mollwollfumble said:
dv said:
Jesus H … there have been free image-to-text converters online since the last millennium
Yes, but they’re all “artificial intelligence” style image to text conversion. Which means, no control of fonts, and hence no friggin way to improve the output.
eg. The first newspaper article from a random search of Trove contains a passage which converts to text as:
- Fresh berf and mutton, Is. Gd.; bacon 2s . ; Lult-r(#i!i) Is. 4 d .; fresh 2s. (id .; ian sron I s .; pn;k (fresh) Is. (Id.- salt ditto G d.; Vle-ese 2 s.; cufiec Is-Gd.; raisins Is.; dried fruits is, Gd.; rice fid .; potatoes 4 d ,; teu 7 s .; sugar is. 4 d .; flour 6 d ; soap Is. 6 d .; starch 2* ; gunpowder 4spvr
The original is:

And even Blind Freddy can read that as: Fresh beef and mutton, 1s. 6d,; bacon 2s.; butter (salt) 1s. 4d.; fresh 2s. 6d.; kangaroo 1s.; pork (fresh) 1s. 6d.; salt ditto 6d.; cheese 2s.; coffee 1s. 6d.; etc.,
If, if ever you find even one image-to-text converter program that allows me to input my own font(s) then let me know immediately.
Sounds like you could get involved with correcting stuff in Trove-
https://help.nla.gov.au/trove/using-trove/getting-to-know-us/trove-is
Image to text can do good things, but there’s the need for humans to get in too. I got involved with a project that had old ship logs with weather observations, they got volunteers to read the old entries which were in often beautiful writing, and to type out all the info so they could crunch the numbers. Beautifully preserved ships books are great, but you can’t feed them into the computer without a bit of help.
I was on Trove a few days ago, searching for new stuff on my grandparents. I’d found an article my grandfather wrote on his experiences at Poziers in WW1 a few months ago, to the delight of my family who had now known about it. I found another one from him written while on leave in Paris, which needs much more editing (I may go back and do it). Amazing to read him, over 100 years later. Found other stuff from both sides of the family, insights into times gone by, like long description of a wedding. There is fantastic stuff to be found in the old papers, worth just diving in for a lucky dip at times.