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: 819656860.1535801
num_examples: 8585
- name: test
num_bytes: 91083592.84641996
num_examples: 954
download_size: 900099233
dataset_size: 910740453
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: test
path: data/test-*