metadata
dataset_info:
features:
- name: audio_id
dtype: string
- name: speaker_id
dtype: string
- name: audio
dtype:
audio:
sampling_rate: 16000
- name: audio_duration
dtype: float64
- name: transcription
dtype: string
- name: gender
dtype: string
- name: age_group
dtype: string
- name: locale
dtype: string
splits:
- name: validation
num_bytes: 890825380.2
num_examples: 1630
- name: test
num_bytes: 1111202096.16
num_examples: 2006
- name: train
num_bytes: 6024513454.8
num_examples: 10839
download_size: 6989609457
dataset_size: 8026540931.16
configs:
- config_name: default
data_files:
- split: validation
path: data/validation-*
- split: test
path: data/test-*
- split: train
path: data/train-*