k2-fsa/sherpa

Speech-to-text server framework with next-gen Kaldi

986

stars

631

commits

C++

primary language

Sep 10, 2026

updated

k2-fsa.github.io/sherpa

asr

cpp

ctc

end-to-end-asr

python

pytorch

speech-recognition

transducer

websocket

README

sherpa

sherpa is an open-source speech-text-text inference framework using PyTorch, focusing exclusively on end-to-end (E2E) models, namely transducer- and CTC-based models. It provides both C++ and Python APIs.

This project focuses on deployment, i.e., using pre-trained models to transcribe speech. If you are interested in how to train or fine-tune your own models, please refer to icefall.

We also have other similar projects that don't depend on PyTorch:

sherpa-onnx and sherpa-ncnn also support iOS, Android and embedded systems.

Installation and Usage

Please refer to the documentation at https://k2-fsa.github.io/sherpa/

Try it in your browser

Try sherpa from within your browser without installing anything: https://huggingface.co/spaces/k2-fsa/automatic-speech-recognition

Contributors

(top 30 of 58)

458 commits

26 commits

24 commits

22 commits

Similar repos

k2-fsa/sherpa-onnx

Speech-to-text, text-to-speech, speaker diarization, speech enhancement, source separation, and VAD…

14,726

same org

k2-fsa/icefall

1,491

same org

k2-fsa/sherpa-ncnn

Real-time speech recognition and voice activity detection (VAD) using next-gen Kaldi with ncnn…

1,781

same org

KoljaB/RealtimeSTT

A robust, efficient, low-latency speech-to-text library with advanced voice activity detection,…

10,117

Banafo/Kroko-ASR

Model

cq27-dev/sherpa-onnx

cq27 fork of k2-fsa/sherpa-onnx — patches sherpa-onnx-sys for Android cross-compile via cargo-ndk

espnet/espnet

End-to-End Speech Processing Toolkit

9,949

wenet-e2e/wenet

Production First and Production Ready End-to-End Speech Recognition Toolkit

5,233

Languages

C++

51.5%

Python

36.6%

Shell

4.8%

CMake

4.0%

JavaScript

2.0%

k2-fsa/sherpa

Speech-to-text server framework with next-gen Kaldi

986

stars

631

commits

C++

primary language

Sep 10, 2026

updated

k2-fsa.github.io/sherpa

asr

cpp

ctc

end-to-end-asr

python

pytorch

speech-recognition

transducer

websocket

README

sherpa

This project focuses on deployment, i.e., using pre-trained models to transcribe speech. If you are interested in how to train or fine-tune your own models, please refer to icefall.

We also have other similar projects that don't depend on PyTorch: