metadata
dataset_info:
features:
- name: audio
dtype:
audio:
sampling_rate: 16000
- name: transcription
dtype: string
- name: speaker_id
dtype: int64
- name: id
dtype: int64
- name: gender
dtype: string
splits:
- name: train
num_bytes: 1981356971.6736403
num_examples: 1231
download_size: 1815407349
dataset_size: 1981356971.6736403
configs:
- config_name: default
data_files:
- split: train
path: data/train-*