/sherpa-ncnn

Real-time speech recognition using next-gen Kaldi with ncnn without Internet connection. Support iOS, Android, Raspberry Pi, etc.

Primary LanguageC++Apache License 2.0Apache-2.0

Introduction

You can use sherpa-ncnn for real-time speech recognition (i.e., speech-to-text) on

  • Linux
  • macOS
  • Windows
  • Embedded Linux (32-bit arm and 64-bit aarch64)
  • Android
  • etc ...

We support all platforms that ncnn supports.

Everything can be compiled from source with static link. The generated executable depends only on system libraries.

HINT: It does not depend on PyTorch or any other inference frameworks other than ncnn.

Please see the documentation https://k2-fsa.github.io/sherpa/ncnn/index.html for installation and usages, e.g.,

  • How to build an Android app
  • How to download and use pre-trained models

We provide a few YouTube videos for demonstration about real-time speech recognition with sherpa-ncnn using a microphone:

See also https://github.com/k2-fsa/sherpa