fix: upgrade @google-cloud/text-to-speech from 5.0.2 to 5.3.0

Snyk has created this PR to upgrade @google-cloud/text-to-speech from 5.0.2 to 5.3.0.

See this package in npm:
@google-cloud/text-to-speech

See this project in Snyk:
https://app.snyk.io/org/davehorton/project/6f41db32-9ff7-45b4-83ae-9512305f387b?utm_source=github&utm_medium=referral&page=upgrade-pr

.github/workflows/ci.yml

@@ -13,6 +13,11 @@ jobs:
         with:
           node-version: lts/*
       - run: npm install
+      - name: Install Docker Compose
+        run: |
+          sudo curl -L "https://github.com/docker/compose/releases/download/1.29.2/docker-compose-$(uname -s)-$(uname -m)" -o /usr/local/bin/docker-compose
+          sudo chmod +x /usr/local/bin/docker-compose
+          docker-compose --version
       - run: npm run jslint
       - run: sudo apt update && sudo apt install -y squid
       - run: sudo cp test/squid.conf /etc/squid/squid.conf

lib/config.js

@@ -1,6 +1,7 @@
 const JAMBONES_TTS_TRIM_SILENCE = process.env.JAMBONES_TTS_TRIM_SILENCE;
 const JAMBONES_DISABLE_TTS_STREAMING = process.env.JAMBONES_DISABLE_TTS_STREAMING;
 const JAMBONES_DISABLE_AZURE_TTS_STREAMING = process.env.JAMBONES_DISABLE_AZURE_TTS_STREAMING;
+const JAMBONES_EAGERLY_PRE_CACHE_AUDIO = process.env.JAMBONES_EAGERLY_PRE_CACHE_AUDIO;
 
 const JAMBONES_HTTP_PROXY_IP = process.env.JAMBONES_HTTP_PROXY_IP;
 const JAMBONES_HTTP_PROXY_PORT = process.env.JAMBONES_HTTP_PROXY_PORT;
@@ -18,6 +19,7 @@ module.exports = {
   JAMBONES_HTTP_PROXY_IP,
   JAMBONES_HTTP_PROXY_PORT,
   JAMBONES_TTS_CACHE_DURATION_MINS,
+  JAMBONES_EAGERLY_PRE_CACHE_AUDIO,
   TMP_FOLDER,
   HTTP_TIMEOUT
 };

lib/synth-audio.js

@@ -44,6 +44,7 @@ const {
   JAMBONES_HTTP_PROXY_IP,
   JAMBONES_HTTP_PROXY_PORT,
   JAMBONES_TTS_CACHE_DURATION_MINS,
+  JAMBONES_EAGERLY_PRE_CACHE_AUDIO,
 } = require('./config');
 const EXPIRES = JAMBONES_TTS_CACHE_DURATION_MINS;
 const OpenAI = require('openai');
@@ -86,7 +87,7 @@ const trimTrailingSilence = (buffer) => {
  */
 async function synthAudio(client, createHash, retrieveHash, logger, stats, { account_sid,
   vendor, language, voice, gender, text, engine, salt, model, credentials, deploymentId,
-  disableTtsCache, renderForCaching, disableTtsStreaming, options
+  disableTtsCache, renderForCaching = false, disableTtsStreaming, options
 }) {
   let audioBuffer;
   let servedFromCache = false;
@@ -151,21 +152,48 @@ async function synthAudio(client, createHash, retrieveHash, logger, stats, { acc
     assert.ok(voice, 'synthAudio requires voice when verbio is used');
     assert.ok(credentials.client_id, 'synthAudio requires client_id when verbio is used');
     assert.ok(credentials.client_secret, 'synthAudio requires client_secret when verbio is used');
+  } else if ('deepgram' === vendor) {
+    if (!credentials.deepgram_tts_uri) {
+      assert.ok(credentials.api_key, 'synthAudio requires api_key when deepgram is used');
+    }
+
   }
+
   const key = makeSynthKey({
     account_sid,
     vendor,
     language: language || '',
     voice: voice || deploymentId,
     engine,
-    text
+    text,
+    renderForCaching
   });
   let filePath;
-  filePath = makeFilePath(vendor, key, salt);
+  filePath = makeFilePath({vendor, key, salt, renderForCaching});
   debug(`synth key is ${key}`);
   let cached;
   if (!disableTtsCache) {
     cached = await client.get(key);
+    /**
+    *  If we are using tts streaming and also precaching audio, audio could have been cached by streaming (r8)
+    * or here in speech-utils due to precaching (mp3), so we need to check for both keys.
+    */
+    if (!cached && JAMBONES_EAGERLY_PRE_CACHE_AUDIO) {
+      const preCachekey = makeSynthKey({
+        account_sid,
+        vendor,
+        language: language || '',
+        voice: voice || deploymentId,
+        engine,
+        text,
+        renderForCaching: true
+      });
+      cached = await client.get(preCachekey);
+      if (cached) {
+        // Precache audio is available update filpath with precache file extension.
+        filePath = makeFilePath({vendor, key, salt, renderForCaching: true});
+      }
+    }
   }
   if (cached) {
     // found in cache - extend the expiry and use it
@@ -909,13 +937,14 @@ const synthWhisper = async(logger, {credentials, stats, voice, text, renderForCa
 };
 
