XRXRX/X-Voice-Dataset-Train
收藏资源简介:
这是一个大规模多语言语音数据集,专为文本到语音和自动语音识别任务设计。数据集包含超过1TB的数据,覆盖多种语言,包括保加利亚语、捷克语、丹麦语、德语、希腊语、英语、西班牙语、爱沙尼亚语、法语、芬兰语、匈牙利语、克罗地亚语、印度尼西亚语、意大利语、日语、韩语、立陶宛语、拉脱维亚语、马耳他语、荷兰语、波兰语、葡萄牙语、罗马尼亚语、俄语、斯洛伐克语、斯洛文尼亚语、瑞典语、泰语、越南语和中文。数据集整合了多个来源的子数据集,如Multilingual LibriSpeech、Emilia、LEMAS、VoxPopuli、Granary(MOSEL部分)、GigaSpeech 2和Reazon Speech,每个子数据集都有其自己的许可证,用户在使用时必须遵守相应子数据集的许可证条款。
This is a large-scale multilingual speech dataset specifically designed for text-to-speech and automatic speech recognition (ASR) tasks. The dataset contains over 1TB of data and covers a wide range of languages, including Bulgarian, Czech, Danish, German, Greek, English, Spanish, Estonian, French, Finnish, Hungarian, Croatian, Indonesian, Italian, Japanese, Korean, Lithuanian, Latvian, Maltese, Dutch, Polish, Portuguese, Romanian, Russian, Slovak, Slovenian, Swedish, Thai, Vietnamese, and Chinese. The dataset integrates sub-datasets from multiple sources, such as Multilingual LibriSpeech, Emilia, LEMAS, VoxPopuli, Granary (MOSEL section), GigaSpeech 2, and Reazon Speech. Each sub-dataset has its own license, and users must comply with the license terms of the corresponding sub-dataset during usage.




