metadata
dataset_info:
features:
- name: audio
dtype:
audio:
sampling_rate: 16000
- name: duration
dtype: float64
- name: language
dtype: string
- name: transcript_text
dtype: string
splits:
- name: train
num_bytes: 26482668
num_examples: 30
download_size: 24994884
dataset_size: 26482668
configs:
- config_name: default
data_files:
- split: train
path: data/train-*