---
dataset_info:
  features:
  - name: image
    dtype: image
  - name: label
    dtype: int64
  - name: speaker
    dtype: int64
  - name: full_path
    dtype: string
  - name: arousal
    dtype: float64
  - name: valence
    dtype: float64
  splits:
  - name: train
    num_bytes: 3674300666.620908
    num_examples: 333683
  - name: validation
    num_bytes: 919030499.525
    num_examples: 83421
  - name: test
    num_bytes: 919030499.525
    num_examples: 83421
  download_size: 5484543444
  dataset_size: 5512361665.670907
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
  - split: validation
    path: data/validation-*
  - split: test
    path: data/test-*
---