metadata
dataset_info:
features:
- name: whisper_text
dtype: string
- name: correct_text
dtype: string
- name: language_probability
dtype: string
splits:
- name: train
num_bytes: 2646316
num_examples: 871
download_size: 1840526
dataset_size: 2646316
configs:
- config_name: default
data_files:
- split: train
path: data/train-*