metadata
license: mit
dataset_info:
features:
- name: audio
dtype: audio
- name: transcript
dtype: string
- name: normalized_transcript
dtype: string
splits:
- name: train
num_bytes: 186166402.01533744
num_examples: 293
- name: val
num_bytes: 20967546.984662578
num_examples: 33
download_size: 204584198
dataset_size: 207133949
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: val
path: data/val-*