metadata
dataset_info:
features:
- name: audio_path
dtype:
audio:
sampling_rate: 16000
- name: transcription
dtype: string
- name: label
dtype: string
splits:
- name: train
num_bytes: 1588104875
num_examples: 11357
download_size: 1435954992
dataset_size: 1588104875
configs:
- config_name: default
data_files:
- split: train
path: data/train-*