metadata
dataset_info:
config_name: split_0
features:
- name: name
dtype: string
- name: audio
dtype:
audio:
sampling_rate: 16000
- name: transcription
dtype: string
- name: whisper_transcript
sequence: int64
splits:
- name: train
num_bytes: 304902975318
num_examples: 2678733
download_size: 302498718984
dataset_size: 304902975318
configs:
- config_name: split_0
data_files:
- split: train
path: split_0/train-*