@jambonz/speech-utils 0.2.23 → 0.2.25

package/lib/synth-audio.js

@@ -210,13 +210,14 @@ async function synthAudio(client, createHash, retrieveHash, logger, stats, { acc
       case 'polly':
         vendorLabel = 'aws';
         audioData = await synthPolly(createHash, retrieveHash, logger,
-          {credentials, stats, language, voice, key, text, engine, renderForCaching, disableTtsStreaming});
+          {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, key, text, deploymentId,
-          renderForCaching, disableTtsStreaming});
+          renderForCaching, disableTtsStreaming, disableTtsCache});
         break;
       case 'nuance':
         model = model || 'enhanced';
@@ -224,7 +225,7 @@ async function synthAudio(client, createHash, retrieveHash, logger, stats, { acc
         break;
       case 'nvidia':
         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, key, text});
@@ -234,44 +235,50 @@ async function synthAudio(client, createHash, retrieveHash, logger, stats, { acc
         break;
       case 'elevenlabs':
         audioData = await synthElevenlabs(logger, {
-          credentials, options, stats, language, voice, key, 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, key, 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, key, 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, key, 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, key, text, renderForCaching, disableTtsStreaming});
+          credentials, options, stats, language, voice, key, text, renderForCaching, disableTtsStreaming,
+          disableTtsCache});
         break;
       case 'whisper':
         audioData = await synthWhisper(logger, {
-          credentials, stats, voice, key, text, instructions, renderForCaching, disableTtsStreaming});
+          credentials, stats, voice, key, text, instructions, renderForCaching, disableTtsStreaming,
+          disableTtsCache});
         break;
       case 'verbio':
         audioData = await synthVerbio(client, logger, {
-          credentials, stats, voice, key, 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, key, text,
-          renderForCaching, disableTtsStreaming});
+          renderForCaching, disableTtsStreaming, disableTtsCache});
         break;
       case 'resemble':
         audioData = await synthResemble(logger, {
-          credentials, stats, voice, key, 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, key, text});
+          {credentials, stats, language, voice, key, text, renderForCaching, disableTtsStreaming, disableTtsCache});
         break;
       default:
         assert(`synthAudio: unsupported speech vendor ${vendor}`);
@@ -306,14 +313,14 @@ async function synthAudio(client, createHash, retrieveHash, logger, stats, { acc
 }
 
 const synthPolly = async(createHash, retrieveHash, logger,
-  {credentials, stats, language, voice, engine, key, text, renderForCaching, disableTtsStreaming}) => {
+  {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=1';
+    params += `,write_cache_file=${disableTtsCache ? 0 : 1}`;
     params += ',vendor=aws';
     if (accessKeyId && secretAccessKey) {
       if (accessKeyId) params += `,accessKeyId=${accessKeyId}`;
@@ -563,7 +570,8 @@ const synthMicrosoft = async(logger, {
   key,
   text,
   renderForCaching,
-  disableTtsStreaming
+  disableTtsStreaming,
+  disableTtsCache
 }) => {
   try {
     const {api_key: apiKey, region, use_custom_tts, custom_tts_endpoint, custom_tts_endpoint_url} = credentials;
@@ -596,7 +604,7 @@ const synthMicrosoft = async(logger, {
       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}`;
@@ -769,7 +777,7 @@ const synthNuance = async(client, logger, {credentials, stats, voice, model, tex
 };
 
 const synthNvidia = async(client, logger, {
-  credentials, stats, language,  voice, model, key, 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) {
@@ -778,7 +786,7 @@ const synthNvidia = async(client, logger, {
     params += `,playback_id=${key}`;
     params += `,voice=${voice}`;
     params += `,language=${language}`;
-    params += ',write_cache_file=1';
+    params += `,write_cache_file=${disableTtsCache ? 0 : 1}`;
     params += '}';
 
     return {
@@ -818,9 +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}`,
@@ -849,9 +875,9 @@ const synthCustomVendor = async(logger, {credentials, stats, language, voice, te
 };
 
