@jambonz/speech-utils 0.2.21 → 0.2.23

package/lib/synth-audio.js

@@ -53,9 +53,6 @@ const EXPIRES = JAMBONES_TTS_CACHE_DURATION_MINS;
 const OpenAI = require('openai');
 const getAwsAuthToken = require('./get-aws-sts-token');
 
-/* counter for playback id */
-let playbackIdCounter = 0;
-
 const trimTrailingSilence = (buffer) => {
   assert.ok(buffer instanceof Buffer, 'trimTrailingSilence - argument is not a Buffer');
 
@@ -207,74 +204,74 @@ 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, renderForCaching, disableTtsStreaming});
+          {credentials, stats, language, voice, key, text, engine, renderForCaching, disableTtsStreaming});
         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});
         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});
         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});
         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});
         break;
       case 'cartesia':
         audioData = await synthCartesia(logger, {
-          credentials, options, stats, language, voice, text, renderForCaching, disableTtsStreaming});
+          credentials, options, stats, language, voice, key, text, renderForCaching, disableTtsStreaming});
         break;
       case 'inworld':
         audioData = await synthInworld(logger, {
-          credentials, options, stats, language, voice, text, renderForCaching, disableTtsStreaming});
+          credentials, options, stats, language, voice, key, text, renderForCaching, disableTtsStreaming});
         break;
       case 'rimelabs':
         audioData = await synthRimelabs(logger, {
-          credentials, options, stats, language, voice, text, renderForCaching, disableTtsStreaming});
+          credentials, options, stats, language, voice, key, text, renderForCaching, disableTtsStreaming});
         break;
       case 'whisper':
         audioData = await synthWhisper(logger, {
-          credentials, stats, voice, text, instructions, renderForCaching, disableTtsStreaming});
+          credentials, stats, voice, key, text, instructions, renderForCaching, disableTtsStreaming});
         break;
       case 'verbio':
         audioData = await synthVerbio(client, logger, {
-          credentials, stats, voice, text, renderForCaching, disableTtsStreaming});
+          credentials, stats, voice, key, text, renderForCaching, disableTtsStreaming});
         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});
         break;
       case 'resemble':
         audioData = await synthResemble(logger, {
-          credentials, stats, voice, text, options, renderForCaching, disableTtsStreaming});
+          credentials, stats, voice, key, text, options, renderForCaching, disableTtsStreaming});
         break;
       case vendor.startsWith('custom') ? vendor : 'cant_match_value':
         audioData = await synthCustomVendor(logger,
-          {credentials, stats, language, voice, text});
+          {credentials, stats, language, voice, key, text});
         break;
       default:
         assert(`synthAudio: unsupported speech vendor ${vendor}`);
@@ -309,13 +306,13 @@ async function synthAudio(client, createHash, retrieveHash, logger, stats, { acc
 }
 
 const synthPolly = async(createHash, retrieveHash, logger,
-  {credentials, stats, language, voice, engine, text, renderForCaching, disableTtsStreaming}) => {
+  {credentials, stats, language, voice, engine, key, text, renderForCaching, disableTtsStreaming}) => {
   const {region, accessKeyId, secretAccessKey, roleArn} = credentials;
   if (!JAMBONES_DISABLE_TTS_STREAMING && !renderForCaching && !disableTtsStreaming) {
 
     let params = '{';
     params += `language=${language}`;
-    params += `,playback_id=${playbackIdCounter++}`;
+    params += `,playback_id=${key}`;
     params += ',write_cache_file=1';
     params += ',vendor=aws';
     if (accessKeyId && secretAccessKey) {
@@ -563,6 +560,7 @@ const synthMicrosoft = async(logger, {
   stats,
   language,
   voice,
+  key,
   text,
   renderForCaching,
   disableTtsStreaming
@@ -594,7 +592,7 @@ const synthMicrosoft = async(logger, {
       !renderForCaching && !disableTtsStreaming) {
       let params = '{';
       params += `api_key=${apiKey}`;
-      params += `,playback_id=${playbackIdCounter++}`;
+      params += `,playback_id=${key}`;
       params += `,language=${language}`;
       params += ',vendor=microsoft';
       params += `,voice=${voice}`;
@@ -771,13 +769,13 @@ 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
 }) => {
   const {riva_server_uri} = credentials;
   if (!JAMBONES_DISABLE_TTS_STREAMING && !renderForCaching && !disableTtsStreaming) {
     let params = '';
     params += `{riva_server_uri=${riva_server_uri}`;
-    params += `,playback_id=${playbackIdCounter++}`;
+    params += `,playback_id=${key}`;
     params += `,voice=${voice}`;
     params += `,language=${language}`;
     params += ',write_cache_file=1';
@@ -851,7 +849,7 @@ 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
 }) => {
   const {api_key, model_id, options: credOpts} = credentials;
   const opts = !!options && Object.keys(options).length !== 0 ? options : JSON.parse(credOpts || '{}');
@@ -860,12 +858,14 @@ const synthElevenlabs = async(logger, {
   if (!JAMBONES_DISABLE_TTS_STREAMING && !renderForCaching && !disableTtsStreaming) {
     let params = '{';
     params += `api_key=${api_key}`;
-    params += `,playback_id=${playbackIdCounter++}`;
+    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=1';
+    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}`;
@@ -915,7 +915,7 @@ 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
 }) => {
   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 || '{}');
@@ -951,7 +951,7 @@ const synthPlayHT = async(client, logger, {
   if (!JAMBONES_DISABLE_TTS_STREAMING && !renderForCaching && !disableTtsStreaming) {
     let params = '{';
     params += `api_key=${api_key}`;
-    params += `,playback_id=${playbackIdCounter++}`;
+    params += `,playback_id=${key}`;
     params += `,user_id=${user_id}`;
     params += ',vendor=playht';
     params += `,voice=${voice}`;
@@ -1010,7 +1010,7 @@ const synthPlayHT = async(client, logger, {
 };
 
