metadata
dataset_info:
config_name: large
features:
- name: audio
dtype:
audio:
sampling_rate: 16000
- name: text
dtype: string
- name: whisper_transcript
sequence: int64
- name: input_length
dtype: int64
- name: labels
sequence: int64
splits:
- name: train
num_bytes: 145659866932.2361
num_examples: 1042883
download_size: 143373138290
dataset_size: 145659866932.2361
configs:
- config_name: large
data_files:
- split: train
path: large/train-*