metadata
dataset_info:
features:
- name: audio
dtype:
audio:
sampling_rate: 16000
- name: speakers
list: string
- name: timestamps_start
list: float64
- name: timestamps_end
list: float64
splits:
- name: train
num_bytes: 566684780
num_examples: 592
download_size: 566701870
dataset_size: 566684780
configs:
- config_name: default
data_files:
- split: train
path: data/train-*