---
title: Mandarin_Chinese_Scripted_Speech_Corpus
canonical_url: "https://www.modelscope.cn/datasets/Magic_Data/Mandarin_Chinese_Scripted_Speech_Corpus"
md_url: "https://www.modelscope.cn/datasets/Magic_Data/Mandarin_Chinese_Scripted_Speech_Corpus.md"
repository: Magic_Data/Mandarin_Chinese_Scripted_Speech_Corpus
chinese_name: "中文普通话朗读音频数据集"
last_updated: 2024-07-16
license: "Apache License 2.0"
storage_size: "50 GB"
downloads: 1031
stars: 1
---

# Mandarin_Chinese_Scripted_Speech_Corpus

> Mandarin_Chinese_Scripted_Speech_Corpus - Magic_Data 在 ModelScope 开源的数据集。下载方法 :modelscope-code[]{type="sdk"} :modelscope-code[]{type="git"}

Magic_Data/Mandarin_Chinese_Scripted_Speech_Corpus 是 ModelScope 魔搭社区上的数据集，存储大小 50 GB，采用 Apache License 2.0 许可。

- **Repository**: Magic_Data/Mandarin_Chinese_Scripted_Speech_Corpus
- **License**: Apache License 2.0
- **Storage size**: 50 GB
- **Downloads**: 1031
- **Stars**: 1
- **Last updated**: 2024-07-16

Source: https://www.modelscope.cn/datasets/Magic_Data/Mandarin_Chinese_Scripted_Speech_Corpus

---

## 目录结构

```
SPKINFO.txt
UTTERANCEINFO.txt
WAV-1/
 --SPEAKER_ID/
  --UTTRANS_ID.wav
```

### SPKINFO.txt
```txt
CHANNEL	SPEAKER_ID	GENDER	AGE	REGION,CITY	DEVICE
C0	14_3466	F	18	Hebei	Mobile phone
C0	14_3664	F	40	Guangdong	Mobile phone
C0	14_3694	M	23	Guangdong	Mobile phone
C0	14_3698	M	23	Guangdong	Mobile phone
```

### UTTERANCEINFO.txt
```txt
CHANNEL	UTTRANS_ID	SPEAKER_ID	PROMPT	TRANSCRIPTION
C0	14_3466_20170826171159.wav	14_3466		请语言播放小说
C0	14_3466_20170826171236.wav	14_3466		这里
C0	14_3466_20170826171323.wav	14_3466		全民唱吧
```

## 使用方法
```python
from modelscope import MsDataset

dataset = MsDataset.load("Magic_Data/Mandarin_Chinese_Scripted_Speech_Corpus", split='train')


# 逐条处理数据
for i, example in enumerate(dataset):
    print(example)
    if i==10:
        break
```


#### 下载方法 
:modelscope-code[]{type="sdk"}
:modelscope-code[]{type="git"}