Skip to content
Google Cloud API based real-time speech-to-text script built on NodeJS and SocketIO. Built for Teli at The University of Hong Kong.
Branch: master
Clone or download
Fetching latest commit…
Cannot retrieve the latest commit at this time.
Type Name Latest commit message Commit time
Failed to load latest commit information.

Google Cloud Speech-to-Text based on NodeJS and SocketIO for Real Time Detection

This is a project based on NodeJS and SocketIO for real-time detection of speech to text. This supports all the languages in Google Cloud Speech API including Cantonese and Chinese.


  • Place the authentication file (auth.json) in the root folder (get it from Google Cloud Speech API dashboard).
  • Run npm install (again on the root folder) to install the dependencies.
  • Run node app.js, and navigate to, to run the script.

Main File Structure

  • app.js - Express based script for sever-side stuff.
  • client.js - Client-side script, handling the audio context, getting the audio buffer, converting it to a format suitable for Google Cloud API and sending it back using SocketIO.
You can’t perform that action at this time.