---
dataset_info:
  features:
  - name: audio
    dtype:
      audio:
        sampling_rate: 16000
  - name: sentence
    dtype: string
  splits:
  - name: train
    num_bytes: 38525556030.958
    num_examples: 780593
  - name: test
    num_bytes: 727918231.539
    num_examples: 17509
  download_size: 26620714783
  dataset_size: 39253474262.497
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
  - split: test
    path: data/test-*
language:
- th
task_categories:
- automatic-speech-recognition
---