metadata
dataset_info:
features:
- name: audio
dtype: audio
- name: intent_class
dtype: int64
- name: transcription
dtype: string
splits:
- name: train
num_bytes: 4815692868.684
num_examples: 20898
download_size: 3117441327
dataset_size: 4815692868.684
configs:
- config_name: default
data_files:
- split: train
path: data/train-*