 const synthDeepgram = async(logger, {credentials, stats, model, text, renderForCaching, disableTtsStreaming}) => {
-  const {api_key} = credentials;
+  const {api_key, deepgram_tts_uri} = credentials;
   if (!JAMBONES_DISABLE_TTS_STREAMING && !renderForCaching && !disableTtsStreaming) {
     let params = '';
     params += `{api_key=${api_key}`;
     params += ',vendor=deepgram';
     params += `,voice=${model}`;
     params += ',write_cache_file=1';
+    if (deepgram_tts_uri) params += `,endpoint=${deepgram_tts_uri}`;
     params += '}';
 
     return {
@@ -925,8 +954,9 @@ const synthDeepgram = async(logger, {credentials, stats, model, text, renderForC
     };
   }
   try {
-    const post = bent('https://api.deepgram.com', 'POST', 'buffer', {
-      'Authorization': `Token ${api_key}`,
+    const post = bent(deepgram_tts_uri || 'https://api.deepgram.com', 'POST', 'buffer', {
+      // on-premise deepgram does not  require to have api_key
+      ...(api_key && {'Authorization': `Token ${api_key}`}),
       'Accept': 'audio/mpeg',
       'Content-Type': 'application/json'
     });

lib/utils.js

@@ -16,29 +16,31 @@ const debug = require('debug')('jambonz:realtimedb-helpers');
  */
 //const nuanceClientMap = new Map();
 
-function makeSynthKey({account_sid = '', vendor, language, voice, engine = '', text}) {
+function makeSynthKey({
+  account_sid = '', vendor, language, voice, engine = '', text,
+  renderForCaching = false}) {
   const hash = crypto.createHash('sha1');
   hash.update(`${language}:${vendor}:${voice}:${engine}:${text}`);
   const hexHashKey = hash.digest('hex');
   const accountKey = account_sid ? `:${account_sid}` : '';
-  const namespace = vendor.startsWith('custom') ? vendor : getFileExtension(vendor);
+  const namespace = vendor.startsWith('custom') ? vendor : getFileExtension({vendor, renderForCaching});
   const key = `tts${accountKey}:${namespace}:${hexHashKey}`;
   return key;
 }
 
-function makeFilePath(vendor, key, salt = '') {
-  const extension = getFileExtension(vendor);
+function makeFilePath({vendor, key, salt = '', renderForCaching = false}) {
+  const extension = getFileExtension({vendor, renderForCaching});
   return `${TMP_FOLDER}/${key.replace('tts:', `tts-${salt}`)}.${extension}`;
 }
 
