metadata
dataset_info:
features:
- name: audio
dtype: audio
- name: text
dtype: string
- name: speaker_id
dtype: int64
splits:
- name: train
num_bytes: 4613704709
num_examples: 22807
- name: validation
num_bytes: 576751387
num_examples: 3075
download_size: 4892060275
dataset_size: 5190456096
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: validation
path: data/validation-*