README.md

stt

Speech to Text for Node.js

Currently support Chinese only. Will add more language(en/etc) and more engine(now stt use baidu api) recently.

Quick Start

JavaScript

const stt = require('stt')

const speechStream = createReadStream('./test/fixture/测试.mp3')

stt(speechStream).then(text => {
  console.log('STT: ' + text)
})

TypeScript

import { stt } from 'stt'

const speechStream = createReadStream('./test/fixture/测试.mp3')

const text = await stt(speechStream)

console.log('STT: ' + text)

Requirement

Ffmpeg is required for converting the speech stream format.

Todo

• Support Languages
  • Chinese 普通话
  • English
• Support Stream Formats
  • MP3 Input Stream
  • ANY Input Stream
• Support Speech Recongnition Engines
  • Google Cloud Speech API
  • Apple Siri Nuance Mix
  • Bing Speech API
  • Watson Speech to Text
  • XFYun Voice Dictation
  • Baidu ASR
  • Tencent Artificial Audio Intelligence
  • HiVoice

Copyright

2016© Huan LI https://git.io/zixia

MIT LICENSE