metadata
dataset_info:
features:
- name: client_id
dtype: string
- name: path
dtype: string
- name: audio
dtype:
audio:
sampling_rate: 16000
- name: sentence
dtype: string
- name: up_votes
dtype: int64
- name: down_votes
dtype: int64
- name: age
dtype: string
- name: gender
dtype: string
- name: accent
dtype: string
- name: locale
dtype: string
- name: segment
dtype: string
- name: variant
dtype: string
splits:
- name: val
num_bytes: 711516922.695
num_examples: 16393
- name: train
num_bytes: 31746094333
num_examples: 780000
download_size: 31885617564
dataset_size: 32457611255.695
configs:
- config_name: default
data_files:
- split: val
path: data/val-*
- split: train
path: data/train-*