metadata
dataset_info:
features:
- name: transcription
dtype: string
- name: audio
dtype:
audio:
sampling_rate: 16000
splits:
- name: train
num_bytes: 2638324759.1
num_examples: 3050
- name: test
num_bytes: 705575156
num_examples: 729
download_size: 2422495881
dataset_size: 3343899915.1
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: test
path: data/test-*