---
dataset_info:
  features:
  - name: audio
    dtype:
      audio:
        sampling_rate: 24000
  - name: sentence
    dtype: string
  splits:
  - name: train
    num_bytes: 23554422884.469326
    num_examples: 637736
  - name: test
    num_bytes: 5844567487.548676
    num_examples: 159435
  download_size: 26651689028
  dataset_size: 29398990372.018
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
  - split: test
    path: data/test-*
language:
- th
task_categories:
- automatic-speech-recognition
license: cc-by-4.0
---