---
dataset_info:
  features:
  - name: transcription
    dtype: string
  - name: mel_spectrogram
    dtype: string
  splits:
  - name: train
    num_bytes: 2490065755
    num_examples: 2602
  download_size: 1959924391
  dataset_size: 2490065755
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
---