runtime-llamacpp-v0.1.1

googleworkspace/cliruntime-llamacpp-v0.1.1Jun 21, 2026by LauraGPT

AI Summary

Introduces a self-contained runtime v0.1.1 for running SenseVoice, Paraformer, and Fun-ASR-Nano locally without Python or build dependencies, featuring built-in VAD and strong performance on Chinese and Cantonese.

Key Highlights

Prebuilt, self-contained binaries for Linux, macOS, and Windows
No Python, build, or torch dependencies required
Supports SenseVoice, Paraformer, and Fun-ASR-Nano models
Built-in FSMN-VAD for voice activity detection
Strong accuracy on Chinese and Cantonese (8.01% CER)

New Features

Local execution of SenseVoice ASR using llama.cpp runtime
On-device ASR with whisper.cpp-style inference
Built-in FSMN-VAD
Cross-platform binary support (x64, arm64)
Zero-dependency execution environment

Full Release Notes

Prebuilt, self-contained binaries to run **SenseVoice** (and Paraformer / Fun-ASR-Nano) locally with the FunASR llama.cpp / GGUF runtime — built-in FSMN-VAD, whisper.cpp-style on-device ASR, strong on Chinese & Cantonese.

```bash
bash download-funasr-model.sh sensevoice ./gguf
llama-funasr-sensevoice -m ./gguf/SenseVoiceSmall-f16.gguf --vad ./gguf/fsmn-vad.gguf -a audio.wav
```

No Python, no build, no torch. Binaries for Linux (x64/arm64), macOS (arm64), Windows (x64). Docs: `runtime/llama.cpp/README.md`. CER (micro-avg): SenseVoice 8.01% vs whisper.cpp small 22%.