metadata
dataset_info:
features:
- name: file_name
dtype: string
- name: audio
dtype: audio
- name: EpId
dtype: string
- name: ClipId
dtype: string
- name: Output
dtype: int64
splits:
- name: train
num_bytes: 732087216.817
num_examples: 9539
download_size: 730438119
dataset_size: 732087216.817
configs:
- config_name: default
data_files:
- split: train
path: data/train-*