metadata
dataset_info:
features:
- name: audio
dtype:
audio:
sampling_rate: 16000
- name: text
dtype: string
- name: labels
dtype: string
- name: speaker
dtype: string
- name: filename
dtype: string
splits:
- name: train
num_bytes: 6365104105.125
num_examples: 50111
download_size: 6345373069
dataset_size: 6365104105.125
configs:
- config_name: default
data_files:
- split: train
path: data/train-*