---
dataset_info:
  features:
  - name: speech
    sequence: float32
  - name: trascript
    dtype: string
  - name: answer
    dtype: string
  splits:
  - name: train
    num_bytes: 38175974440
    num_examples: 123433
  download_size: 37886689488
  dataset_size: 38175974440
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
---