---
dataset_info:
  features:
  - name: audio
    dtype: audio
  - name: caption
    dtype: string
  - name: language
    dtype: string
  - name: speaker
    dtype: string
  - name: sample_id
    dtype: int64
  splits:
  - name: train
    num_bytes: 233339927.2
    num_examples: 400
  - name: validation
    num_bytes: 29167490.9
    num_examples: 50
  - name: test
    num_bytes: 29167490.9
    num_examples: 50
  download_size: 255972099
  dataset_size: 291674909.0
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
  - split: validation
    path: data/validation-*
  - split: test
    path: data/test-*
---