metadata
dataset_info:
features:
- name: file_id
dtype: string
- name: speaker_id
dtype: string
- name: text
dtype: string
- name: audio
dtype:
audio:
sampling_rate: 16000
- name: duration
dtype: float64
splits:
- name: train
num_bytes: 11387400487.566462
num_examples: 148234
- name: validation
num_bytes: 1430228270.6948454
num_examples: 18529
- name: test
num_bytes: 1456593933.2726939
num_examples: 18530
download_size: 13175735009
dataset_size: 14274222691.534
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: validation
path: data/validation-*
- split: test
path: data/test-*