 const synthInworld = async(logger, {
-  credentials, options, stats, voice, text, renderForCaching, disableTtsStreaming
+  credentials, options, stats, voice, key, text, renderForCaching, disableTtsStreaming
 }) => {
   const {api_key, model_id, options: credOpts} = credentials;
   const opts = !!options && Object.keys(options).length !== 0 ? options : JSON.parse(credOpts || '{}');
@@ -1019,7 +1019,7 @@ const synthInworld = async(logger, {
   if (!JAMBONES_DISABLE_TTS_STREAMING && !renderForCaching && !disableTtsStreaming) {
     let params = '{';
     params += `api_key=${api_key}`;
-    params += `,playback_id=${playbackIdCounter++}`;
+    params += `,playback_id=${key}`;
     params += `,model_id=${model_id}`;
     params += ',vendor=inworld';
     params += `,voice=${voice}`;
@@ -1075,7 +1075,7 @@ 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
 }) => {
   const {api_key, model_id, options: credOpts} = credentials;
   const opts = !!options && Object.keys(options).length !== 0 ? options : JSON.parse(credOpts || '{}');
@@ -1084,7 +1084,7 @@ const synthRimelabs = async(logger, {
   if (!JAMBONES_DISABLE_TTS_STREAMING && !renderForCaching && !disableTtsStreaming) {
     let params = '{';
     params += `api_key=${api_key}`;
-    params += `,playback_id=${playbackIdCounter++}`;
+    params += `,playback_id=${key}`;
     params += `,model_id=${model_id}`;
     params += ',vendor=rimelabs';
     params += `,language=${language}`;
@@ -1132,7 +1132,9 @@ const synthRimelabs = async(logger, {
     throw err;
   }
 };
-const synthVerbio = async(client, logger, {credentials, stats, voice, text, renderForCaching, disableTtsStreaming}) => {
+const synthVerbio = async(client, logger, {
+  credentials, stats, voice, key, text, renderForCaching, disableTtsStreaming
+}) => {
   //https://doc.speechcenter.verbio.com/#tag/Text-To-Speech-REST-API
   if (text.length > 2000) {
     throw new Error('Verbio cannot synthesize for the text length larger than 2000 characters');
@@ -1141,7 +1143,7 @@ const synthVerbio = async(client, logger, {credentials, stats, voice, text, rend
   if (!process.env.JAMBONES_DISABLE_TTS_STREAMING && !renderForCaching && !disableTtsStreaming) {
     let params = '{';
     params += `access_token=${token.access_token}`;
-    params += `,playback_id=${playbackIdCounter++}`;
+    params += `,playback_id=${key}`;
     params += ',vendor=verbio';
     params += `,voice=${voice}`;
     params += ',write_cache_file=1';
@@ -1178,14 +1180,14 @@ 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}) => {
   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 = '{';
     params += `api_key=${api_key}`;
-    params += `,playback_id=${playbackIdCounter++}`;
+    params += `,playback_id=${key}`;
     params += `,model_id=${model_id}`;
     params += ',vendor=whisper';
     params += `,voice=${voice}`;
@@ -1227,12 +1229,12 @@ 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}) => {
   const {api_key, deepgram_tts_uri} = credentials;
   if (!JAMBONES_DISABLE_TTS_STREAMING && !renderForCaching && !disableTtsStreaming) {
     let params = '{';
     params += `api_key=${api_key}`;
-    params += `,playback_id=${playbackIdCounter++}`;
+    params += `,playback_id=${key}`;
     params += ',vendor=deepgram';
     params += `,voice=${model}`;
     params += ',write_cache_file=1';
@@ -1268,7 +1270,7 @@ 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
 }) => {
   const {api_key, model_id, embedding, options: credOpts} = credentials;
   const opts = !!options && Object.keys(options).length !== 0 ? options : JSON.parse(credOpts || '{}');
@@ -1276,7 +1278,7 @@ const synthCartesia = async(logger, {
   if (!JAMBONES_DISABLE_TTS_STREAMING && !renderForCaching && !disableTtsStreaming) {
     let params = '{';
     params += `api_key=${api_key}`;
-    params += `,playback_id=${playbackIdCounter++}`;
+    params += `,playback_id=${key}`;
     params += `,model_id=${model_id}`;
     params += ',vendor=cartesia';
     params += `,voice=${voice}`;
@@ -1339,7 +1341,7 @@ const synthCartesia = async(logger, {
 };
 
 const synthResemble = async(logger, {
-  credentials, options, stats, voice, text, renderForCaching, disableTtsStreaming
+  credentials, options, stats, voice, key, text, renderForCaching, disableTtsStreaming
 }) => {
   const {api_key, resemble_tts_uri, resemble_tts_use_tls} = credentials;
   const {project_uuid, use_hd} = options || {};
@@ -1348,7 +1350,7 @@ const synthResemble = async(logger, {
   if (!JAMBONES_DISABLE_TTS_STREAMING && !renderForCaching && !disableTtsStreaming) {
     let params = '{';
     params += `api_key=${api_key}`;
-    params += `,playback_id=${playbackIdCounter++}`;
+    params += `,playback_id=${key}`;
     params += ',vendor=resemble';
     params += `,voice=${voice}`;
     params += ',write_cache_file=1';

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@jambonz/speech-utils",
-  "version": "0.2.21",
+  "version": "0.2.23",
   "description": "TTS-related speech utilities for jambonz",
   "main": "index.js",
   "author": "Dave Horton",