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: 11492951424.472
num_examples: 3099
- name: test
num_bytes: 11394089657.564
num_examples: 3101
- name: train
num_bytes: 103283207115.864
num_examples: 28371
download_size: 123964397980
dataset_size: 126170248197.9
configs:
- config_name: default
data_files:
- split: validation
path: data/validation-*
- split: test
path: data/test-*
- split: train
path: data/train-*