metadata

dataset_info:
  features:
    - name: image
      dtype: image
    - name: text
      dtype: string
  splits:
    - name: train
      num_bytes: 27613382844.995
      num_examples: 7576785
    - name: test
      num_bytes: 74400308.776
      num_examples: 22964
  download_size: 24805635073
  dataset_size: 27687783153.771
configs:
  - config_name: default
    data_files:
      - split: train
        path: data/train-*
      - split: test
        path: data/test-*
language:
  - my
pretty_name: Myanmar OCR Dataset (Synthetic)
license: mit
task_categories:
  - image-to-text

Myanmar OCR Dataset

A synthetic dataset for training and fine-tuning Optical Character Recognition (OCR) models specifically for the Myanmar language.

Dataset Description

This dataset contains synthetically generated OCR images created specifically for Myanmar text recognition tasks. The images were generated using myanmar-ocr-data-generator, a fork of TextRecognitionDataGenerator with fixes for proper Myanmar character splitting.

Direct Download

Available from UT Austin SharePoint

The downloadable data is specifically formatted for PaddleOCR
Includes organized folder structure optimized for PaddleOCR training
Labels are provided in tab-separated format
Ready to fine-tuned for PaddleOCR

Key Features

Synthetic Myanmar text images with corresponding text labels
Maximum text length of 100 characters per image
Various fonts, styles, skews, various distortion types, blur levels, and backgrounds to improve model robustness

Generation Method

The dataset was created using a forked and customized version of the TextRecognitionDataGenerator:

Original repository: Belval/TextRecognitionDataGenerator
Customized fork: chuuhtetnaing/myanmar-ocr-data-generator

Text Sources

The text content used to generate the OCR images was extracted from:

chuuhtetnaing/myanmar-wikipedia-dataset: A collection of Myanmar language text from Wikipedia
chuuhtetnaing/myanmar-facebook-flores-dataset: A subset of Meta's FLORES dataset containing only Myanmar language text
chuuhtetnaing/myanmar-instruction-tuning-dataset: Compilation of question-answer pairs in the Myanmar language
chuuhtetnaing/mm-lib-book-dataset: A dataset of 437 books extracted from MM-Lib (Only 421 unicode font books are used for OCR dataset)

Usage

This dataset is particularly useful for:

Training OCR models for Myanmar language text recognition
Fine-tuning existing OCR models to improve performance on Myanmar script
Developing applications for Myanmar text digitization

License

This dataset is free to use for any purpose - personal, commercial, or educational.