v1.4.0

moeru-ai/airiv1.4.0Jul 31, 2026by github-actions[bot]

AI Summary

This release focuses on the FunASR Python package and runtime support, fixing common user errors, restoring specific normalization features, and providing self-contained llama.cpp runtimes for various platforms.

Key Highlights

  • Rejects the common `vda_model` misspelling to prevent silent VAD failures.
  • Restores SenseVoice English spelling normalization by packaging `english.json`.
  • Restores RWKV-BAT CUDA kernel compilation by packaging source files.
  • Includes self-contained llama.cpp/GGUF runtimes for multiple operating systems and hardware configurations.

New Features

  • Self-contained llama.cpp runtimes for SenseVoice, Paraformer, and Fun-ASR-Nano.
  • Finalized legacy WebSocket file-input session termination logic.
  • Support for RWKV-BAT CUDA kernel compilation in installed wheels.

Full Release Notes

## FunASR 1.4.0

Install or upgrade the Python package:

```bash
pip install -U funasr==1.4.0
```

### Python package

- Rejects the common `vda_model` misspelling immediately and points users to the supported `vad_model` argument, instead of silently running without VAD.
- Restores SenseVoice English spelling normalization in installed wheels by packaging `english.json`.
- Restores RWKV-BAT CUDA kernel compilation from installed wheels by packaging the encoder and decoder `.cpp` and `.cu` sources.

### Tagged source

- Finalizes legacy WebSocket file-input sessions correctly when the client sends the end-control message. This runtime is distributed in the tagged GitHub source and is not part of the PyPI wheel.

### Prebuilt llama.cpp runtime

The release assets include self-contained FunASR llama.cpp / GGUF runtimes for SenseVoice, Paraformer, and Fun-ASR-Nano:

- Linux arm64
- Linux x64
- Linux x64 AVX2
- Linux x64 Vulkan
- macOS arm64
- Windows x64
- Windows x64 AVX2
- Windows x64 CUDA
- Windows x64 Vulkan

Use `SHA256SUMS-v1.4.0` to verify every Python and runtime artifact after download.

**Full Changelog:** https://github.com/modelscope/FunASR/compare/v1.3.30...v1.4.0