Voicemaker

This Api make your nodeJS app or script speak.

With this API you can use and compare available TTS APIs from Google, voicemaker.in, IBM Watson, more are comming.

Configure Environment

Google cloud

To make Google cloud TTS works, define a GOOGLE_APPLICATION_CREDENTIALS env variable:

• Go to your google console
• Create a project
• Activate Google text-to-speech API for your project
• Enable Billing
• create a client
• create and download a json certificate
• set your GOOGLE_APPLICATION_CREDENTIALS environment variable to a path to your json certificate file.

Powershell:

$Env:GOOGLE_APPLICATION_CREDENTIALS = "c://pathtoyourCERT"

Unix:

export GOOGLE_APPLICATION_CREDENTIALS="/pathtoyourCERT"

Alternatively you can save your certificate in ~/.tts/googlecloud/key.json.

ElevenLabs

To make elevenlabs TTS works, define an ELEVENLABS_IO_TOKEN env variable, log on https://beta.elevenlabs.io/ and go to Profile Settings to get one.

Then put it onto your ELEVENLABS_IO_TOKEN valiable.

Powershell:

$Env:ELEVENLABS_IO_TOKEN = "123456781234567890abcdef12345678"

Unix:

export ELEVENLABS_IO_TOKEN="123456781234567890abcdef12345678"

Alternatively you can save your certificate in ~/.tts/elevenlabs/key.json (as text data).

Voicemaker

To make voicemaker TTS works, define an VOICEMAKER_IN_TOKEN env variable, to get this token, send an E-Mail to [email protected] asking for a developper token.

Then put it onto your VOICEMAKER_IN_TOKEN valiable.

Powershell:

$Env:VOICEMAKER_IN_TOKEN = "12345678-abcd-1234-1234-1234567890ab"

Unix:

export VOICEMAKER_IN_TOKEN="12345678-abcd-1234-1234-1234567890ab"

Alternatively you can save your certificate in ~/.tts/voiceMaker/key.json (as text data).

IBM Watson

To make IBM watson TTS works, define an TEXT_TO_SPEECH_APIKEY and TEXT_TO_SPEECH_URL env variable, you can get them from an ibm-credentials.env file from https://cloud.ibm.com/

Alternatively you can save your certificate in ~/.tts/watson/key.json (as text data or a json file).

Usage as a script

This package includes a binary that can use every voice from any script; this script is optimized to let you easily switch between engines, volume, speed, and pitch values, which are normalized to -100% to +100%.

npm install -g voicemaker
voicemaker --help

Help message

Usage: voicemaker [options] [command]

Options:
  -V, --version            output the version number
  -h, --help               display help for command

Commands:
  list <engine>            list available voices
  say [options] <text...>  use a TTS engin to make your computer speak
  help [command]           display help for command

Usage: voicemaker say [options] <text...>

use a TTS engine to make your computer speak

Options:
  -v, --voice <name>     select a voiceMaker, IBM watson, or a google cloud TTS voice.
  -s, --speed <speed>    customise talk speed
  -p, --pitch <pitch>    customise talk pitch
  -l, --lang <lang>      specify a lang code
  -V, --volume <volume>  increate or decrease speech volume
  -h, --help             display help for command

List available voices

For google voices

voicemaker list google

For voicemaker voices

voicemaker list voicemaker

Make it talk

Using google free voices

voicemaker say hello word
voicemaker say -l es-ES vamos a la playa

Using google cloud voices

voicemaker say -v fr-FR-Standard-C bienvenue chez nous

Using voicemaker voices

voicemaker say -v ai2-de-DE-Patrick Willkommen
voicemaker say -v ai2-ja-JP-Ayaka こんにちは

Useage as an API

npm install voicemaker

This package integrates 3 TTS API:

• the free legacy translate.google.com API.
• the google cloud texttospeech API.
• the voicemaker.in API.

Engine translate.google.com

You do not need any configuration to use the legacy google TTS just use the GoogleTTS class.

import { GoogleTTS, GoogleTTSRequest } from 'voicemaker';

const engine = new GoogleTTS();
const request = new GoogleTTSRequest('hello');
const filename = await engine.getTts(request);
console.log(`your speech is available in ${filename}`);
// Let it speak using translate.google.com engine
await engine.say(request);

Engine google cloud TTS

import { GoogleTTS2, GoogleTTS2Request } from 'voicemaker';

const engine = new GoogleTTS2();
const request = new GoogleTTS2Request('hello');
const filename = await engine.getTts(request);
console.log(`your speech is available in ${filename}`);
// Let it speak using cloud.google.com engine
await engine.say(request);

Advanced usage:

import { GoogleTTS2, GoogleTTS2Request } from 'voicemaker';

const engine = new GoogleTTS2();
const request = new GoogleTT2SRequest('hello');
const filename = await engine.getTts(request);
// change the speaker to Ai1 Ivy
request.setVoice("en-GB-Wavenet-D");
// increate pitch by 4 (min value is -20, max is 20)
request.pitch = 2;
// increate volume by 10 (min value is -20, max is 20)
request.volume = 1;
// increate speed by -10 (min value is -96 Db, max is +16Db)
request.speed = -3;

console.log(`your speech is available in ${filename}`);
// Let it speak using cloud.google.com engine
await engine.say(request);

Engine voicemaker.in

To use voicemaker.in API you must define your developer token in the VOICEMAKER_IN_TOKEN environment variable.

import { VoiceMaker, VoiceMakerRequest } from 'voicemaker';


const engine = new VoiceMaker();
const request = new VoiceMakerRequest('hello');
// change the speaker, and this call will also change the language to es-ES
request.setVoice("ai4-es-ES-Matlab");
const filename = await engine.getTts(request);
console.log(`your speech is available in ${filename}`);
// Let it speak using voicemaker.in
await engine.say(request);

Advanced usage:

import { VoiceMaker, VoiceMakerRequest } from 'voicemaker';

const engine = new VoiceMaker();
const request = new VoiceMakerRequest('Your very secret password is 123456');
// change the speaker to Ai1 Ivy
request.setVoice("ai1-Ivy");
// increate pitch by 4 (min value is -100, max is 100)
request.pitch = 4;
// increate volume by 10 (min value is -20, max is 20)
request.volume = 10;
// increate speed by -10 (min value is -100, max is +100)
request.speed = -10;
// whisper the message
request.effect = "whispered";
// whisper you password
await engine.say(request);

Notes

Cache and log

By default, every generated speech will be cached in the ~/.tts directory; a log.txt will also be added and will contain all cached requests.