atlasia/MoulSot-Full-80

该数据集是一个多模态音频数据集，包含音频样本及其相关的质量评估指标和文本信息。每个样本具有唯一的ID、音频数据（以音频格式存储）、采样率、通道数，以及PESQ（语音质量感知评估）、STOI（短时客观可懂度）和SI-SDR（尺度不变信噪比）等音频质量评估指标。此外，数据集还提供与音频对应的文本内容和通道标识，以及音频的持续时间（以秒为单位）。数据集分为训练集和测试集，训练集包含1000个样本，测试集包含200个样本，总大小约为164MB，适用于音频处理、语音质量评估和语音识别等任务。

This dataset is a multimodal audio dataset containing audio samples along with associated quality assessment metrics and text information. Each sample includes a unique ID, audio data (stored in audio format), sample rate, number of channels, and audio quality assessment metrics such as PESQ (Perceptual Evaluation of Speech Quality), STOI (Short-Time Objective Intelligibility), and SI-SDR (Scale-Invariant Signal-to-Distortion Ratio). Additionally, the dataset provides corresponding text content and channel identifiers for the audio, as well as the duration of the audio in seconds. The dataset is divided into training and test sets, with the training set containing 1000 samples and the test set containing 200 samples, totaling approximately 164MB in size. It is suitable for tasks such as audio processing, speech quality assessment, and speech recognition.