shantipriya/

odia-ocr-merged

	Odia OCR - Merged Multi-Source Dataset




	
	Overview

This is a comprehensive merged Odia Optical Character Recognition (OCR) dataset combining three major public datasets:

OdiaGenAIOCR/Odia-lipi-ocr-data (64 samples) tell2jyoti/odia-handwritten-ocr (182,152 samples) darknight054/indic-mozhi-ocr - Odia subset (10,000+ samples)

Total: 192,000+ Odia OCR samples ready for training!

	Dataset Contents




	


	Source Breakdown




	

Dataset Samples Type License… See the full description on the dataset page: https://huggingface.co/datasets/shantipriya/odia-ocr-merged.