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: 965600351.85
num_examples: 3800
- name: test
num_bytes: 50821072.55
num_examples: 200
download_size: 1016347834
dataset_size: 1016421424.4
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: test
path: data/test-*