metadata
dataset_info:
features:
- name: mfcc
sequence:
sequence: float32
- name: phoneme_align
sequence: float64
splits:
- name: train
num_bytes: 622446240
num_examples: 9999
- name: validation
num_bytes: 93360300
num_examples: 2703
- name: test
num_bytes: 93599280
num_examples: 2619
download_size: 713568979
dataset_size: 809405820
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: validation
path: data/validation-*
- split: test
path: data/test-*