metadata
dataset_info:
features:
- name: audio
dtype: audio
- name: audio_id
dtype: string
- name: transcription
dtype: string
- name: normalized_text
dtype: string
splits:
- name: train
num_bytes: 40041320
num_examples: 542
download_size: 38275595
dataset_size: 40041320
configs:
- config_name: default
data_files:
- split: train
path: data/train-*