metadata
dataset_info:
features:
- name: emotion
dtype: string
- name: speaker_id
dtype: string
- name: duration
dtype: string
- name: code
dtype: string
- name: sentence
dtype: string
- name: translation
dtype: string
- name: audio
dtype: audio
splits:
- name: train
num_bytes: 4846166384.28
num_examples: 12420
download_size: 4743416379
dataset_size: 4846166384.28
configs:
- config_name: default
data_files:
- split: train
path: data/train-*