metadata
dataset_info:
features:
- name: audio
dtype:
audio:
sampling_rate: 16000
- name: speaker_id
dtype: string
- name: text
dtype: string
- name: language
dtype: string
- name: gender
dtype: string
- name: age_range
dtype: string
- name: phase
dtype: string
- name: normalized_text
dtype: string
splits:
- name: train
num_bytes: 52133287467.784
num_examples: 195644
download_size: 50841725483
dataset_size: 52133287467.784
configs:
- config_name: default
data_files:
- split: train
path: data/train-*