TTS_eval_datasets

# TTS evaluation datasets

This repository contains three testsets for zero-shot TTS models:

- **dialog_testset**: Chinese and English testsets for spoken dialogue generation models, introduced in paper [ZipVoice-Dialog](https://arxiv.org/abs/2507.09318).
- **librispeech_pc_testset**: English testset for zero-shot TTS models, introduced in paper [F5-TTS](https://arxiv.org/abs/2410.06885).
- **seedtts_testset**: Chinese and English testsets for zero-shot TTS models, introduced in paper [Seed-TTS](https://arxiv.org/abs/2406.02430).
- **minimax_multilingual_24**: 24-language testset for zero-shot TTS models, introduced in paper [MiniMax-Speech](https://arxiv.org/abs/2505.07916).
- **fleurs_multilingual_102**: 102-language testset for zero-shot TTS models, introduced in paper [OmniVoice](https://arxiv.org/abs/2604.00688).

# 文本转语音（Text-to-Speech，TTS）评测数据集

本仓库收录三款适配零样本（Zero-shot）文本转语音模型的测试集：
- **dialog_testset（对话测试集）**：面向口语对话生成模型的中英双语测试集，由论文《ZipVoice-Dialog》（https://arxiv.org/abs/2507.09318）提出。
- **librispeech_pc_testset（LibriSpeech_pc测试集）**：面向零样本文本转语音模型的英语测试集，由论文《F5-TTS》（https://arxiv.org/abs/2410.06885）提出。
- **seedtts_testset（Seed-TTS测试集）**：面向零样本文本转语音模型的中英双语测试集，由论文《Seed-TTS》（https://arxiv.org/abs/2406.02430）提出。