0.1.13

deepgram tts support endpoint for on-premise

.husky/pre-commit

@@ -1,3 +1,3 @@
-npm audit
+#npm audit
 npm run jslint:fix || true
 npm test

lib/get-aws-sts-token.js

@@ -1,20 +1,22 @@
 const { STSClient, GetSessionTokenCommand, AssumeRoleCommand } = require('@aws-sdk/client-sts');
 const {makeAwsKey, noopLogger} = require('./utils');
 const debug = require('debug')('jambonz:speech-utils');
-const EXPIRY = 3600;
+const EXPIRY = process.env.AWS_STS_SESSION_DURATION || 3600;
+// by default reset aws session before expiry time 10 mins
+const CACHE_EXPIRY = process.env.AWS_STS_SESSION_RESET_EXPIRY || (EXPIRY - 600);
 
 async function getAwsAuthToken(
   logger, createHash, retrieveHash,
-  awsAccessKeyId, awsSecretAccessKey, awsRegion, roleArn = null) {
+  {accessKeyId, secretAccessKey, region, roleArn}) {
   logger = logger || noopLogger;
   try {
-    const key = makeAwsKey(roleArn || awsAccessKeyId);
+    const key = makeAwsKey(roleArn || accessKeyId);
     const obj = await retrieveHash(key);
     if (obj) return {...obj, servedFromCache: true};
 
     let data;
     if (roleArn) {
-      const stsClient = new STSClient({ region: awsRegion});
+      const stsClient = new STSClient({ region });
       const roleToAssume = { RoleArn: roleArn, RoleSessionName: 'Jambonz_Speech', DurationSeconds: EXPIRY};
       const command = new AssumeRoleCommand(roleToAssume);
 
@@ -22,10 +24,10 @@ async function getAwsAuthToken(
     } else {
       /* access token not found in cache, so generate it using STS */
       const stsClient = new STSClient({
-        region: awsRegion,
+        region,
         credentials: {
-          accessKeyId: awsAccessKeyId,
-          secretAccessKey: awsSecretAccessKey,
+          accessKeyId,
+          secretAccessKey,
         }
       });
       const command = new GetSessionTokenCommand({DurationSeconds: EXPIRY});
@@ -39,8 +41,7 @@ async function getAwsAuthToken(
       securityToken: data.Credentials.SessionToken
     };
 
-    /* expire 10 minutes before the hour, so we don't lose the use of it during a call */
-    createHash(key, credentials, EXPIRY - 600)
+    createHash(key, credentials, CACHE_EXPIRY)
       .catch((err) => logger.error(err, `Error saving hash for key ${key}`));
 
     return {...credentials, servedFromCache: false};

lib/get-tts-voices.js

@@ -107,7 +107,12 @@ const getAwsVoices = async(_client, createHash, retrieveHash, logger, credential
     } else if (roleArn) {
       client = new PollyClient({
         region,
-        credentials: await getAwsAuthToken(logger, createHash, retrieveHash, null, null, region, roleArn),
+        credentials: await getAwsAuthToken(
+          logger, createHash, retrieveHash,
+          {
+            region,
+            roleArn
+          }),
       });
     } else {
       client = new PollyClient({region});

lib/synth-audio.js

@@ -151,6 +151,11 @@ 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,
@@ -281,7 +286,12 @@ const synthPolly = async(createHash, retrieveHash, logger,
     } else if (roleArn) {
       polly = new PollyClient({
         region,
-        credentials: await getAwsAuthToken(logger, createHash, retrieveHash, null, null, region, roleArn),
+        credentials: await getAwsAuthToken(
+          logger, createHash, retrieveHash,
+          {
+            region,
+            roleArn
+          }),
       });
     } else {
       // AWS RoleArn assigned to Instance profile
@@ -904,13 +914,14 @@ 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 {
@@ -920,8 +931,9 @@ 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'
     });

package.json

@@ -1,6 +1,6 @@
 {
   "name": "@jambonz/speech-utils",
-  "version": "0.1.9",
+  "version": "0.1.13",
   "description": "TTS-related speech utilities for jambonz",
   "main": "index.js",
   "author": "Dave Horton",
@@ -36,7 +36,7 @@
     "form-urlencoded": "^6.1.4",
     "google-protobuf": "^3.21.2",
     "ibm-watson": "^8.0.0",
-    "microsoft-cognitiveservices-speech-sdk": "1.36.0",
+    "microsoft-cognitiveservices-speech-sdk": "1.38.0",
     "openai": "^4.25.0",
     "undici": "^6.4.0"
   },

test/aws.js

@@ -19,12 +19,20 @@ test('AWS - create and cache auth token', async(t) => {
       return;
   }
   try {
-    let obj = await getAwsAuthToken(process.env.AWS_ACCESS_KEY_ID, process.env.AWS_SECRET_ACCESS_KEY, process.env.AWS_REGION);
+    let obj = await getAwsAuthToken({
+      accessKeyId: process.env.AWS_ACCESS_KEY_ID, 
+      secretAccessKey: process.env.AWS_SECRET_ACCESS_KEY,
+      region: process.env.AWS_REGION
+    });
     //console.log({obj}, 'received auth token from AWS');
     t.ok(obj.securityToken && !obj.servedFromCache, 'successfullY generated auth token from AWS');
 
     await sleep(250);
-    obj = await getAwsAuthToken(process.env.AWS_ACCESS_KEY_ID, process.env.AWS_SECRET_ACCESS_KEY, process.env.AWS_REGION);
+    obj = await getAwsAuthToken({
+      accessKeyId: process.env.AWS_ACCESS_KEY_ID, 
+      secretAccessKey: process.env.AWS_SECRET_ACCESS_KEY,
+      region: process.env.AWS_REGION
+    });
     //console.log({obj}, 'received auth token from AWS - second request');
     t.ok(obj.securityToken && obj.servedFromCache, 'successfully received access token from cache');