PHBJT/mls-annotated

该数据集是对非英语多语言LibriSpeech（MLS）数据集的注释版本。MLS数据集是一个适用于语音研究的大型多语言语料库，包含来自LibriVox的有声读物，涵盖8种语言（英语、德语、荷兰语、西班牙语、法语、意大利语、葡萄牙语、波兰语）。该数据集提供了对说话者和话语特征的自然语言注释，这些注释是使用Data-Speech存储库生成的。该数据集用于训练Parler-TTS多语言模型，并提供了训练脚本和工具。

This dataset consists of annotations of the Non-English subset of the Multilingual LibriSpeech (MLS) dataset. The MLS dataset is a large multilingual corpus suitable for speech research, derived from read audiobooks from LibriVox and includes 8 languages (English, German, Dutch, Spanish, French, Italian, Portuguese, Polish). The dataset provides natural language annotations on the characteristics of speakers and utterances, generated using the Data-Speech repository. It was used to train the Parler-TTS multilingual model and includes training scripts and utilities.