xtz999/jordanian_arabic_speech_amir

该数据集是一个音频-文本配对数据集，包含119,918个训练示例。每个示例由音频文件和对应的文本组成，音频采样率为22050 Hz，文本为字符串格式。数据集总大小为24,233,260,419字节，下载大小为24,052,712,934字节。数据文件位于train-*路径下，适用于语音识别、音频生成或文本到语音等任务。This dataset is an audio-text paired dataset containing 119,918 training examples. Each example consists of an audio file and its corresponding text, with audio sampled at 22050 Hz and text in string format. The total dataset size is 24,233,260,419 bytes, with a download size of 24,052,712,934 bytes. Data files are located under the train-* path, suitable for tasks such as speech recognition, audio generation, or text-to-speech.