metadata
dataset_info:
features:
- name: speaker_id
dtype: int64
- name: age
dtype: int64
- name: gender
dtype: string
- name: location
dtype: string
- name: audio
dtype:
audio:
sampling_rate: 16000
- name: md5sum
dtype: string
- name: duration
dtype: float64
- name: pdp_score
dtype: float64
- name: transcription
dtype: string
splits:
- name: train
num_bytes: 1153968331
num_examples: 11990
- name: valid
num_bytes: 230752160
num_examples: 2499
- name: test
num_bytes: 230685363
num_examples: 2160
download_size: 1590976543
dataset_size: 1615405854
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: valid
path: data/valid-*
- split: test
path: data/test-*