OpenDialog

# OpenDialog

OpenDialog is a 6.8k hours spoken dialogue dataset, introduced in the paper [ZipVoice-Dialog: Non-Autoregressive Spoken Dialogue Generation with Flow Matching](https://arxiv.org/abs/2507.09318).

- **Paper:** [https://arxiv.org/abs/2507.09318](https://arxiv.org/abs/2507.09318)
- **GitHub:** [https://github.com/k2-fsa/ZipVoice](https://github.com/k2-fsa/ZipVoice)
- **Project Page:** [https://zipvoice-dialog.github.io](https://zipvoice-dialog.github.io)

OpenDialog is the first large-scale (6.8k hours) open-source spoken dialogue dataset derived from in-the-wild speech data. It consists of:
- **English data:** 5074 hours
- **Chinese data:** 1759 hours

This dataset is also available at [ModelScope](https://www.modelscope.cn/datasets/k2-fsa/OpenDialog) (more friendly for users from China mainland).

## Citation

```bibtex
@article{zhu2025zipvoicedialog,
title={ZipVoice-Dialog: Non-Autoregressive Spoken Dialogue Generation with Flow Matching},
author={Zhu, Han and Kang, Wei and Guo, Liyong and Yao, Zengwei and Kuang, Fangjun and Zhuang, Weiji and Li, Zhaoqing and Han, Zhifeng and Zhang, Dong and Zhang, Xin and Song, Xingchen and Lin, Long and Povey, Daniel},
journal={arXiv preprint arXiv:2507.09318},
year={2025}
}
```

# OpenDialog
OpenDialog 是一款总时长达6.8千小时的口语对话数据集，相关研究成果刊载于论文《ZipVoice-Dialog》（https://arxiv.org/abs/2507.09318），详细信息请参阅该论文。

该数据集包含1759小时的中文语料与5074小时的英文语料。

搜集汇总

背景与挑战

背景概述

OpenDialog是一个大规模开源口语对话数据集，总时长6.8k小时，包含5074小时英语和1759小时中文数据，源自野外语音。该数据集基于ZipVoice-Dialog论文，并在ModelScope平台提供访问。