metadata
dataset_info:
features:
- name: logits
sequence: float32
- name: mfcc
sequence:
sequence: float64
splits:
- name: train
num_bytes: 1103191984
num_examples: 216652
download_size: 1123489065
dataset_size: 1103191984
configs:
- config_name: default
data_files:
- split: train
path: data/train-*