metadata
dataset_info:
features:
- name: file_name
dtype: string
- name: uni
dtype: string
- name: url
dtype: string
- name: sentence
dtype: string
- name: label
dtype: string
- name: speaker_id
dtype: string
- name: duration
dtype: float64
splits:
- name: train
num_bytes: 289109221.21559685
num_examples: 534753
- name: test
num_bytes: 72277710.78440313
num_examples: 133689
download_size: 165848047
dataset_size: 361386932
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: test
path: data/test-*