Summary of Unlocking the Archives: Using Large Language Models to Transcribe Handwritten Historical Documents, by Mark Humphries et al.
Unlocking the Archives: Using Large Language Models to Transcribe Handwritten Historical Documents
by Mark Humphries, Lianne C. Leddy, Quinn Downton, Meredith Legace, John McConnell, Isabella Murray, Elizabeth Spence
First submitted to arxiv on: 2 Nov 2024
Categories
- Main: Computer Vision and Pattern Recognition (cs.CV)
- Secondary: Computation and Language (cs.CL); Digital Libraries (cs.DL); Machine Learning (cs.LG)
GrooveSquid.com Paper Summaries
GrooveSquid.com’s goal is to make artificial intelligence research accessible by summarizing AI papers in simpler terms. Each summary below covers the same AI paper, written at different levels of difficulty. The medium difficulty and low difficulty versions are original summaries written by GrooveSquid.com, while the high difficulty version is the paper’s original abstract. Feel free to learn from the version that suits you best!
| Summary difficulty | Written by | Summary |
|---|---|---|
| High | Paper authors | High Difficulty Summary Read the original abstract here |
| Medium | GrooveSquid.com (original content) | Medium Difficulty Summary The study demonstrates the capabilities of Large Language Models (LLMs) in transcribing historical handwritten documents with higher accuracy than specialized Handwritten Text Recognition (HTR) software. The researchers introduce an open-source tool called Transcription Pearl that leverages commercially available LLMs from OpenAI, Anthropic, and Google to automatically transcribe and correct batches of handwritten documents. In tests on a diverse corpus of 18th/19th century English language handwritten documents, LLMs achieved Character Error Rates (CER) of 5.7 to 7% and Word Error Rates (WER) of 8.9 to 15.9%, improving upon state-of-the-art HTR software like Transkribus by 14% and 32% respectively. Furthermore, when LLMs were used to correct transcriptions as well as texts generated by conventional HTR software, they achieved near-human levels of accuracy, with CERs as low as 1.8% and WERs of 3.5%. The study highlights the potential of incorporating LLMs into software tools like Transcription Pearl for mass transcription of historical handwritten documents, significantly streamlining the digitization process. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary The paper shows that Large Language Models (LLMs) can read old handwriting better than special computers designed just for that task. They created a tool called Transcription Pearl that uses these LLMs to quickly and accurately write down lots of old letters and documents. The LLMs got the words mostly right, with errors happening only rarely. When they corrected mistakes made by other computers, they did an amazing job, getting almost all the words correct! This is a big deal because it means we can make copies of these old documents much faster and cheaper than before. |
Keywords
* Artificial intelligence * Cer
