---
license: mit
dataset_info:
  features:
  - name: speaker
    dtype: string
  - name: source
    dtype: string
  - name: start
    dtype: float64
  - name: end
    dtype: float64
  - name: style
    dtype: string
  - name: noise
    dtype: string
  - name: transcription
    dtype: string
  - name: audio
    dtype: audio
  - name: duration
    dtype: float64
  splits:
  - name: train
    num_bytes: 13550244048.138
    num_examples: 64659
  download_size: 14945729120
  dataset_size: 13550244048.138
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
---