talha-waqar17/spc_r_with_id

该数据集是一个包含音频和文本的多语言数据集，主要特征包括音频（采样率16000）、文本内容、语言标识、字幕（srt）以及多个音频质量评估指标（如平均对数概率、BLEU分数等）。数据集分为训练集（12900个样本）和测试集（679个样本），适用于语音识别、自然语言处理等任务。

This dataset is a multilingual collection containing audio and text data, with key features including audio (sampling rate 16000), text content, language identifiers, subtitles (srt), and various audio quality metrics (e.g., average log probability, BLEU scores). It is divided into a training set (12,900 samples) and a test set (679 samples), suitable for tasks such as speech recognition and natural language processing.