metadata
dataset_info:
features:
- name: file
dtype:
audio:
sampling_rate: 16000
- name: text
dtype: string
- name: phonemes
dtype: string
splits:
- name: train
num_bytes: 5532639166.179
num_examples: 28539
- name: validation
num_bytes: 295164149.124
num_examples: 2703
- name: test
num_bytes: 322058605.08
num_examples: 2620
download_size: 6355290744
dataset_size: 6149861920.382999
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: validation
path: data/validation-*
- split: test
path: data/test-*