metadata
dataset_info:
features:
- name: file
dtype: string
- name: is_valid
dtype: bool
- name: language
dtype:
class_label:
names:
'0': et
- name: speaker_id
dtype: string
- name: gender
dtype:
class_label:
names:
'0': MALE
'1': FEMALE
'2': OTHER
'3': NAN
- name: keyword
dtype: string
- name: audio
dtype:
audio:
sampling_rate: 16000
- name: label
dtype: int64
- name: pitch
dtype: float64
splits:
- name: train
num_bytes: 206363153.125
num_examples: 6367
- name: validation
num_bytes: 28651620
num_examples: 884
- name: test
num_bytes: 28813682
num_examples: 889
download_size: 75201381
dataset_size: 263828455.125
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: validation
path: data/validation-*
- split: test
path: data/test-*