metadata
dataset_info:
features:
- name: path
dtype: string
- name: audio
dtype:
audio:
sampling_rate: 16000
- name: original_sentence
dtype: string
- name: preprocessed_sentence
dtype: string
splits:
- name: train
num_bytes: 1722176870.5
num_examples: 11660
- name: test
num_bytes: 86118591
num_examples: 760
download_size: 1772673347
dataset_size: 1808295461.5
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: test
path: data/test-*