Adapting the Tesseract Open-Source OCR Engine for Tamil and Sinhala Legacy Fonts and Creating a Parallel Corpus for Tamil-Sinhala-English
This addresses the challenge of building monolingual and parallel corpora for low-resource languages in government documents, though it is incremental as it fine-tunes an existing OCR engine.
The paper tackled the problem of extracting text from PDFs with legacy fonts for low-resource languages like Tamil and Sinhala by adapting the Tesseract OCR engine with LSTM-based training on over 20 fonts, reducing character-level error rates from 6.03 to 2.61 for Tamil and 7.61 to 4.74 for Sinhala, and creating a parallel corpus with over 180k sentences per language.
Most low-resource languages do not have the necessary resources to create even a substantial monolingual corpus. These languages may often be found in government proceedings but mainly in Portable Document Format (PDF) that contains legacy fonts. Extracting text from these documents to create a monolingual corpus is challenging due to legacy font usage and printer-friendly encoding, which are not optimized for text extraction. Therefore, we propose a simple, automatic, and novel idea that can scale for Tamil, Sinhala, English languages, and many documents along with parallel corpora. Since Tamil and Sinhala are Low-Resource Languages, we improved the performance of Tesseract by employing LSTM-based training on more than 20 legacy fonts to recognize printed characters in these languages. Especially, our model detects code-mixed text, numbers, and special characters from the printed document. It is shown that this approach can reduce the character-level error rate of Tesseract from 6.03 to 2.61 for Tamil (-3.42% relative change) and 7.61 to 4.74 for Sinhala (-2.87% relative change), as well as the word-level error rate from 39.68 to 20.61 for Tamil (-19.07% relative change) and 35.04 to 26.58 for Sinhala (-8.46% relative change) on the test set. Also, our newly created parallel corpus consists of 185.4k, 168.9k, and 181.04k sentences and 2.11M, 2.22M, and 2.33M Words in Tamil, Sinhala, and English respectively. This study shows that fine-tuning Tesseract models on multiple new fonts help to understand the texts and enhances the performance of the OCR. We made newly trained models and the source code for fine-tuning Tesseract, freely available.