Transcrbing Print

[Basil_Bowdler]

Hi there, as a little experiment I just tried putting some printed text from a late C17 Dutch newspaper through Leo. While the site is saying that a transcription has been carried out, the transcription box is blank. This may be unrelated to the fact that it was attempting to transcribe printed text, but I haven’t noticed this problem with transcibing manuscript material.

[Jon]

Hi Basil. Yes, there is a known issue with print where the model ignores text. See here:

Transcription failures

Thanks both for sharing, this is really useful feedback! I’ll add a little to what Jon says. For your images Amelia: I’ve had a look through the logs and the transcription jobs were technically successful (i.e. no software bugs), but the AI failed to return text in a large proportion of your images. I see that your images are mostly or entirely printed text – I believe what’s going on here is that Leo has been optimised to extract handwriting from images and, when presented with printed text, a…

The good news is that we’ve fixed the issue for our next model, which is currently training and if all goes well will be released next week.