metadata
dataset_info:
features:
- name: image
dtype: image
- name: emotion
dtype: string
- name: emotion_id
dtype: int64
- name: EMOV_Labels
dtype: int64
- name: transcript
dtype: string
- name: speaker_name
dtype: string
- name: speaker_ID
dtype: int64
- name: file
dtype: string
splits:
- name: train
num_bytes: 283581963.755
num_examples: 6599
download_size: 267699667
dataset_size: 283581963.755
configs:
- config_name: default
data_files:
- split: train
path: data/train-*