metadata
dataset_info:
features:
- name: audio
dtype:
audio:
sampling_rate: 16000
- name: text
dtype: string
- name: predictions
dtype: int64
- name: gender
dtype: string
- name: speaker_id
dtype: string
- name: index
dtype: int64
splits:
- name: test
num_bytes: 47519269
num_examples: 300
- name: train
num_bytes: 15241326764.5
num_examples: 99700
download_size: 15269004822
dataset_size: 15288846033.5
configs:
- config_name: default
data_files:
- split: test
path: data/test-*
- split: train
path: data/train-*