metadata
dataset_info:
features:
- name: waveforms
sequence: float64
- name: labels
sequence:
sequence: uint8
- name: nb_speakers
sequence: int8
splits:
- name: train
num_bytes: 1314366132
num_examples: 1024
- name: validation
num_bytes: 137336686
num_examples: 107
- name: test
num_bytes: 1095345524
num_examples: 853
download_size: 468163122
dataset_size: 2547048342
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: validation
path: data/validation-*
- split: test
path: data/test-*