Sinhala Synthetic OCR-Large: Synthetic Images for Optical Character Recognition

Sinhala_synthetic_ocr-large is a dataset for optical character recognition tasks in the Sinhala language, created by Ransaka Ravihara and published on Hugging Face in 2024. The dataset's description suggests it contains synthetic images paired with text, likely generated for training machine learning models. Its specific size, format, and license details are not provided in the available metadata.

Use Cases

• Train optical character recognition models based on synthetic Sinhala text images.
• Benchmark OCR model performance on the Sinhala language.
• Augment real-world Sinhala text datasets with synthetic data to improve model robustness.

Strengths

• Dataset is specifically designed for the Sinhala language, addressing a less common script.
• Published on a major platform (Hugging Face) with a DOI (10.57967/hf/1809) for citation.
• Last updated on 2024-02-24, indicating recent maintenance.

Limitations

• Row count, file size, and specific data formats are unknown, limiting suitability assessment.
• Column-level documentation is absent; field semantics must be inferred after download.
• Data may reflect bias inherent to the synthetic generation method used.

Provenance

Source

Hugging Face dataset repository by author Ransaka Ravihara.

Collection Method

Likely contains synthetically generated images and text for OCR.

Freshness

Last updated 2024-02-24 13:51:00