metadata
dataset_info:
features:
- name: utterance_id
dtype: string
- name: audio
dtype:
audio:
sampling_rate: 22050
- name: text
dtype: string
- name: speaker
dtype: string
- name: duration
dtype: float32
- name: accent
dtype: string
splits:
- name: train
num_bytes: 2056733391
num_examples: 27788
- name: valid
num_bytes: 106441476.4
num_examples: 1385
- name: test
num_bytes: 174349135.2
num_examples: 2222
download_size: 2337031847
dataset_size: 2337524002.6
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: valid
path: data/valid-*
- split: test
path: data/test-*