metadata
dataset_info:
features:
- name: file_name
dtype: string
- name: transcription
dtype: string
- name: transcription_normalised
dtype: string
- name: audio
dtype: audio
splits:
- name: train
num_bytes: 3296777822
num_examples: 13496
download_size: 3154329880
dataset_size: 3296777822
configs:
- config_name: default
data_files:
- split: train
path: data/train-*