sherpa-onnx
Next-gen Kaldi toolkit for offline speech recognition, text-to-speech, speaker diarization, VAD and more on top of onnxruntime, from servers down to phones and embedded boards.
About sherpa-onnx
Cloud speech APIs need a network connection and send audio off-device, which rules them out for embedded, privacy-sensitive or offline apps.
Who it’s for
- Developers adding on-device speech recognition or TTS to mobile, desktop or embedded apps
- Engineers deploying speech models on Raspberry Pi, RISC-V or NPU hardware
When to consider it
Consider sherpa-onnx when you need speech-to-text, text-to-speech, keyword spotting or diarization to run without Internet access, with bindings for 12 languages including Python, C++, Java, Swift, Go and Rust, plus WebAssembly.
Tradeoffs & limitations
- Apache-2.0 licensed library; you pick, download and ship the models yourself, and it runs on your own hardware with no hosted service.
SOTA overview · Documentation-based assessment · Sources & review method
Updates
No updates shared yet.
Discussion
Newest firstAsk a question or share how you use sherpa-onnx.
Keep it helpful. Community rules
Loading discussion…
Sources & review method
Documentation-based assessment · Nov 1, 2026 · Prepared with AI assistance; not a hands-on benchmark.
Checked by SOTA · AI-assisted documentation review. Selection advice is our assessment; verify current requirements for your deployment.
Import history & original evidence
sherpa-onnx on GitHub
Next-gen Kaldi toolkit for offline speech recognition, text-to-speech, speaker diarization, VAD and more on top of onnxruntime, from servers down to phones and embedded boards.
Based on official pages and announcements checked on 2026-11-01. No hands-on test, performance benchmark or popularity ranking is claimed.