Indic languages represent a significant aspect of India’s cultural heritage, embodying collective knowledge, traditions, and customs. Preserving this heritage is crucial. Optical Character Recognition (OCR) technology aids in simplifying text recognition tasks by extracting text from images. This study uses an established OCR model to digitize document images of extremely low-resource Indian languages, which previous OCR efforts did not focus on. Preparing corpora for such languages is challenging due to the scarcity of expert linguists and the required time and resources. We introduce a synthetic dataset, Mozhi-LR(S), and a real dataset, Mozhi-LR(R), comprising word level images with textual transcriptions for these nine languages. We trained our model using synthetic datasets and used real ones to fine-tune them, achieving high accuracy on synthetic and real datasets. Additionally, we provide web-based apps that integrate our OCR models with APIs. This integration facilitates the digitization of Indic printed documents in extremely low-resource languages. The codebase, trained models, and datasets utilized in this work are publicly accessible at https://github.com/ALIKSARKAR/Printed-ELRIL-OCR.