metadata
dataset_info:
features:
- name: audio
dtype:
audio:
sampling_rate: 16000
- name: transcripts
dtype: string
splits:
- name: train
num_bytes: 3459182456.8
num_examples: 8000
- name: test
num_bytes: 448898880.1
num_examples: 1000
- name: valid
num_bytes: 421215489.1
num_examples: 1000
download_size: 4294875689
dataset_size: 4329296826
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: test
path: data/test-*
- split: valid
path: data/valid-*