metadata
dataset_info:
features:
- name: audio
dtype: audio
- name: timestamps_start
sequence: float64
- name: timestamps_end
sequence: float64
- name: speakers
sequence: string
splits:
- name: train
num_bytes: 186825817
num_examples: 61
download_size: 172710498
dataset_size: 186825817
configs:
- config_name: default
data_files:
- split: train
path: data/train-*