metadata
dataset_info:
features:
- name: text
dtype: string
- name: topic
dtype: string
- name: emotion
dtype: string
- name: audio
dtype:
audio:
sampling_rate: 24000
- name: voice_description
dtype: string
splits:
- name: train
num_bytes: 306878327
num_examples: 800
download_size: 284832260
dataset_size: 306878327
configs:
- config_name: default
data_files:
- split: train
path: data/train-*