the numebr of genre label is different from the audio data

[ShungJhon]

I load the genre label.npy in the data folder, the shape is (6632,10), while the aist train dataset only has 2581 audios and motions. why?