sherpa-onnx

by k2-fsa

ONNX-runtime ASR: Whisper, Zipformer, Paraformer on every platform.

Looking at sherpa-onnx? Try this first.

Drop your audio. Transcript in seconds. First transcript free, then $4 = 300 min

TL;DR

Best for cross-platform on-device ASR — Android/iOS/Linux/Web, no Python at runtime. Pricing: free.

Category
Open source
License
Apache-2.0
Stars
★ 14.0k
Last push
2026-08-07
Pricing
free
Platforms
Linux, macOS, Windows, iOS, Android, Web

What it is

ONNX-runtime inference for ASR/TTS/VAD/diarization. Ships Java/Kotlin/Swift/JS/.NET bindings. Apache-2.0.

Best for: Cross-platform on-device ASR — Android/iOS/Linux/Web, no Python at runtime.
Watch out for: ONNX export quality varies per model.

Install / use

pip install sherpa-onnx

Features

Speaker diarizationYes
Word-level timestampsYes
Streaming / real-timeYes
Languages supported60
HIPAA eligibleNo

Links

GitHub repo ↗

sherpa-onnx vs Whipscribe

Featuresherpa-onnxWhipscribe
CategoryOpen sourceTranscription APIs
Pricingfree$4–$24 one-time packs (credits never expire) · $2 single unlock · 60 min free on signup
Speaker diarizationYesYes
Word timestampsYesYes
StreamingYesNo
Languages6099
PlatformsLinux, macOS, Windows, iOS, Android, WebWeb, API, MCP

Alternatives to sherpa-onnx

OpenAI Whisper
OpenAI

The reference open-source multilingual ASR model from OpenAI.

OSS · MIT ★ 107.0k
whisper.cpp
Georgi Gerganov

C/C++ port of Whisper — runs on anything, from a Raspberry Pi to Apple Silicon.

OSS · MIT ★ 52.8k
faster-whisper
SYSTRAN

4× faster than reference Whisper using CTranslate2 — production sweet spot.

OSS · MIT ★ 24.8k

Whipscribe is a managed faster-whisper + whisperX service. If you want transcripts without running infrastructure, paste a URL or drop a file in the form below — you'll have a transcript in seconds.