---
dataset_info:
  features:
  - name: audio
    dtype:
      audio:
        sampling_rate: 16000
  - name: text
    dtype: string
  - name: path
    dtype: string
  - name: condition_on_prev
    dtype: int64
  - name: whisper_transcript
    dtype: string
  splits:
  - name: train
    num_bytes: 1064819610.125
    num_examples: 1215
  - name: validation
    num_bytes: 267679544.0
    num_examples: 306
  download_size: 1329977753
  dataset_size: 1332499154.125
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
  - split: validation
    path: data/validation-*
---