metadata
dataset_info:
features:
- name: audio
dtype: audio
- name: transcription
dtype: string
- name: audio_id
dtype: string
- name: sub_dir
dtype: string
splits:
- name: train
num_bytes: 1488787368
num_examples: 10000
download_size: 921579917
dataset_size: 1488787368
configs:
- config_name: default
data_files:
- split: train
path: data/train-*