metadata
dataset_info:
features:
- name: audio
dtype:
audio:
sampling_rate: 16000
- name: text
dtype: string
- name: speaker
dtype: string
- name: id
dtype: string
- name: locale
dtype: string
- name: gender
dtype: string
- name: split
dtype: string
- name: duration_sec
dtype: float64
- name: sample_rate
dtype: int64
- name: duration_bucket
dtype: string
splits:
- name: train
num_bytes: 120556257
num_examples: 536
- name: validation
num_bytes: 15937646
num_examples: 73
- name: test
num_bytes: 15783815
num_examples: 71
download_size: 152265158
dataset_size: 152277718
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: validation
path: data/validation-*
- split: test
path: data/test-*