metadata
dataset_info:
features:
- name: audio
dtype:
audio:
sampling_rate: 16000
- name: id
dtype: int64
- name: transcription
dtype: string
- name: gender
dtype: string
- name: speaker_id
dtype: int64
splits:
- name: train
num_bytes: 11447403584.423744
num_examples: 41694
download_size: 10902441184
dataset_size: 11447403584.423744
configs:
- config_name: default
data_files:
- split: train
path: data/train-*