---
dataset_info:
  config_name: small
  features:
  - name: audio
    dtype:
      audio:
        sampling_rate: 16000
  - name: text
    dtype: string
  - name: whisper_transcript
    sequence: int64
  - name: input_length
    dtype: int64
  - name: labels
    sequence: int64
  splits:
  - name: train
    num_bytes: 2906406187.0
    num_examples: 20900
  download_size: 2866131051
  dataset_size: 2906406187.0
configs:
- config_name: small
  data_files:
  - split: train
    path: small/train-*
---