metadata
dataset_info:
features:
- name: audio
dtype: audio
- name: label
dtype: int64
- name: valence
dtype: float64
- name: arousal
dtype: float64
- name: domination
dtype: float64
- name: arousal_norm
dtype: float64
- name: valence_norm
dtype: float64
- name: speakerID
dtype: int64
- name: utterance_id
dtype: string
- name: transcript
dtype: string
- name: speaker_id
dtype: int64
- name: emotion
dtype: string
- name: difficulty
dtype: float64
splits:
- name: train
num_bytes: 646837070.77
num_examples: 4490
download_size: 630481260
dataset_size: 646837070.77
configs:
- config_name: default
data_files:
- split: train
path: data/train-*