README.md

---
license: apache-2.0
task_categories:
- text-to-speech
- automatic-speech-recognition
language:
- ja
size_categories:
- 100K<n<1M
---

# This dataset contains NSFW content. Please use with caution.


## Changelog
    
```
- 2024-04-27 Audio classified using [Japanese Ero Voice Classifier](https://huggingface.co/spaces/litagin/Japanese-Ero-Voice-Classifier)， Results are uploaded in transcript_20240427.csv
- 2024-04-09 Initial release
```

This dataset is only permitted for use for data analysis (such as machine learning) purposes. Any use for purposes other than those specified is prohibited.

## Dataset Language
Japanese

## Dataset Information

This dataset is a medium-quality dataset of character acting speech audio by Japanese dou-jin voice actors. The initial release contains **835736 files**, **2235.72 hours** of audio. The dataset is in Japanese. Audio files are in their original format, and the dataset is provided as-is. Most of the audio files are 44.1 kHz or 48 kHz, while some are 22.05 kHz. The dataset is provided with zip compression.
## Dataset Structure
```
├── transcript.csv
├── dataset
│   ├── (files)
```

## Ttranscript
The transcript file contains the transcript of the audio files in the dataset.  The format is:


```
filename, chara, duration, samplerate, transcript, SFW 
```
## Dataset Analysis

Duration:

![](./duration_distribution.png)

Classification Result:

[Japanese Ero Voice Classifier](https://huggingface.co/spaces/litagin/Japanese-Ero-Voice-Classifier)

sfw_count:  413432,  sfw_duration:  906.86h

nsfw_count: 422304, nsfw_duration: 1328.86h