metadata
dataset_info:
features:
- name: audio
dtype: audio
- name: sentence
dtype: string
- name: start_time
dtype: string
- name: end_time
dtype: string
- name: speaker
dtype: int64
splits:
- name: train
num_bytes: 124429787.65911151
num_examples: 882
- name: validation
num_bytes: 31177985.340888485
num_examples: 221
download_size: 155005627
dataset_size: 155607773
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: validation
path: data/validation-*