Open, reproducible benchmark that compares OCR engines on Urdu text lines, with a weekly automated leaderboard, a synthetic data generator, a trainable model and a photo reader app.
-
Updated
Sep 21, 2026 - Python
Open, reproducible benchmark that compares OCR engines on Urdu text lines, with a weekly automated leaderboard, a synthetic data generator, a trainable model and a photo reader app.
Description: Production-grade RAG pipeline for IT helpdesk — hybrid search, confidence-gated responses, OCR support, and streaming APIs.
[🎯Challenge] DIO's Artificial Intelligence Challenge, with a presentation on how AI works to create images and recognize text from those images.
To associate your repository with the tessearct topic, visit your repo's landing page and select "manage topics."