metadata
dataset_info:
features:
- name: file_name
dtype: string
- name: transcript
dtype: string
- name: duration
dtype: float32
- name: audio
dtype:
audio:
sampling_rate: 16000
splits:
- name: train
num_bytes: 10373489972.532295
num_examples: 12177
- name: test
num_bytes: 1152609996.9480329
num_examples: 1353
download_size: 11464599564
dataset_size: 11526099969.480328
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: test
path: data/test-*