Vikhrmodels/ToneSpeak_quantized-bigcodec

这是一个包含音频、文本和相关信息的语音数据集。数据集由训练集和验证集组成，其中每个部分包含了音频文件和对应的文本信息。音频采样率为24000Hz，并且数据集中还包含了提示信息、声音名称和音频令牌序列。训练集包含6298个示例，验证集包含700个示例。

This is a speech dataset containing audio, text, and related information. The dataset is divided into training and validation sets, each of which includes audio files and corresponding text information. The audio sampling rate is 24000Hz, and the dataset also includes prompt information, voice names, and audio token sequences. The training set contains 6298 examples, and the validation set contains 700 examples.