Downloads · 30 days
0
luigi000/fun-asr-mlt-nano-onnx-int8
fun-asr-mlt-nano-onnx-int8 is a automatic speech recognition model from luigi000. Use it when you need speech turned into text. It is set up for sherpa-onnx. The card lists the license as apache-2.0.
int8 ONNX export of FunAudioLLM/Fun-ASR-MLT-Nano-2512 (≈0.8B, SenseVoice encoder + Qwen3-0.6B LLM decoder), packaged for sherpa-onnx's OfflineRecognizer.fromfunasrnano. Runs on CPU with no PyTorch/transformers at infe…
Downloads · 30 days
0
Access
Public
Updated Jul 9, 2026
Repo size
1 GB
Likes
0
Public
Click a slice to open those files.
.onnx994 MB · 98%
From the Hugging Face model README
int8 ONNX export of FunAudioLLM/Fun-ASR-MLT-Nano-2512
(≈0.8B, SenseVoice encoder + Qwen3-0.6B LLM decoder), packaged for
sherpa-onnx's OfflineRecognizer.from_funasr_nano.
Runs on CPU with no PyTorch/transformers at inference. Covers the 31 languages of the
multilingual Fun-ASR-Nano, including Chinese, English, Vietnamese, Indonesian, Thai, Malay,
Korean, and more.
| file | size | purpose |
|---|---|---|
encoder_adaptor.int8.onnx | ~238 MB | SenseVoice audio encoder + adaptor |
embedding.int8.onnx | ~156 MB | token embedding |
llm.int8.onnx | ~600 MB | Qwen3-0.6B decoder |
tokenizer/ | — | Qwen3-0.6B tokenizer (vocab / merges / config) |
Total ≈ 994 MB.
import sherpa_onnx # pip install sherpa-onnx>=1.13.3
rec = sherpa_onnx.OfflineRecognizer.from_funasr_nano(
encoder_adaptor="encoder_adaptor.int8.onnx",
llm="llm.int8.onnx",
embedding="embedding.int8.onnx",
tokenizer="tokenizer",
num_threads=4,
sample_rate=16000,
provider="cpu",
language="", # "" = auto language id; or a Chinese language name, e.g. 英文 / 印尼语 / 越南语
max_new_tokens=512,
)
s = rec.create_stream()
s.accept_waveform(16000, audio_float32_mono_16k) # numpy float32 in [-1, 1]
rec.decode_stream(s)
print(s.result.text)
language="" runs the model's built-in language identification (recommended for mixed /
auto-detect audio). To force a language, pass the Chinese language name the model was
trained on (e.g. 英文 English, 印尼语 Indonesian, 越南语 Vietnamese) — the string is
interpolated into a Chinese instruction prompt, so English names do not work.
model.pt was exported to ONNX (encoder-adaptor / embedding / LLM) and dynamic int8-quantized; the tokenizer is copied unchanged from the base checkpoint. No fine-tuning or weight changes beyond quantization.Apache-2.0, inherited from the base model. See the LICENSE and NOTICE files in this repo.
This is a redistributed derivative work — attribution above.