k2-fsa/sherpa-onnx

Speech-to-text, text-to-speech, speaker diarization, speech enhancement, source separation, and VAD using next-gen Kaldi with onnxruntime without Internet connection. Support embedded systems, Android, iOS, HarmonyOS, Raspberry Pi, RISC-V, RK NPU, Axera NPU, Ascend NPU, x86_64 servers, websocket server/client, support 12 programming languages

88
/ 100
Verified

This tool helps you process audio information directly on your device without needing an internet connection. It can convert spoken audio into text, turn text into natural-sounding speech, identify who is speaking, or even separate different voices or instruments in a recording. This is for anyone working with audio data, such as content creators, transcriptionists, or developers building offline voice applications on various devices.

10,885 stars and 1,502 monthly downloads. Actively maintained with 134 commits in the last 30 days. Available on PyPI.

Use this if you need to perform advanced audio processing tasks like speech recognition, text-to-speech, or speaker identification directly on a local device, even embedded systems, without relying on cloud services.

Not ideal if you primarily work with text-based data and do not have a need for offline, on-device audio processing capabilities.

audio-transcription voice-assistance multimedia-production accessibility-tech embedded-systems
Maintenance 25 / 25
Adoption 17 / 25
Maturity 25 / 25
Community 21 / 25

How are scores calculated?

Stars

10,885

Forks

1,235

Language

C++

License

Apache-2.0

Last pushed

Mar 18, 2026

Monthly downloads

1,502

Commits (30d)

134

Dependencies

1

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/voice-ai/k2-fsa/sherpa-onnx"

Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.