---
dataset_info:
  features:
  - name: audio
    dtype:
      audio:
        sampling_rate: 16000
  - name: speakers
    list: string
  - name: timestamps_start
    list: float64
  - name: timestamps_end
    list: float64
  splits:
  - name: train
    num_bytes: 3325286223
    num_examples: 3469
  download_size: 3325376446
  dataset_size: 3325286223
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
---