metadata
dataset_info:
features:
- name: logits
sequence: float32
- name: mfcc
sequence:
sequence: float64
splits:
- name: train
num_bytes: 741868756
num_examples: 145693
download_size: 756597580
dataset_size: 741868756
configs:
- config_name: default
data_files:
- split: train
path: data/train-*