Starstruk Logo MarkStarstruk

v1.12.12

k2-fsa/sherpa-onnxv1.12.12Sep 10, 2025by github-actions[bot]

AI Summary

This release focuses on fixing build issues for RISC-V and Android ARMv8l architectures, alongside improvements to Cantonese VITS text-to-speech models. It introduces new support for streaming Russian ASR models via the T-one project and expands language bindings for various non-streaming models.

Key Highlights

  • Fixed build compatibility for RISC-V architecture and CMake sub-project integration.
  • Resolved Android ARMv8l system call exceptions and disabled loading native libraries from JAR files.
  • Added T-one streaming Russian ASR models with support for C++, Python, and various language bindings.
  • Exported KittenTTS mini v0.1 and fixed Cantonese VITS TTS output issues.
  • Added support for Yue (Cantonese) ASR models and non-streaming Wenet CTC models.

New Features

  • T-one streaming Russian ASR models (C++, Python, and bindings)
  • KittenTTS mini v0.1 TTS model
  • Yue (Cantonese) u2pp_conformer_yue ASR model
  • Wenet non-streaming CTC models (various language bindings)
  • Sense-voice models

Full Release Notes

## What's Changed
* Fix building for risc-v by @csukuangfj in https://github.com/k2-fsa/sherpa-onnx/pull/2549
* Fix using sherpa-onnx as a cmake sub-project. by @csukuangfj in https://github.com/k2-fsa/sherpa-onnx/pull/2550
* Update kaldifst and kaldi-decoder by @csukuangfj in https://github.com/k2-fsa/sherpa-onnx/pull/2551
* Bug fix-修复armv8l架构下的Android系统调用java-api时启动异常 by @fengin in https://github.com/k2-fsa/sherpa-onnx/pull/2556
* Disable loading libs from jar on Android. by @csukuangfj in https://github.com/k2-fsa/sherpa-onnx/pull/2557
* Fix cantonese vits tts by @csukuangfj in https://github.com/k2-fsa/sherpa-onnx/pull/2558
* Avoid appending blanks for Cantonese vits tts. by @csukuangfj in https://github.com/k2-fsa/sherpa-onnx/pull/2559
* Add hint for loading model files from SD card on Android. by @csukuangfj in https://github.com/k2-fsa/sherpa-onnx/pull/2564
* Update README to include https://github.com/Mentra-Community/MentraOS by @csukuangfj in https://github.com/k2-fsa/sherpa-onnx/pull/2565
* Export models from https://github.com/voicekit-team/T-one to sherpa-onnx by @csukuangfj in https://github.com/k2-fsa/sherpa-onnx/pull/2571
* Add C++ and Python support for T-one streaming Russian ASR models by @csukuangfj in https://github.com/k2-fsa/sherpa-onnx/pull/2575
* Add various language bindings for streaming T-one Russian ASR models by @csukuangfj in https://github.com/k2-fsa/sherpa-onnx/pull/2576
* Fix the missing online punctuation in android aar by @csukuangfj in https://github.com/k2-fsa/sherpa-onnx/pull/2577
* Export KittenTTS mini v0.1 to sherpa-onnx by @csukuangfj in https://github.com/k2-fsa/sherpa-onnx/pull/2578
* Upload new sense-voice models by @csukuangfj in https://github.com/k2-fsa/sherpa-onnx/pull/2580
* Export ASLP-lab/WSYue-ASR/tree/main/u2pp_conformer_yue to sherpa-onnx by @csukuangfj in https://github.com/k2-fsa/sherpa-onnx/pull/2582
* Add various languge bindings for Wenet non-streaming CTC models by @csukuangfj in https://github.com/k2-fsa/sherpa-onnx/pull/2584
* Release v1.12.12 by @csukuangfj in https://github.com/k2-fsa/sherpa-onnx/pull/2586

## New Contributors
* @fengin made their first contribution in https://github.com/k2-fsa/sherpa-onnx/pull/2556

**Full Changelog**: https://github.com/k2-fsa/sherpa-onnx/compare/v1.12.11...v1.12.12