sherpa-onnx Releases
152 releases of k2-fsa/sherpa-onnx
- v1.12.8
Introduces KittenTTS with multi-language API support and adds NeMo CTC model support.
Aug 8, 2025
- v2.0.0
Refactors BreezeApp into engine and client components.
Aug 5, 2025
- v1.12.7
Adds NeMo ASR models for Portuguese and German, VAD+ASR support for WearOS, and more TTS models.
Jul 27, 2025
- v1.12.6
Adds Ten-VAD support with APIs for many languages and Silero-VAD v4.
Jul 12, 2025
- v1.12.5
Adds NeMo Canary models support with APIs for multiple languages and LODR support.
Jul 10, 2025
- v1.12.4
Adds non-streaming Zipformer CTC ASR models and improved Dart/Flutter Linux support.
Jul 4, 2025
- v1.12.3
Adds Zipformer ASR models with Whisper features.
Jun 27, 2025
- v1.12.2
This release focuses on fixing Windows CI, adding Java 24 support, and introducing new Python APIs for source separation and TTS export capabilities.
Jun 24, 2025
- V0.9.0
This release describes a major update to an autonomous agent application focusing on database implementations, static typing, and AI-assisted coding workflows.
Jun 9, 2025
- v1.12.1
This release addresses RKNN wheel build issues and expands source separation capabilities by integrating Spleeter and UVR models with new C++ runtimes.
Jun 3, 2025
- source-separation-models
This release provides pre-converted audio samples and model weights for Spleeter and UVR source separation models.
May 22, 2025
- v1.0.6
This release fixes a configuration issue with LLM settings in the BreezeApp.
May 16, 2025
- v1.12.0
This major feature release adds real-time speech recognition capabilities for SenseVoice, new C++ and Android examples, and WebSocket updates.
May 15, 2025
- v1.0.5
This release fixes UI issues related to screen rotation and LLM display in the BreezeApp.
May 15, 2025
- v1.0.4
This release enhances the UI for TTS triggering and adds an LLM model configuration page in the BreezeApp.
May 14, 2025
- v1.11.5
This release introduces support for the Parakeet TDT model and fixes a normalization issue in feature extraction.
May 8, 2025
- v1.11.4
This comprehensive release adds homophone replacer support across multiple programming languages (C++, Python, Java, JS, etc.) and new ASR models like GigaAM.
May 1, 2025
- hr-files
This release provides a utility to generate a replace.fst file for mapping Chinese Pinyin to Hanzi.
Apr 27, 2025
- v0.9.1
Apr 24, 2025
- v1.11.3
This release focuses on bug fixes for TTS and VAD models, adds RKNN and DirectML support, and introduces comprehensive multi-language APIs for Dolphin CTC models.
Apr 3, 2025