metadata
dataset_info:
features:
- name: audio
dtype:
audio:
sampling_rate: 24000
- name: instruct
dtype: string
splits:
- name: train
num_bytes: 2556630
num_examples: 10
download_size: 2318769
dataset_size: 2556630
configs:
- config_name: default
data_files:
- split: train
path: data/train-*