponsors/ljharb"
+      }
+    },
+    "node_modules/side-channel-weakmap": {
+      "version": "1.0.2",
+      "resolved": "https://registry.npmjs.org/side-channel-weakmap/-/side-channel-weakmap-1.0.2.tgz",
+      "integrity": "sha512-WPS/HvHQTYnHisLo9McqBHOJk2FkHO/tlpvldyrnem4aeQp4hai3gythswg6p01oSoTl58rcpiFAjF2br2Ak2A==",
+      "license": "MIT",
+      "dependencies": {
+        "call-bound": "^1.0.2",
+        "es-errors": "^1.3.0",
+        "get-intrinsic": "^1.2.5",
+        "object-inspect": "^1.13.3",
+        "side-channel-map": "^1.0.1"
+      },
+      "engines": {
+        "node": ">= 0.4"
       },
       "funding": {
         "url": "https://github.com/sponsors/ljharb"
@@ -7969,17 +8052,17 @@
       }
     },
     "@cartesia/cartesia-js": {
-      "version": "2.1.0",
+      "version": "2.2.7",
-      "resolved": "https://registry.npmjs.org/@cartesia/cartesia-js/-/cartesia-js-2.1.0.tgz",
+      "resolved": "https://registry.npmjs.org/@cartesia/cartesia-js/-/cartesia-js-2.2.7.tgz",
-      "integrity": "sha512-r4uqhFWxVmFbm0A6aDeY0y3RVJHAc6PQ8hTwHmb2ZO2kmyAP4WoZdEEQ9GIU/5c1XpV5n3Y/y6zKjMzUwgG64A==",
+      "integrity": "sha512-yXIK/rEQzY3LPdrFd+oWlW39dxuAb5mnP3NOemaeLiMaRKAYatMdh154NyB3lAf3BgqMXpGWUg46vSzBCDiEnQ==",
       "requires": {
         "emittery": "^0.13.1",
         "form-data": "^4.0.0",
         "form-data-encoder": "^4.0.2",
         "formdata-node": "^6.0.3",
         "human-id": "^4.1.1",
-        "node-fetch": "2.7.0",
+        "node-fetch": "^2.7.0",
-        "qs": "6.11.2",
+        "qs": "^6.13.1",
         "readable-stream": "^4.5.2",
         "url-join": "4.0.1",
         "ws": "^8.15.13"
@@ -9147,6 +9230,7 @@
       "version": "1.0.7",
       "resolved": "https://registry.npmjs.org/call-bind/-/call-bind-1.0.7.tgz",
       "integrity": "sha512-GHTSNSYICQ7scH7sZ+M2rFopRoLh8t2bLSW6BbgrtLsahOIB5iyAVJf9GjWK3cYTDaMj4XdBpM1cA6pIS0Kv2w==",
+      "dev": true,
       "requires": {
         "es-define-property": "^1.0.0",
         "es-errors": "^1.3.0",
@@ -9164,6 +9248,15 @@
         "function-bind": "^1.1.2"
       }
     },
+    "call-bound": {
+      "version": "1.0.4",
+      "resolved": "https://registry.npmjs.org/call-bound/-/call-bound-1.0.4.tgz",
+      "integrity": "sha512-+ys997U96po4Kx/ABpBCqhA9EuxJaQWDQg7295H4hBphv3IZg0boBKuwYpt4YXp6MZ5AmZQnU/tyMTlRpaSejg==",
+      "requires": {
+        "call-bind-apply-helpers": "^1.0.2",
+        "get-intrinsic": "^1.3.0"
+      }
+    },
     "callsites": {
       "version": "3.1.0",
       "resolved": "https://registry.npmjs.org/callsites/-/callsites-3.1.0.tgz",
@@ -9344,6 +9437,7 @@
       "version": "1.1.4",
       "resolved": "https://registry.npmjs.org/define-data-property/-/define-data-property-1.1.4.tgz",
       "integrity": "sha512-rBMvIzlpA8v6E+SJZoo++HAYqsLrkg7MSfIinMPFhmkorw7X+dOXVJQs+QT69zGkzMyfDnIMN2Wid1+NbL3T+A==",
+      "dev": true,
       "requires": {
         "es-define-property": "^1.0.0",
         "es-errors": "^1.3.0",
@@ -10165,6 +10259,7 @@
       "version": "1.0.2",
       "resolved": "https://registry.npmjs.org/has-property-descriptors/-/has-property-descriptors-1.0.2.tgz",
       "integrity": "sha512-55JNKuIW+vq4Ke1BjOTjM2YctQIvCT7GFzHwmfZPGo5wnrgkid0YQtnAleFSqumZm4az3n2BS+erby5ipJdgrg==",
+      "dev": true,
       "requires": {
         "es-define-property": "^1.0.0"
       }
@@ -11240,9 +11335,9 @@
       "integrity": "sha512-RSn9F68PjH9HqtltsSnqYC1XXoWe9Bju5+213R98cNGttag9q9yAOTzdbsqvIa7aNm5WffBZFpWYr2aWrklWAw=="
     },
     "object-inspect": {
-      "version": "1.13.1",
+      "version": "1.13.4",
-      "resolved": "https://registry.npmjs.org/object-inspect/-/object-inspect-1.13.1.tgz",
