adaadig/candorconv001

这是一个包含音频数据和对应转录文本的数据集，同时提供了说话者信息、对话轮次、文本描述等额外信息。数据集通过wav2vec模型提供了转录文本，并且还提供了清洗后的转录文本。此外，数据集包含了一些音频质量相关的指标，如信噪比、c50值等。数据集被划分为训练集，可用于语音识别、语音质量评估等研究。

This dataset includes audio data and their corresponding transcripts, along with additional information such as speaker information, conversation turns, and text descriptions. The dataset provides transcripts generated by the wav2vec model and also offers cleaned versions of the transcripts. In addition, it contains audio quality-related metrics such as signal-to-noise ratio and c50 values. The dataset is split into a training set, which can be used for research in speech recognition, speech quality assessment, etc.