metadata
dataset_info:
features:
- name: speaker
dtype: int64
- name: hour
dtype: string
- name: file_path
dtype: string
- name: gender
dtype: string
- name: quality
dtype: string
- name: transcript
dtype: string
- name: label
dtype: float64
- name: audio
dtype: audio
splits:
- name: train
num_bytes: 575241298
num_examples: 735
- name: test
num_bytes: 42812541
num_examples: 53
download_size: 603067900
dataset_size: 618053839
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: test
path: data/test-*