An HTR-LLM Workflow for High-Accuracy Transcription and Analysis of Abbreviated Latin Court Hand

TL;DR

This paper introduces a four-stage hybrid workflow combining specialized HTR models and large language models to transcribe and analyze challenging medieval Latin legal documents with high accuracy, automating complex tasks.

Contribution

It presents a novel multi-stage process integrating LLMs and HTR for high-accuracy transcription and analysis of medieval Latin documents, validated through case studies.

Findings

Achieved Word Error Rates of 2-7% against scholarly ground truths.

Demonstrated effective automation of complex transcription tasks.

Validated workflow through detailed case studies.

Abstract

This article presents and validates an ideal, four-stage workflow for the high-accuracy transcription and analysis of challenging medieval legal documents. The process begins with a specialized Handwritten Text Recognition (HTR) model, itself created using a novel "Clean Ground Truth" curation method where a Large Language Model (LLM) refines the training data. This HTR model provides a robust baseline transcription (Stage 1). In Stage 2, this baseline is fed, along with the original document image, to an LLM for multimodal post-correction, grounding the LLM's analysis and improving accuracy. The corrected, abbreviated text is then expanded into full, scholarly Latin using a prompt-guided LLM (Stage 3). A final LLM pass performs Named-Entity Correction (NEC), regularizing proper nouns and generating plausible alternatives for ambiguous readings (Stage 4). We validate this workflow through detailed case studies, achieving Word Error Rates (WER) in the range of 2-7% against scholarly ground truths. The results demonstrate that this hybrid, multi-stage approach effectively automates the most laborious aspects of transcription while producing a high-quality, analyzable output, representing a powerful and practical solution for the current technological landscape.