metadata
dataset_info:
features:
- name: file_name
dtype: string
- name: transcription
dtype: string
- name: transcription_normalized
dtype: string
- name: audio
dtype:
audio:
sampling_rate: 44100
splits:
- name: train
num_bytes: 9775310185.688154
num_examples: 7921
- name: test
num_bytes: 1059142455.4878436
num_examples: 881
download_size: 7496246656
dataset_size: 10834452641.175999
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: test
path: data/test-*