+      "resolved": "https://registry.npmjs.org/object-inspect/-/object-inspect-1.13.4.tgz",
-      "integrity": "sha512-5qoj1RUiKOMsCCNLV1CBiPYE10sziTsnmNxkAI/rZhiD63CF7IqdFGC/XzjWjpSgLf0LxXX3bDFIh0E18f6UhQ=="
+      "integrity": "sha512-W67iLl4J2EXEGTbfeHCffrjDfitvLANg0UlX3wFUUSTx92KXRFegMHUVgSqE+wvhAbi4WqjGg9czysTV2Epbew=="
     },
     "object-is": {
       "version": "1.1.5",
@@ -11583,11 +11678,11 @@
       "integrity": "sha512-vYt7UD1U9Wg6138shLtLOvdAu+8DsC/ilFtEVHcH+wydcSpNE20AfSOduf6MkRFahL5FY7X1oU7nKVZFtfq8Fg=="
     },
     "qs": {
-      "version": "6.11.2",
+      "version": "6.14.0",
-      "resolved": "https://registry.npmjs.org/qs/-/qs-6.11.2.tgz",
+      "resolved": "https://registry.npmjs.org/qs/-/qs-6.14.0.tgz",
-      "integrity": "sha512-tDNIz22aBzCDxLtVH++VnTfzxlfeK5CbqohpSqpJgj1Wg/cQbStNAz3NuqCs5vV+pjBsK4x4pN9HlVh7rcYRiA==",
+      "integrity": "sha512-YWWTjgABSKcvs/nWBi9PycY/JiPJqOD4JA6o9Sej2AtvSGarXxKC3OQSk4pAarbdQlKAh5D4FCQkJNkW+GAn3w==",
       "requires": {
-        "side-channel": "^1.0.4"
+        "side-channel": "^1.1.0"
       }
     },
     "querystringify": {
@@ -11803,6 +11898,7 @@
       "version": "1.2.2",
       "resolved": "https://registry.npmjs.org/set-function-length/-/set-function-length-1.2.2.tgz",
       "integrity": "sha512-pgRc4hJ4/sNjWCSS9AmnS40x3bNMDTknHgL5UaMBTMyJnU90EgWh1Rz+MC9eFu4BuN/UwZjKQuY/1v3rM7HMfg==",
+      "dev": true,
       "requires": {
         "define-data-property": "^1.1.4",
         "es-errors": "^1.3.0",
@@ -11839,13 +11935,47 @@
       "dev": true
     },
     "side-channel": {
-      "version": "1.0.4",
+      "version": "1.1.0",
-      "resolved": "https://registry.npmjs.org/side-channel/-/side-channel-1.0.4.tgz",
+      "resolved": "https://registry.npmjs.org/side-channel/-/side-channel-1.1.0.tgz",
-      "integrity": "sha512-q5XPytqFEIKHkGdiMIrY10mvLRvnQh42/+GoBlFW3b2LXLE2xxJpZFdm94we0BaoV3RwJyGqg5wS7epxTv0Zvw==",
+      "integrity": "sha512-ZX99e6tRweoUXqR+VBrslhda51Nh5MTQwou5tnUDgbtyM0dBgmhEDtWGP/xbKn6hqfPRHujUNwz5fy/wbbhnpw==",
       "requires": {
-        "call-bind": "^1.0.0",
+        "es-errors": "^1.3.0",
-        "get-intrinsic": "^1.0.2",
+        "object-inspect": "^1.13.3",
-        "object-inspect": "^1.9.0"
+        "side-channel-list": "^1.0.0",
+        "side-channel-map": "^1.0.1",
+        "side-channel-weakmap": "^1.0.2"
+      }
+    },
+    "side-channel-list": {
+      "version": "1.0.0",
+      "resolved": "https://registry.npmjs.org/side-channel-list/-/side-channel-list-1.0.0.tgz",
+      "integrity": "sha512-FCLHtRD/gnpCiCHEiJLOwdmFP+wzCmDEkc9y7NsYxeF4u7Btsn1ZuwgwJGxImImHicJArLP4R0yX4c2KCrMrTA==",
+      "requires": {
+        "es-errors": "^1.3.0",
+        "object-inspect": "^1.13.3"
+      }
+    },
+    "side-channel-map": {
+      "version": "1.0.1",
+      "resolved": "https://registry.npmjs.org/side-channel-map/-/side-channel-map-1.0.1.tgz",
+      "integrity": "sha512-VCjCNfgMsby3tTdo02nbjtM/ewra6jPHmpThenkTYh8pG9ucZ/1P8So4u4FGBek/BjpOVsDCMoLA/iuBKIFXRA==",
+      "requires": {
+        "call-bound": "^1.0.2",
+        "es-errors": "^1.3.0",
+        "get-intrinsic": "^1.2.5",
+        "object-inspect": "^1.13.3"
+      }
+    },
+    "side-channel-weakmap": {
+      "version": "1.0.2",
+      "resolved": "https://registry.npmjs.org/side-channel-weakmap/-/side-channel-weakmap-1.0.2.tgz",
+      "integrity": "sha512-WPS/HvHQTYnHisLo9McqBHOJk2FkHO/tlpvldyrnem4aeQp4hai3gythswg6p01oSoTl58rcpiFAjF2br2Ak2A==",
+      "requires": {
+        "call-bound": "^1.0.2",
+        "es-errors": "^1.3.0",
+        "get-intrinsic": "^1.2.5",
+        "object-inspect": "^1.13.3",
+        "side-channel-map": "^1.0.1"
       }
     },
     "signal-exit": {

