confit/librispeech-wds

LibriSpeech ASR是一个面向自动语音识别任务的公共数据集，包含了经过预处理的纯净语音和其它类型的语音数据。数据集分为clean和other两种配置，每种配置都包括训练集、验证集和测试集。数据集中的语音为英语，适用于自动语音识别模型的训练和评估。

LibriSpeech ASR is a public dataset for automatic speech recognition tasks, including pre-processed clean speech and other types of speech data. The dataset is divided into two configurations: clean and other, each with training, validation, and test sets. The speech in the dataset is in English, suitable for training and evaluating automatic speech recognition models.