Datasets:
metadata
language:
- vi
size_categories:
- 1K<n<10K
task_categories:
- automatic-speech-recognition
dataset_info:
features:
- name: path
dtype: string
- name: audio
dtype:
audio:
sampling_rate: 48000
- name: text
dtype: string
- name: duration
dtype: float64
splits:
- name: train
num_bytes: 379500674.644
num_examples: 4198
- name: validation
num_bytes: 79133689
num_examples: 900
- name: test
num_bytes: 80838113
num_examples: 900
download_size: 534859580
dataset_size: 539472476.644
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: validation
path: data/validation-*
- split: test
path: data/test-*