 const synthElevenlabs = async(logger, {
-  credentials, options, stats, voice, key, 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 */
@@ -862,7 +888,8 @@ const synthElevenlabs = async(logger, {
     params += ',vendor=elevenlabs';
     params += `,voice=${voice}`;
     params += `,model_id=${model_id}`;
-    params += ',write_cache_file=1';
+    params += `,write_cache_file=${disableTtsCache ? 0 : 1}`;
+    if (api_uri) params += `,api_uri=${api_uri}`;
     if (opts.optimize_streaming_latency !== null && opts.optimize_streaming_latency !== undefined) {
       params += `,optimize_streaming_latency=${opts.optimize_streaming_latency}`;
     }
@@ -888,7 +915,7 @@ const synthElevenlabs = async(logger, {
   const optimize_streaming_latency = opts.optimize_streaming_latency ?
     `?optimize_streaming_latency=${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'
@@ -915,7 +942,7 @@ const synthElevenlabs = async(logger, {
 };
 
 const synthPlayHT = async(client, logger, {
-  credentials, options, stats, voice, language, key, 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 || '{}');
@@ -957,7 +984,7 @@ const synthPlayHT = async(client, logger, {
     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}`;
@@ -1010,7 +1037,7 @@ const synthPlayHT = async(client, logger, {
 };
 
 const synthInworld = async(logger, {
-  credentials, options, stats, voice, key, 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 || '{}');
@@ -1023,7 +1050,7 @@ const synthInworld = async(logger, {
     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}`;
@@ -1075,7 +1102,7 @@ const synthInworld = async(logger, {
 };
 
 const synthRimelabs = async(logger, {
-  credentials, options, stats, language, voice, key, 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 || '{}');
@@ -1089,7 +1116,7 @@ const synthRimelabs = async(logger, {
     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=${opts.reduceLatency}`;
     // Arcana model parameters
@@ -1133,7 +1160,7 @@ const synthRimelabs = async(logger, {
   }
 };
 const synthVerbio = async(client, logger, {
-  credentials, stats, voice, key, text, renderForCaching, disableTtsStreaming
+  credentials, stats, voice, key, text, renderForCaching, disableTtsStreaming, disableTtsCache
 }) => {
   //https://doc.speechcenter.verbio.com/#tag/Text-To-Speech-REST-API
   if (text.length > 2000) {
@@ -1146,7 +1173,7 @@ const synthVerbio = async(client, logger, {
     params += `,playback_id=${key}`;
     params += ',vendor=verbio';
     params += `,voice=${voice}`;
-    params += ',write_cache_file=1';
+    params += `,write_cache_file=${disableTtsCache ? 0 : 1}`;
     params += '}';
 
     return {
@@ -1181,7 +1208,7 @@ const synthVerbio = async(client, logger, {
 };
 
 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) {
@@ -1191,7 +1218,7 @@ const synthWhisper = async(logger, {credentials, stats, voice, key, text, instru
     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, ';')}`;
@@ -1229,7 +1256,8 @@ const synthWhisper = async(logger, {credentials, stats, voice, key, text, instru
   }
 };
 
-const synthDeepgram = async(logger, {credentials, stats, model, key, 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 = '{';
@@ -1237,7 +1265,7 @@ const synthDeepgram = async(logger, {credentials, stats, model, key, text, rende
     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 += '}';
 
@@ -1270,7 +1298,7 @@ const synthDeepgram = async(logger, {credentials, stats, model, key, text, rende
 };
 
 const synthCartesia = async(logger, {
-  credentials, options, stats, voice, language, key, 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 || '{}');
@@ -1282,7 +1310,7 @@ const synthCartesia = async(logger, {
     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}`;
@@ -1341,7 +1369,7 @@ const synthCartesia = async(logger, {
 };
 
 const synthResemble = async(logger, {
-  credentials, options, stats, voice, key, text, renderForCaching, disableTtsStreaming
+  credentials, options, stats, voice, key, text, renderForCaching, disableTtsStreaming, disableTtsCache
 }) => {
   const {api_key, resemble_tts_uri, resemble_tts_use_tls} = credentials;
   const {project_uuid, use_hd} = options || {};
@@ -1353,7 +1381,7 @@ const synthResemble = async(logger, {
     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}`;

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@jambonz/speech-utils",
-  "version": "0.2.23",
+  "version": "0.2.25",
   "description": "TTS-related speech utilities for jambonz",
   "main": "index.js",
   "author": "Dave Horton",

package/test/synth.js

@@ -554,6 +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');
@@ -575,6 +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');
@@ -589,6 +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`);