metadata
dataset_info:
features:
- name: audio
dtype:
audio:
sampling_rate: 16000
- name: text
dtype: string
- name: speaker_id
dtype: string
- name: gender
dtype:
class_label:
names:
'0': unknown
'1': female
'2': male
- name: file
dtype: string
- name: id
dtype: string
splits:
- name: train
num_bytes: 13628871343.61
num_examples: 56803
- name: validation
num_bytes: 197784478
num_examples: 591
- name: test
num_bytes: 352769289.03
num_examples: 1469
download_size: 14152473470
dataset_size: 14179425110.640001
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: validation
path: data/validation-*
- split: test
path: data/test-*