Aratako/LiquidAI-Hackathon-Tokyo-CPT-Data

该数据集包含四个配置：emilia-en、emilia-ja、wikipedia-en和wikipedia-ja。每个配置都详细说明了特征（input_ids、labels、attention_mask），训练部分中的示例数量和字节数、下载大小和数据集大小。该数据集用于自动语音识别、文本转语音和音频转音频等任务。包含的语言有日语（ja）和英语（en）。数据集遵循cc-by-nc-4.0许可。根据提供的信息，没有具体描述数据集的内容或用途，只提供了结构和技术细节。

The dataset consists of four configurations: emilia-en, emilia-ja, wikipedia-en, and wikipedia-ja. Each configuration details the features (input_ids, labels, attention_mask), the number of examples and bytes in the train split, the download size, and the dataset size. The dataset is used for tasks such as automatic-speech-recognition, text-to-speech, and audio-to-audio. The languages included are Japanese (ja) and English (en). The dataset is licensed under cc-by-nc-4.0. Based on the information provided, there is no specific description of the datasets content or purpose, only structural and technical details are given.