-function getFileExtension(vendor) {
+function getFileExtension({vendor, renderForCaching = false}) {
   const mp3Extension = 'mp3';
   const r8Extension = 'r8';
 
   switch (vendor) {
     case 'azure':
     case 'microsoft':
-      if (!JAMBONES_DISABLE_TTS_STREAMING || JAMBONES_TTS_TRIM_SILENCE) {
+      if (!renderForCaching && !JAMBONES_DISABLE_TTS_STREAMING || JAMBONES_TTS_TRIM_SILENCE) {
         return r8Extension;
       } else {
         return mp3Extension;
@@ -46,7 +48,7 @@ function getFileExtension(vendor) {
     case 'deepgram':
     case 'elevenlabs':
     case 'rimlabs':
-      if (!JAMBONES_DISABLE_TTS_STREAMING) {
+      if (!renderForCaching && !JAMBONES_DISABLE_TTS_STREAMING) {
         return r8Extension;
       } else {
         return mp3Extension;

package-lock.json

@@ -1,17 +1,17 @@
 {
   "name": "@jambonz/speech-utils",
-  "version": "0.1.12",
+  "version": "0.1.15",
   "lockfileVersion": 2,
   "requires": true,
   "packages": {
     "": {
       "name": "@jambonz/speech-utils",
-      "version": "0.1.12",
+      "version": "0.1.15",
       "license": "MIT",
       "dependencies": {
         "@aws-sdk/client-polly": "^3.496.0",
         "@aws-sdk/client-sts": "^3.496.0",
-        "@google-cloud/text-to-speech": "^5.0.2",
+        "@google-cloud/text-to-speech": "^5.3.0",
         "@grpc/grpc-js": "^1.9.14",
         "@jambonz/realtimedb-helpers": "^0.8.7",
         "bent": "^7.3.12",
@@ -1230,9 +1230,10 @@
       }
     },
     "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",
-      "integrity": "sha512-Q11Ddh9eHKSDA3E/KSqMITgVprXb0XgIKuJP9F5ScJ1T9h+DNrbgIU7shd0QOlPqb8ruQRiTOqL08+Mq5R89Ow==",
+      "version": "5.3.0",
+      "resolved": "https://registry.npmjs.org/@google-cloud/text-to-speech/-/text-to-speech-5.3.0.tgz",
+      "integrity": "sha512-jc0TEHkSGrQErlaFCJ59YF/NKZUMBfrdpEc5it+AXVWbLvnVTbgCrxZn4ny+PFs310kyguaqLQ2qwScTDAaFxA==",
+      "license": "Apache-2.0",
       "dependencies": {
         "google-gax": "^4.0.3"
       },
@@ -8338,9 +8339,9 @@
       "dev": true
     },
     "@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",
-      "integrity": "sha512-Q11Ddh9eHKSDA3E/KSqMITgVprXb0XgIKuJP9F5ScJ1T9h+DNrbgIU7shd0QOlPqb8ruQRiTOqL08+Mq5R89Ow==",
+      "version": "5.3.0",
+      "resolved": "https://registry.npmjs.org/@google-cloud/text-to-speech/-/text-to-speech-5.3.0.tgz",
+      "integrity": "sha512-jc0TEHkSGrQErlaFCJ59YF/NKZUMBfrdpEc5it+AXVWbLvnVTbgCrxZn4ny+PFs310kyguaqLQ2qwScTDAaFxA==",
       "requires": {
         "google-gax": "^4.0.3"
       }

package.json

@@ -1,6 +1,6 @@
 {
   "name": "@jambonz/speech-utils",
-  "version": "0.1.12",
+  "version": "0.1.15",
   "description": "TTS-related speech utilities for jambonz",
   "main": "index.js",
   "author": "Dave Horton",
@@ -28,7 +28,7 @@
   "dependencies": {
     "@aws-sdk/client-polly": "^3.496.0",
     "@aws-sdk/client-sts": "^3.496.0",
-    "@google-cloud/text-to-speech": "^5.0.2",
+    "@google-cloud/text-to-speech": "^5.3.0",
     "@grpc/grpc-js": "^1.9.14",
     "@jambonz/realtimedb-helpers": "^0.8.7",
     "bent": "^7.3.12",