metadata
dataset_info:
features:
- name: audio
dtype:
audio:
sampling_rate: 16000
- name: sentence
dtype: string
- name: phonemes
dtype: string
splits:
- name: train
num_bytes: 701808401
num_examples: 3136
- name: eval
num_bytes: 44758374
num_examples: 200
download_size: 708302833
dataset_size: 746566775
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: eval
path: data/eval-*