metadata
dataset_info:
features:
- name: audio
dtype:
audio:
sampling_rate: 16000
- name: original_sentence
dtype: string
- name: preprocessed_sentence
dtype: string
splits:
- name: train
num_bytes: 93566852.04
num_examples: 2835
- name: test
num_bytes: 34744852.9
num_examples: 1290
download_size: 112056156
dataset_size: 128311704.94
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: test
path: data/test-*