metadata
dataset_info:
features:
- name: __key__
dtype: string
- name: transcription
dtype: string
splits:
- name: train
num_bytes: 256915526
num_examples: 2595649
- name: test
num_bytes: 1334400
num_examples: 14000
download_size: 140261128
dataset_size: 258249926
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: test
path: data/test-*