fixbug hotwords

游雁

2024-06-24 1596f6f414f6f41da66506debb1dff19fffeb3ec

 README_zh.md

@@ -2,7 +2,11 @@

(简体中文|[English](./README.md))

# FunASR: A Fundamental End-to-End Speech Recognition Toolkit


[![SVG Banners](https://svg-banners.vercel.app/api?type=origin&text1=FunASR🤠&text2=💖%20A%20Fundamental%20End-to-End%20Speech%20Recognition%20Toolkit&width=800&height=210)](https://github.com/Akshay090/svg-banners)

[//]: # (# FunASR: A Fundamental End-to-End Speech Recognition Toolkit)

[![PyPI](https://img.shields.io/pypi/v/funasr)](https://pypi.org/project/funasr/)

@@ -60,18 +64,28 @@
<a name="安装教程"></a>
## 安装教程

- 安装funasr之前，确保已经安装了下面依赖环境:
```text
python>=3.8
torch>=1.13
torchaudio
```

- pip安装
```shell
pip3 install -U funasr
```
或者从源代码安装

- 或者从源代码安装
``` sh
git clone https://github.com/alibaba/FunASR.git && cd FunASR
pip3 install -e ./
```
如果需要使用工业预训练模型，安装modelscope（可选）

如果需要使用工业预训练模型，安装modelscope与huggingface_hub（可选）

```shell
pip3 install -U modelscope
pip3 install -U modelscope huggingface huggingface_hub
```

## 模型仓库
@@ -87,7 +101,7 @@
| paraformer-zh-streaming <br> ( [⭐](https://modelscope.cn/models/damo/speech_paraformer-large_asr_nat-zh-cn-16k-common-vocab8404-online/summary) [🤗](https://huggingface.co/funasr/paraformer-zh-streaming) ) |      语音识别，实时       |   60000小时，中文   |  220M  |
|         paraformer-en <br> ( [⭐](https://www.modelscope.cn/models/damo/speech_paraformer-large-vad-punc_asr_nat-en-16k-common-vocab10020/summary) [🤗](https://huggingface.co/funasr/paraformer-en) )         |      语音识别，非实时      |   50000小时，英文   |  220M  |
|                      conformer-en <br> ( [⭐](https://modelscope.cn/models/damo/speech_conformer_asr-en-16k-vocab4199-pytorch/summary) [🤗](https://huggingface.co/funasr/conformer-en) )                      |      语音识别，非实时      |   50000小时，英文   |  220M  |
|                        ct-punc <br> ( [⭐](https://modelscope.cn/models/damo/punc_ct-transformer_cn-en-common-vocab471067-large/summary) [🤗](https://huggingface.co/funasr/ct-punc) )                         |        标点恢复        |   100M，中文与英文   |  1.1B  | 
|                        ct-punc <br> ( [⭐](https://modelscope.cn/models/damo/punc_ct-transformer_cn-en-common-vocab471067-large/summary) [🤗](https://huggingface.co/funasr/ct-punc) )                         |        标点恢复        |   100M，中文与英文   |  290M  | 
|                            fsmn-vad <br> ( [⭐](https://modelscope.cn/models/damo/speech_fsmn_vad_zh-cn-16k-common-pytorch/summary) [🤗](https://huggingface.co/funasr/fsmn-vad) )                             |     语音端点检测，实时      |  5000小时，中文与英文  |  0.4M  | 
|                              fa-zh <br> ( [⭐](https://modelscope.cn/models/damo/speech_timestamp_prediction-v1-16k-offline/summary) [🤗](https://huggingface.co/funasr/fa-zh) )                               |      字级别时间戳预测      |   50000小时，中文   |  38M   |
|                                 cam++ <br> ( [⭐](https://modelscope.cn/models/iic/speech_campplus_sv_zh-cn_16k-common/summary) [🤗](https://huggingface.co/funasr/campplus) )                                 |      说话人确认/分割      |     5000小时     |  7.2M  | 
@@ -153,6 +167,8 @@

注：`chunk_size`为流式延时配置，`[0,10,5]`表示上屏实时出字粒度为`10*60=600ms`，未来信息为`5*60=300ms`。每次推理输入为`600ms`（采样点数为`16000*0.6=960`），输出为对应文字，最后一个语音片段输入需要设置`is_final=True`来强制输出最后一个字。

<details><summary>更多例子</summary>

### 语音端点检测（非实时）
```python
from funasr import AutoModel
@@ -216,9 +232,24 @@
res = model.generate(input=(wav_file, text_file), data_type=("sound", "text"))
print(res)
```

### 情感识别
```python
from funasr import AutoModel

model = AutoModel(model="emotion2vec_plus_large")

wav_file = f"{model.model_path}/example/test.wav"

res = model.generate(wav_file, output_dir="./outputs", granularity="utterance", extract_embedding=False)
print(res)
```

更详细（[教程文档](docs/tutorial/README_zh.md)），
更多（[模型示例](https://github.com/alibaba-damo-academy/FunASR/tree/main/examples/industrial_data_pretraining)）

</details>

## 导出ONNX
### 从命令行导出
```shell