metadata
dataset_info:
features:
- 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
- name: audio_token
sequence:
sequence:
sequence: int64
- name: audio_duration
dtype: float64
splits:
- name: train
num_bytes: 1078679482
num_examples: 268263
- name: validation
num_bytes: 3954505
num_examples: 591
- name: test
num_bytes: 7151488
num_examples: 1469
download_size: 213280748
dataset_size: 1089785475
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: validation
path: data/validation-*
- split: test
path: data/test-*