metadata
dataset_info:
config_name: en
features:
- name: audio
dtype:
audio:
sampling_rate: 16000
- name: path
dtype: string
- name: sentence
dtype: string
- name: length
dtype: float64
- name: whisper_transcript
sequence: int64
splits:
- name: train
num_bytes: 618348530.525
num_examples: 3009
download_size: 606619139
dataset_size: 618348530.525
configs:
- config_name: en
data_files:
- split: train
path: en/train-*