---
title: speech_paraformer_large_asr_mtl-16k-common-vocab11666-pytorch
canonical_url: "https://www.modelscope.cn/models/dengcunqin/speech_paraformer_large_asr_mtl-16k-common-vocab11666-pytorch"
md_url: "https://www.modelscope.cn/models/dengcunqin/speech_paraformer_large_asr_mtl-16k-common-vocab11666-pytorch.md"
repository: dengcunqin/speech_paraformer_large_asr_mtl-16k-common-vocab11666-pytorch
chinese_name: "Paraformer语音识别-多方言-通用-16k-离线-large"
last_updated: 2024-12-11
license: "Apache License 2.0"
pipeline_tag: auto-speech-recognition
tasks:
  - auto-speech-recognition
model_type:
  - Non-autoregressive
library_name:
  - pytorch
frameworks:
  - pytorch
language:
  - cn
domain:
  - audio
downloads: 4516
stars: 10
tags:
  - FunASR
  - Paraformer
  - Alibaba
  - ICASSP2024
---

# speech_paraformer_large_asr_mtl-16k-common-vocab11666-pytorch

> speech_paraformer_large_asr_mtl-16k-common-vocab11666-pytorch - dengcunqin 在 ModelScope 开源的模型。Paraformer语音识别-多方言-通用-16k-离线-large

dengcunqin/speech_paraformer_large_asr_mtl-16k-common-vocab11666-pytorch 是 ModelScope 魔搭社区上的auto-speech-recognition模型，采用 Apache License 2.0 许可。

- **Repository**: dengcunqin/speech_paraformer_large_asr_mtl-16k-common-vocab11666-pytorch
- **License**: Apache License 2.0
- **Tasks**: auto-speech-recognition
- **Tags**: FunASR, Paraformer, Alibaba, ICASSP2024
- **Downloads**: 4516
- **Stars**: 10
- **Last updated**: 2024-12-11

Source: https://www.modelscope.cn/models/dengcunqin/speech_paraformer_large_asr_mtl-16k-common-vocab11666-pytorch

---

# 模型介绍
# 基于Paraformer large(iic/speech_paraformer-large_asr_nat-zh-cn-16k-common-vocab8404-pytorch)，更换vocab为11666，增加kespeech以及粤语部分字，通过在普通话1w小时、粤语200小时、英语1w小时+整个kespeech train音频数据集上进行训练，当前版本为训练0.1 epoch。语音数据混合合并训练，并且增加噪音，所以模型可无缝识别多语言，抗噪音能力较强。
# 训练loss如下图，可见loss下降空间仍很大，继续训练效果应该挺不错的。
<div align=center>
<img src="screenshot-1722838198936.png" width="700" />
</div>

## 测试代码
```python
from funasr import AutoModel

model = AutoModel(model="dengcunqin/speech_paraformer_large_asr_mtl-16k-common-vocab11666-pytorch",
                  model_revision="master"
                  )

wav_root_url="https://www.modelscope.cn/api/v1/models/dengcunqin/speech_paraformer_large_asr_mtl-16k-common-vocab11666-pytorch/repo?Revision=master&FilePath="
res = model.generate(input=wav_root_url+"example/asr_example.wav")
print(res)

res = model.generate(input=wav_root_url+"example/asr_example_普通话.wav")
print(res)

res = model.generate(input=wav_root_url+"example/asr_example_粤语.wav")
print(res)
```
### 微信交流
Funasr群找小邓

### 微信小程序
<img src="https://www.modelscope.cn/models/dengcunqin/speech_seaco_paraformer_large_asr_nat-zh-cantonese-en-16k-common-vocab11666-pytorch/resolve/master/%E5%BE%AE%E4%BF%A1%E5%B0%8F%E7%A8%8B%E5%BA%8F%E7%A0%81.jpg" alt="微信小程序"  width="500" />

## 相关论文以及引用信息

```BibTeX
@article{shi2023seaco,
  title={SeACo-Paraformer: A Non-Autoregressive ASR System with Flexible and Effective Hotword Customization Ability},
  author={Shi, Xian and Yang, Yexin and Li, Zerui and Zhang, Shiliang},
  journal={arXiv preprint arXiv:2308.03266 (accepted by ICASSP2024)},
  year={2023}
}
```