Comparative evaluation and combination of automatic rhythm description system...

The dataset DatasetVocal contains 75 excerpts with highly predominant vocals in WAV format and has been used for beat tracking evaluation.数据集DatasetVocal包含75段人声占绝对主导的音频片段，所有片段均采用WAV格式，该数据集已被用于节拍跟踪（beat tracking）的评估工作。