Upload README.md with huggingface_hub

0cbba64 verified 3 months ago

1.52 kB

license: apache-2.0
language:
  - zh
  - en
library_name: gguf
tags:
  - automatic-speech-recognition
  - asr
  - sensevoice
  - funasr
  - llama.cpp
  - ggml
  - cpu
  - chinese
pipeline_tag: automatic-speech-recognition

SenseVoiceSmall · GGUF (FunASR llama.cpp runtime)

GGUF build of SenseVoiceSmall (SAN-M encoder + CTC) for the zero-Python, CPU/edge FunASR llama.cpp runtime — multilingual ASR with language / emotion / event tags, ~20× real-time on CPU.

Files

file	size	notes
`sensevoice-small-f16.gguf`	470 MB	recommended (f16 matmul weights)
`sensevoice-small.gguf`	936 MB	f32 reference

Usage

# 1. get the VAD too (for long audio): huggingface-cli download FunAudioLLM/fsmn-vad-GGUF
llama-funasr-sensevoice -m sensevoice-small-f16.gguf -a audio.wav --vad fsmn-vad.gguf > ids.txt

On CPU (8 threads) this reaches 8.01 % CER on the 184-clip Mandarin benchmark — vs whisper.cpp 22–31 %. See the benchmark.

FunAudioLLM
/

SenseVoiceSmall-GGUF

SenseVoiceSmall · GGUF (FunASR llama.cpp runtime)

Files

Usage

Links