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: 2092459546.394
num_examples: 2994
- name: validation
num_bytes: 275319524
num_examples: 361
- name: test
num_bytes: 692444021
num_examples: 857
download_size: 3040363730
dataset_size: 3060223091.394
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: validation
path: data/validation-*
- split: test
path: data/test-*