---

language:

• zh

license: other

tags:

• wav2vec2
• gguf
• asr
• speech
• audio

library_name: wav2vec2.cpp

base_model:

• jonatasgrosman/wav2vec2-large-xlsr-53-chinese-zh-cn

---

wav2vec2-large-xlsr-53-chinese-zh-cn

GGUF quantizations of jonatasgrosman/wav2vec2-large-xlsr-53-chinese-zh-cn for use with wav2vec2.cpp.

Files

| Filename | Quant | Size | Precision |

|----------|-------|------|-----------|

| model_f16.gguf | F16 | 638 MB | ~fp16 (near-lossless) |

| model_q8_0.gguf | Q8_0 | 377 MB | 8-bit quantized linear layers |

| model_q4_0.gguf | Q4_0 | 224 MB | 4-bit quantized linear layers |

Parity vs HuggingFace Reference

Tested on 10 audio samples from FLEURS cmn_hans_cn.

| Quant | Mean CER (vs HF) | Mean WER (vs HF) | Status |

|-------|-----------------|-----------------|--------|

| F16 | 0.50% | 25.00% | PASS |

| Q8_0 | 1.31% | 37.50% | PASS |

| Q4_0 | 6.96% | 87.50% | FAIL |

CER/WER measured against HF model output (not ground truth) to isolate quantization error.

Usage

C++ CLI

# Clone and build
git clone --recursive https://github.com/liodon-ai/wav2vec2.cpp
cd wav2vec2.cpp && mkdir build && cd build && cmake .. && make -j

# Download and run
huggingface-cli download liodon-ai/wav2vec2-large-xlsr-53-chinese-zh-cn-GGUF model_q8_0.gguf --local-dir .
./wav2vec2-cli -m model_q8_0.gguf -f audio.wav

# With word timestamps
./wav2vec2-cli -m model_q8_0.gguf -f audio.wav -w

# SRT subtitle output
./wav2vec2-cli -m model_q8_0.gguf -f audio.wav --format srt > output.srt

Python (via subprocess)

import subprocess, json
result = subprocess.run(
    ['./wav2vec2-cli', '-m', 'model_q8_0.gguf', '-f', 'audio.wav', '--format', 'json'],
    capture_output=True, text=True
)
data = json.loads(result.stdout)
print(data['transcript'])

Source Model

• Original: jonatasgrosman/wav2vec2-large-xlsr-53-chinese-zh-cn
• Architecture: XLS-R
• Language: zh
• Task: Automatic Speech Recognition

Conversion

Converted using wav2vec2.cpp scripts/convert_to_gguf.py.

python scripts/convert_to_gguf.py jonatasgrosman/wav2vec2-large-xlsr-53-chinese-zh-cn model_f16.gguf --dtype f16
python scripts/convert_to_gguf.py jonatasgrosman/wav2vec2-large-xlsr-53-chinese-zh-cn model_q8_0.gguf --dtype q8_0
python scripts/convert_to_gguf.py jonatasgrosman/wav2vec2-large-xlsr-53-chinese-zh-cn model_q4_0.gguf --dtype q4_0

License

Follows the license of the source model.