GSoC 2026
RenAIssance: Automating transcription of historical documents with weighted CRNN and LLM integration
This project will focus on applying hybrid end-to-end models that combine weighted convolutional-recurrent architectures (CNN-RNN) with large language models (LLMs) to recognize text from Spanish printed sources dating back to the seventeenth century. The ultimate goal is to create a user-friendly application that allows researchers to generate transcripts simply by uploading images in PDF or PNG format.
Project details
Technologies
Not listed in the archive