Speech Synthesis in JavaScript

Backstory

Hi. I want to develop my personal assistant. I needed to implement speech synthesis. I had two ways:

1. Use online speech synthesis like googleTTS
2. Use offline npm modules and my system

So, I chose the second one.

And, here I’ll tell you everything about my adventure and how to use speech synthesis on your device using node.js

New era

I decided to use Say.js but the repository is abandoned by the author, and I needed to set “UTF-8” encoding for my voice. Originally I wanted to create a fork for Say.js module, but more convenient was to create my own module (I renamed variables, added encodings and created convenient documentation README). So, I developed my module based on Say.js named “speakertts”. If you want to contribute follow github

How it works

This module uses node.js module “child-process” to call powershell and execute built-in SpeechSynthesizer object (I talk about windows). You can use your powershell and paste this code:

Add-Type -AssemblyName System.speech
$speak = New-Object System.Speech.Synthesis.SpeechSynthesizer
$speak.Speak('Hello...')

You can read more here

How to use

1. Install speakertts module
   

   npm i speakertts
   

2. Get your voices using getInstalledVoices() function
   

   const speaker = require('speakertts')
   
   speaker.getInstalledVoices((error, voices) => {
       if (error) {
           console.error('Error:', error);
       } else {
           console.log('Installed voices:', voices);
       }
   });
   

3. Call SpeechSynthesizer using speak function
   

   speaker.speak('Hello, world', 'Your voice', 1, 'ASCII', 
   (error) => {
     if (error) {
       return console.error('Error speaking!', error)
     }
   
     console.log('text to speech complete')
   })
   

It was basics, other functions you can look up on github or npm

How to set your not system voice (for windows)

You just need to install SAPI format voice, and repeat previous chapter