ben.carrasco/speech-utils
1
0
Fork 0
mirror of https://github.com/jambonz/speech-utils.git synced 2026-01-25 02:08:26 +00:00
Code Issues Packages Projects Releases Wiki Activity

Merge branch 'main' into feat/azure_tts

Browse Source
This commit is contained in:
Hoan Luu Huu
2024-03-10 06:39:00 +07:00
committed by GitHub
parent 31a54f595b dbfabeaddf
commit 2988e800b1
2 changed files with 14 additions and 19 deletions
Download Patch File Download Diff File Expand all files Collapse all files

31
lib/synth-audio.js
View File

@@ -77,7 +77,7 @@ const trimTrailingSilence = (buffer) => {
*/
async function synthAudio(client, logger, stats, { account_sid,
vendor, language, voice, gender, text, engine, salt, model, credentials, deploymentId,
disableTtsCache, renderForCaching, options
disableTtsCache, renderForCaching, disableTtsStreaming, options
}) {
let audioBuffer;
let servedFromCache = false;
@@ -208,21 +208,14 @@ async function synthAudio(client, logger, stats, { account_sid,
break;
case 'elevenlabs':
audioBuffer = await synthElevenlabs(logger, {
credentials, options, stats, language, voice, text, renderForCaching, filePath
credentials, options, stats, language, voice, text, renderForCaching, disableTtsStreaming, filePath
});
if (typeof audioBuffer === 'object' && audioBuffer.filePath) {
return audioBuffer;
}
else {
audioBuffer = await synthElevenlabs(logger, {credentials, options, stats, language, voice, text, filePath});
}
if (audioBuffer?.filePath) return audioBuffer;
break;
case 'whisper':
audioBuffer = await synthWhisper(logger, {credentials, stats, voice, text, renderForCaching});
if (typeof audioBuffer === 'object' && audioBuffer.filePath) {
return audioBuffer;
}
audioBuffer = await synthWhisper(logger, {credentials, stats, voice, text});
audioBuffer = await synthWhisper(logger, {
credentials, stats, voice, text, renderForCaching, disableTtsStreaming});
if (audioBuffer?.filePath) return audioBuffer;
break;
case 'deepgram':
audioBuffer = await synthDeepgram(logger, {credentials, stats, model, text});
@@ -637,12 +630,14 @@ const synthCustomVendor = async(logger, {credentials, stats, language, voice, te
}
};
const synthElevenlabs = async(logger, {credentials, options, stats, language, voice, text, renderForCaching}) => {
const synthElevenlabs = async(logger, {
credentials, options, stats, voice, text, renderForCaching, disableTtsStreaming
}) => {
const {api_key, model_id, 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 */
if (!process.env.JAMBONES_DISABLE_TTS_STREAMING && !renderForCaching) {
if (!process.env.JAMBONES_DISABLE_TTS_STREAMING && !renderForCaching && !disableTtsStreaming) {
let params = '';
params += `{api_key=${api_key}`;
params += `,model_id=${model_id}`;
@@ -686,10 +681,10 @@ const synthElevenlabs = async(logger, {credentials, options, stats, language, vo
}
};
const synthWhisper = async(logger, {credentials, stats, voice, text, renderForCaching}) => {
const synthWhisper = async(logger, {credentials, stats, voice, text, 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 (!process.env.JAMBONES_DISABLE_TTS_STREAMING && !renderForCaching) {
if (!process.env.JAMBONES_DISABLE_TTS_STREAMING && !renderForCaching && !disableTtsStreaming) {
let params = '';
params += `{api_key=${api_key}`;
params += `,model_id=${model_id}`;
@@ -728,7 +723,7 @@ const synthWhisper = async(logger, {credentials, stats, voice, text, renderForCa
const synthDeepgram = async(logger, {credentials, stats, model, text}) => {
const {api_key} = credentials;
try {
const post = bent('https://api.beta.deepgram.com', 'POST', 'buffer', {
const post = bent('https://api.deepgram.com', 'POST', 'buffer', {
'Authorization': `Token ${api_key}`,
'Accept': 'audio/mpeg',
'Content-Type': 'application/json'

2
test/synth.js
View File

@@ -539,7 +539,7 @@ test('Deepgram speech synth tests', async(t) => {
credentials: {
api_key: process.env.DEEPGRAM_API_KEY
},
model: 'alpha-asteria-en-v2',
model: 'aura-asteria-en',
text,
});
t.ok(!opts.servedFromCache, `successfully synthesized deepgram audio to ${opts.filePath}`);