Starstruk Logo MarkStarstruk

v1.11.3

k2-fsa/sherpa-onnxv1.11.3Apr 3, 2025by csukuangfj

AI Summary

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.

Key Highlights

  • RKNN support for silero_vad v4 and Dolphin CTC models
  • DirectML support for Windows/Linux
  • Multi-language API support (C++, Python, Java, Swift, Go, etc.) for Dolphin CTC
  • Fixes for VITS dict and Kokoro TTS length scale

New Features

  • RKNN runtime for silero_vad v4
  • Dolphin CTC support across 12+ languages
  • C++ API for VAD
  • DirectML support

Full Release Notes

## What's Changed
* fix vits dict dir config by @amutu in https://github.com/k2-fsa/sherpa-onnx/pull/2036
* fix case by @amutu in https://github.com/k2-fsa/sherpa-onnx/pull/2037
* Fix building wheels for RKNN by @csukuangfj in https://github.com/k2-fsa/sherpa-onnx/pull/2041
* 缩放因子应该是32767？ by @yourengod in https://github.com/k2-fsa/sherpa-onnx/pull/2056
* Fix length scale for kokoro tts by @csukuangfj in https://github.com/k2-fsa/sherpa-onnx/pull/2060
* Allow building repository as CMake subdirectory by @niansa in https://github.com/k2-fsa/sherpa-onnx/pull/2059
* Export silero_vad v4 to RKNN by @csukuangfj in https://github.com/k2-fsa/sherpa-onnx/pull/2067
* 修复 DirectML 支持 by @endink in https://github.com/k2-fsa/sherpa-onnx/pull/2066
* Fix building aar to include speech denoiser by @csukuangfj in https://github.com/k2-fsa/sherpa-onnx/pull/2069
* Add CXX API for VAD by @csukuangfj in https://github.com/k2-fsa/sherpa-onnx/pull/2077
* Add C++ runtime for silero_vad with RKNN by @csukuangfj in https://github.com/k2-fsa/sherpa-onnx/pull/2078
* Refactor rknn code by @csukuangfj in https://github.com/k2-fsa/sherpa-onnx/pull/2079
* Fix building for android by @csukuangfj in https://github.com/k2-fsa/sherpa-onnx/pull/2081
* Add C++ and Python API for Dolphin CTC models by @csukuangfj in https://github.com/k2-fsa/sherpa-onnx/pull/2085
* Add Kotlin and Java API for Dolphin CTC models by @csukuangfj in https://github.com/k2-fsa/sherpa-onnx/pull/2086
* Add C and CXX API for Dolphin CTC models by @csukuangfj in https://github.com/k2-fsa/sherpa-onnx/pull/2088
* Preserve more context after endpointing in transducer by @vsd-vector in https://github.com/k2-fsa/sherpa-onnx/pull/2061
* Add C# API for Dolphin CTC models by @csukuangfj in https://github.com/k2-fsa/sherpa-onnx/pull/2089
* Add Go API for Dolphin CTC models by @csukuangfj in https://github.com/k2-fsa/sherpa-onnx/pull/2090
* Add Swift API for Dolphin CTC models by @csukuangfj in https://github.com/k2-fsa/sherpa-onnx/pull/2091
* Add Javascript (WebAssembly) API for Dolphin CTC models by @csukuangfj in https://github.com/k2-fsa/sherpa-onnx/pull/2093
* Add Javascript (node-addon) API for Dolphin CTC models by @csukuangfj in https://github.com/k2-fsa/sherpa-onnx/pull/2094
* Add Dart API for Dolphin CTC models by @csukuangfj in https://github.com/k2-fsa/sherpa-onnx/pull/2095
* Add Pascal API for Dolphin CTC models by @csukuangfj in https://github.com/k2-fsa/sherpa-onnx/pull/2096
* Release v1.11.3 by @csukuangfj in https://github.com/k2-fsa/sherpa-onnx/pull/2097

## New Contributors
* @amutu made their first contribution in https://github.com/k2-fsa/sherpa-onnx/pull/2036
* @yourengod made their first contribution in https://github.com/k2-fsa/sherpa-onnx/pull/2056
* @niansa made their first contribution in https://github.com/k2-fsa/sherpa-onnx/pull/2059

**Full Changelog**: https://github.com/k2-fsa/sherpa-onnx/compare/v1.11.2...v1.11.3