metadata
dataset_info:
features:
- name: file_name
dtype: string
- name: transcript
dtype: string
- name: transcript_normalized
dtype: string
- name: gender
dtype: string
- name: audio
dtype:
audio:
sampling_rate: 16000
splits:
- name: train
num_bytes: 6273825386.76
num_examples: 20440
- name: test
num_bytes: 1560223303.76
num_examples: 5110
download_size: 7804588208
dataset_size: 7834048690.52
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: test
path: data/test-*