metadata
dataset_info:
features:
- name: audio_filepath
dtype:
audio:
sampling_rate: 16000
- name: duration
dtype: float64
- name: transcription
dtype: string
splits:
- name: train
num_bytes: 352705145.69
num_examples: 1929
download_size: 344783867
dataset_size: 352705145.69
configs:
- config_name: default
data_files:
- split: train
path: data/train-*