@jambonz/speech-utils 0.2.25 → 0.2.27

package/.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: |

package/lib/synth-audio.js

@@ -1316,6 +1316,7 @@ const synthCartesia = async(logger, {
     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 {
@@ -1328,7 +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: {
@@ -1341,13 +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',
@@ -1355,8 +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
     };

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@jambonz/speech-utils",
-  "version": "0.2.25",
+  "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",

package/test/synth.js

@@ -716,7 +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));