metadata
dataset_info:
features:
- name: audio
dtype:
audio:
sampling_rate: 16000
- name: sentence
dtype: string
splits:
- name: train
num_bytes: 1046279526.75
num_examples: 2994
download_size: 928678233
dataset_size: 1046279526.75
configs:
- config_name: default
data_files:
- split: train
path: data/train-*