Second dip into AI for transcribing documents

 Again using the new ScribeAI from My Heritage (MH) I tried another difficult document with it. This was from the court of common pleas from 1523. It was a short document which had previously been transcribed and translated professionally by someone well versed in them. The document is in highly abbreviated Latin in a neat secretary hand, and the image was clear.

The description of the document from MH was "This document is an entry from the Court of Common Pleas (identified by the 'CP 40' reference in the filename)", and gave more information on this.

The transcription and translation was much better than expected although some text was added that did not come from the document. The transcription did show missing characters from the abbreviations in square brackets which was good. The transcription unfortunately also showed a couple of wrong names.

The results were good (impressive even) for a quick look. Care needs to be taken if a document looks to be of interest to give it more evaluation as the AI output on its own can mislead.

It is interesting that AI transcriptions (the ones that I have played with at least) do not show missing characters where there is a mark on the parchment or paper, or a crease, or a tear at the edge of a page, but the missing characters seem to be guessed. I assume it needs human intelligence to identify such things.

First step in AI for transcribing old documents

 

The new Scribe AI from My Heritage - first look

I have been resisting  using AI tools for genealogy (and everything else) but talk about them is everywhere. I have been warned about their "hallucinations" and so on.

My palaeographic skills are fair but not expert back to Tudor times. 

I have tried Transkribus a couple of times without much success, getting quite mangled results on documents which were in English although not with a particularly good hand.

I received an email about the new feature at My Heritage a couple of days ago which describes itself as "a powerful new feature that transcribes, translates, and interprets historical documents and photos" , so it seemed worth giving it a go.  It has a drag and drop interface to upload a file which is easy.

I uploaded the will of Roger Allis of Glentworth, Lincolnshire (Lincolnshire archives ref. Stow wills 1563/47). It is a single page in an untidy secretary hand which I had previously transcribed with a handful of words missing that I couldn't interpret. Maybe not a fair test but I was interested to see what would happen, and I was pleasantly surprised.

 ScribeAI gave a summary of the document telling me it was in secretary hand and what a husbandman was, and discussed some of the items bequeathed. It also noted that the date (which was in January 1562) was in old style dating and corrected it to give the modern year.

It also gave a summary of the people mentioned in the will and their relationship to Roger and the bequests made.

Then the actual transcription. The first part of Glentworth was missing due to the edge of the page being torn, so it transcribed this as Panton (another parish in Lincolnshire) which was probaly fair given the writing. The problem was that later when the word Glentworth was complete it still carried on with Panton and did not correct itself. It did insert some words that weren't there and had some trouble with money, e.g ij s (2 shillings) was given as 2 sheep.

It did fill in some of my gaps which was excellent, but got one name wrong, and corrected one of my transcriptions.

Overall it was impressive. I would say that it could be of great use but used as a guide and not to take the given transcription as gospel. I gave it a pretty difficult document and it gave back a pretty good result.

Original documents

 It really pays to view the original documents if you can.

At Lincoln archives in the Nettleham bishops transcripts a marriage was found in 1599 for Edward Allys that hasn't been seen anywhere else or in the transcripts of the BTs (I hope I'm not wrong there!). 

That helped with a mystery but as usual prompts some more questions.

Audrey Allis and cyclist magazine

 Back in 2022 I wondered about an Audrey Allis who had been a contact in a cyclists magazine around the WW2 years.

After doing a little research it seems the mystery is solved and this was found to be a pen name for the person concerned. She was indeed a niece of "ragged staff" through his marriage to Ann Palmer, and used her first two Christian names of Audrey Alice to become Audrey Allis.

It really seems to be that simple.