package.json

@@ -1,6 +1,6 @@
 {
   "name": "@jambonz/speech-utils",
-  "version": "0.2.16",
+  "version": "0.2.27",
   "description": "TTS-related speech utilities for jambonz",
   "main": "index.js",
   "author": "Dave Horton",
@@ -29,7 +29,7 @@
     "23": "^0.0.0",
     "@aws-sdk/client-polly": "^3.496.0",
     "@aws-sdk/client-sts": "^3.496.0",
-    "@cartesia/cartesia-js": "^2.1.0",
+    "@cartesia/cartesia-js": "^2.2.7",
     "@google-cloud/text-to-speech": "^5.5.0",
     "@grpc/grpc-js": "^1.9.14",
     "@jambonz/realtimedb-helpers": "^0.8.7",

test/synth.js

@@ -185,6 +185,7 @@ test('AWS speech synth tests', async(t) => {
       language: 'en-US',
       voice: 'Joey',
       text: 'This is a test.  This is only a test',
+      renderForCaching: true,
     });
     t.ok(!opts.servedFromCache, `successfully synthesized aws audio to ${opts.filePath}`);
 
@@ -198,6 +199,7 @@ test('AWS speech synth tests', async(t) => {
       language: 'en-US',
       voice: 'Joey',
       text: 'This is a test.  This is only a test',
+      renderForCaching: true,
     });
     t.ok(opts.servedFromCache, `successfully retrieved aws audio from cache ${opts.filePath}`);
   } catch (err) {
@@ -552,6 +554,7 @@ test('Custom Vendor speech synth tests', async(t) => {
       language: 'en-US',
       voice: 'English-US.Female-1',
       text: 'This is a test.  This is only a test',
+      renderForCaching: true
     });
     t.ok(!opts.servedFromCache, `successfully synthesized custom vendor audio to ${opts.filePath}`);
     t.ok(opts.filePath.endsWith('wav'), 'audio is cached as wav file');
@@ -573,6 +576,7 @@ test('Custom Vendor speech synth tests', async(t) => {
       language: 'en-US',
       voice: 'English-US.Female-1',
       text: 'This is a test.  This is only a test',
+      renderForCaching: true
     });
     t.ok(opts.servedFromCache, `successfully get custom vendor cached audio to ${opts.filePath}`);
     t.ok(opts.filePath.endsWith('wav'), 'audio is cached as wav file');
@@ -587,6 +591,7 @@ test('Custom Vendor speech synth tests', async(t) => {
       language: 'en-US',
       voice: 'English-US.Female-1',
       text: '<speak>This is a test.  This is only a test</speak>',
+      renderForCaching: true
     });
     t.ok(!opts.servedFromCache, `successfully synthesized Custom Vendor audio to ${opts.filePath}`);
     obj = await getJSON(`http://127.0.0.1:3100/lastRequest/somethingnew2`);
@@ -711,7 +716,7 @@ test('Cartesia speech synth tests', async(t) => {
       text,
       renderForCaching: true
     });
-    t.ok(!opts.servedFromCache, `successfully playht eleven audio to ${opts.filePath}`);
+    t.ok(!opts.servedFromCache, `successfully cartesia eleven audio to ${opts.filePath}`);
 
   } catch (err) {
     console.error(JSON.stringify(err));