About

This is a server for highly accurate offline speech recognition using Kaldi and Vosk-API.

There are four different servers which support four major communication protocols - MQTT, GRPC, WebRTC and Websocket

The server can be used locally to provide the speech recognition to smart home, PBX like freeswitch or asterisk. The server can also run as a backend for streaming speech recognition on the web, it can power chatbots, websites and telephony.

Documentation

For documentation and instructions see Vosk Website