---
dataset_info:
  features:
  - name: audio
    dtype: audio
  - name: vocals
    dtype: audio
  - name: instrumentals
    dtype: audio
  splits:
  - name: train
    num_bytes: 532177050.0
    num_examples: 72
  download_size: 448632890
  dataset_size: 532177050.0
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
---