metadata
dataset_info:
features:
- name: audio
dtype:
audio:
sampling_rate: 22050
- name: filename
dtype: string
- name: transcription
dtype: string
splits:
- name: train
num_bytes: 11526915288.63
num_examples: 15270
download_size: 11879454648
dataset_size: 11526915288.63
configs:
- config_name: default
data_files:
- split: train
path: data/train-*