metadata
dataset_info:
features:
- name: audio_files
dtype: string
- name: emotion
dtype: string
- name: audio_data
dtype: binary
splits:
- name: train
num_bytes: 606526150
num_examples: 7442
download_size: 605703113
dataset_size: 606526150
configs:
- config_name: default
data_files:
- split: train
path: data/train-*