metadata
dataset_info:
features:
- name: speaker_id
dtype: string
- name: audio
dtype:
audio:
sampling_rate: 48000
- name: file
dtype: string
- name: text
dtype: string
- name: text_id
dtype: string
- name: age
dtype: string
- name: gender
dtype: string
- name: accent
dtype: string
- name: region
dtype: string
- name: comment
dtype: string
- name: country
dtype: string
- name: resampled_audio
sequence:
sequence: float64
splits:
- name: train
num_bytes: 49246544243.724
num_examples: 66026
- name: test
num_bytes: 13225829772.684
num_examples: 17066
download_size: 61783691699
dataset_size: 62472374016.408
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: test
path: data/test-*