sherpa-onnx
A next-generation Kaldi-based toolkit for offline speech-to-text, text-to-speech, and audio processing across 12 languages and diverse hardware.
What is sherpa-onnx?
Sherpa-onnx is an open-source toolkit for offline speech and audio processing, built on next-generation Kaldi and ONNX Runtime. It provides speech-to-text, text-to-speech, speaker diarization, and other audio AI functions that run entirely locally without an internet connection. The project solves the need for privacy-preserving, low-latency audio processing on devices ranging from servers to embedded systems and mobile platforms.
Developers building voice-enabled applications for embedded systems, mobile apps (Android/iOS), IoT devices, or desktop software where offline operation, data privacy, or cross-platform support is critical. It's also suited for researchers and hobbyists working on multilingual speech projects.
Developers choose Sherpa-onnx for its unique combination of offline capability, extensive hardware support (including NPUs), and broad language/API coverage. Unlike cloud-dependent services, it offers complete data privacy and reduces latency, while its cross-platform nature and pre-built models accelerate development for diverse deployment targets.
Overview
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
Use Cases
Best For
Found a gem we're missing?
Open-Awesome is built by the community, for the community. Submit a project, suggest an awesome list, or help improve the catalog on GitHub.