metadata
dataset_info:
features:
- name: video_id
dtype: string
- name: dialect
dtype: string
- name: gender
dtype: string
- name: country
dtype: string
- name: category
dtype: string
- name: text
dtype: string
- name: audio
dtype:
audio:
sampling_rate: 16000
splits:
- name: train
num_bytes: 121908978086.559
num_examples: 942813
- name: test
num_bytes: 13533591273.409
num_examples: 104757
download_size: 134450266666
dataset_size: 135442569359.968
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: test
path: data/test-*