Fun-ASR-Nano-2512: transcribe.cpp GGUF

GGUF conversions of FunAudioLLM/Fun-ASR-Nano-2512 for use with transcribe.cpp.

Ported from upstream commit a7088d6, pinned 2026-05-06. Validated against the FunASR reference at transcribe.cpp commit f094d28 on 2026-05-06.

Offline speech-to-text in Chinese, English, and Japanese, plus 7 Chinese dialects (Wu, Cantonese, Min, Hakka, Gan, Xiang, Jin) and 26 regional Mandarin accents. ~800M trainable parameters wrapping a frozen SenseVoiceEncoderSmall (50 SAN-M main blocks + 20 transformer blocks), a 2-layer audio adaptor (512 → 1024), and a bundled Qwen3-0.6B LLM (28 layers, 16/8 GQA, BF16) that produces the transcript autoregressively. Takes a 16 kHz mono WAV and emits text. Not a streaming model, no translation, no built-in long-form chunking, no timestamps. ITN (inverse text normalization) is supported by the model and exposed via the --itn CLI flag and transcribe_funasr_nano_params { use_itn } in the library API.

Downloads

WER on the full LibriSpeech test-clean split (2,620 utterances). Figures without a commit were published before provenance was recorded.

Greedy LLM decoding via the bundled Qwen3-0.6B head. Publisher reports 1.76% on this split (model card "Open-Source Dataset Performance" table). Our FunASR 1.3.1 reference run scores 1.79% (95% CI [1.63%, 1.95%]), within bootstrap noise of the publisher's number. transcribe.cpp's BF16 port matches that baseline within -0.01 percentage-points. LibriSpeech is an English-only benchmark; Chinese (AISHELL-1, WenetSpeech) and Japanese (CommonVoice JA) are the recommended complementary checks.

Usage

Build transcribe.cpp from source:

git clone git@github.com:handy-computer/transcribe.cpp.git
cd transcribe.cpp
cmake -B build && cmake --build build

Run on a 16 kHz mono WAV:

build/bin/transcribe-cli \
  -m Fun-ASR-Nano-2512-Q8_0.gguf \
  input.wav

If your audio isn't already 16 kHz mono WAV, convert it first:

ffmpeg -i input.mp3 -ar 16000 -ac 1 output.wav

See the transcribe.cpp model page for performance numbers, numerical validation, and reproduction steps.

License

Inherited from the base model: FunASR Model Open Source License Agreement v1.1. See the upstream model card for full terms.

Original Model Card

The section below is reproduced from FunAudioLLM/Fun-ASR-Nano-2512 at commit a7088d6 for offline reference. The upstream card is the authoritative source.

Fun-ASR

「简体中文」|「English」

Fun-ASR is an end-to-end speech recognition large model launched by Tongyi Lab. It is trained on tens of millions of hours of real speech data, possessing powerful contextual understanding capabilities and industry adaptability. It supports low-latency real-time transcription and covers 31 languages. It excels in vertical domains such as education and finance, accurately recognizing professional terminology and industry expressions, effectively addressing challenges like "hallucination" generation and language confusion, achieving "clear hearing, understanding meaning, and accurate writing."

Model Repository: modelscope, huggingface

Online Experience: ModelScope Community Space, huggingface space

What's New 🔥

• 2025/12: Fun-ASR-Nano-2512 is an end-to-end speech recognition large model trained on tens of millions of hours real speech data. It supports low-latency real-time transcription and covers 31 languages.
• 2024/7: FunASR is a fundamental speech recognition toolkit that offers a variety of features, including speech recognition (ASR), Voice Activity Detection (VAD), Punctuation Restoration, Language Models, Speaker Verification, Speaker Diarization and multi-talker ASR.

