Spellchecking Latin texts?

[Carolus Raeticus]

Salvete!

Transcribing texts is quite a bit of work, and even though one tries to avoid typos or at least get rid of them during the proofreading phase, some of them are always being missed. The main reason for this is that one gets blind to errors, especially if you are familiar with the text. The brain automatically edits the perception (quite a feat, actually!).

For example, it turned out that my Herodotus-transcription had (at least) 88 typos (mostly letter swaps and spacing errors). Some of these I was able to weed out using regular expressions, but I would like to improve quality even more (and more efficiently). There are programmes for Windows, but since I use Linux I had to do something else...

I used Whitaker's WORDS as a spellchecker. It has an option named WRITE_UNKNOWNS_TO_FILE. When set to "Y" and letting WORDS work on an input file, all unknown words are written to a file. This approach is far from perfect. It creates a lot of false positives, but it helped me a lot and adds another layer of checking a Latin text without adding too much work (you can mostly simply breeze through the list without examining the actual text).

Since this approach is working to a certain degree but is less than perfect I am wondering if anyone knows how to check Latin texts (ideally not using Windows-tools): software, specific regular expressions, etc. Any idea?

Valete,

Carolus Raeticus

[truks]

I've had success with the Latin spellchecker which you can download separately for use in LibreOffice. There's also a good spellchecker for ancient Greek available for LibreOffice. As far as I know, LibreOffice runs on Linux.

https://www.libreoffice.org/download/download/

[Carolus Raeticus]

Thank you for the link. I will give it a try.

Vale,

Carolus Raeticus

[Jandar]

Someone mentioned a spellchecker here.