Core Features 🎯

Fun-ASR focuses on high-precision speech recognition, multi-language support, and industry customization capabilities

• Far-field High-noise Recognition: Deeply optimized for far-distance sound pickup and high-noise scenarios (such as conference rooms, in-vehicle environments, industrial sites, etc.), improving recognition accuracy to 93%.
• Chinese Dialects and Regional Accents:
  • Supports 7 major dialects: Wu, Cantonese, Min, Hakka, Gan, Xiang, Jin
  • Covers 26 regional accents: including Henan, Shaanxi, Hubei, Sichuan, Chongqing, Yunnan, Guizhou, Guangdong, Guangxi and more than 20 other regions
• Multi-language Free Speech: Supports recognition of 31 languages, with focused optimization on East and Southeast Asian languages, supporting free language switching and mixed recognition.
• Music Background Lyric Recognition: Enhanced speech recognition performance under music background interference, supporting accurate recognition of lyric content in songs.

Environment Setup 🐍

git clone https://github.com/FunAudioLLM/Fun-ASR.git
cd Fun-ASR
pip install -r requirements.txt

TODO

• Support returning timestamps
• Support speaker diarization
• Support model training

Usage 🛠️

Inference

Using funasr for inference

from funasr import AutoModel


def main():
    model_dir = "FunAudioLLM/Fun-ASR-Nano-2512"
    model = AutoModel(
        model=model_dir,
        trust_remote_code=True,
        remote_code="./model.py",
        device="cuda:0",
    )

    wav_path = f"{model.model_path}/example/zh.mp3"
    res = model.generate(
        input=[wav_path],
        cache={},
        batch_size=1,
        hotwords=["开放时间"],
        # 中文、英文、日文 for Fun-ASR-Nano-2512
        # 中文、英文、粤语、日文、韩文、越南语、印尼语、泰语、马来语、菲律宾语、阿拉伯语、
        # 印地语、保加利亚语、克罗地亚语、捷克语、丹麦语、荷兰语、爱沙尼亚语、芬兰语、希腊语、
        # 匈牙利语、爱尔兰语、拉脱维亚语、立陶宛语、马耳他语、波兰语、葡萄牙语、罗马尼亚语、
        # 斯洛伐克语、斯洛文尼亚语、瑞典语 for Fun-ASR-MLT-Nano-2512
        language="中文",
        itn=True, # or False
    )
    text = res[0]["text"]
    print(text)

    model = AutoModel(
        model=model_dir,
        trust_remote_code=True,
        vad_model="fsmn-vad",
        vad_kwargs={"max_single_segment_time": 30000},
        remote_code="./model.py",
        device="cuda:0",
    )
    res = model.generate(input=[wav_path], cache={}, batch_size=1)
    text = res[0]["text"]
    print(text)


if __name__ == "__main__":
    main()

Direct Inference

from model import FunASRNano


def main():
    model_dir = "FunAudioLLM/Fun-ASR-Nano-2512"
    m, kwargs = FunASRNano.from_pretrained(model=model_dir, device="cuda:0")
    m.eval()

    wav_path = f"{kwargs['model_path']}/example/zh.mp3"
    res = m.inference(data_in=[wav_path], **kwargs)
    text = res[0][0]["text"]
    print(text)


if __name__ == "__main__":
    main()

• model_dir: Model name or local disk model path.
• trust_remote_code: Whether to trust remote code for loading custom model implementations.
• remote_code: Specify the location of specific model code (e.g., model.py in the current directory), supporting both absolute and relative paths.
• device: Specify the device to use, such as "cuda:0" or "cpu".

Performance 📝

We evaluated Fun-ASR against other state-of-the-art models on open-source benchmarks, Chinese dialect datasets, and industry-specific test sets. The results demonstrate that Fun-ASR achieves superior performance across various scenarios.

1. Open-Source Dataset Performance (WER %)

Note: Seed-ASR* results are evaluated using the official API on volcengine; GLM-ASR-nano* results are evaluated using the open-source checkpoint.

2. Industry Dataset Performance (WER %)

Citations

@article{an2025fun,
  title={Fun-ASR Technical Report},
  author={An, Keyu and Chen, Yanni and Deng, Chong and Gao, Changfeng and Gao, Zhifu and Gong, Bo and Li, Xiangang and Li, Yabin and Lv, Xiang and Ji, Yunjie and others},
  journal={arXiv preprint arXiv:2509.12508},
  year={2025}
}