Anagnostis – OCR for ancient Greek

I had an email from Dominique Gonnet at the Sources Chretiennes, asking if I had a copy of the OCR program Anagnostis.  I’d never heard of it, but this turns out to be an OCR utility for Greek which can handle Ancient Greek! 

Of course I want one!  But I was quickly put off, by the unbelievable price-tag of 585 euros for the ‘pro’ version — almost £400 or $800.  A ‘standard’ version exists for 78 euros, but this doesn’t do ancient Greek, so is useless to us all.  I don’t really think that I can justify spending such a very large sum.  I wonder if anyone else has seen or used this product?  If so, how well did it work?  A trial download is available: but I have not tried this, and I have seen a post suggesting that it doesn’t let you try the OCR!

I’ve also written to the company suggesting that they do an ‘ancient Greek’-only version for our market, for 78 euros. You might like to do the same, if you think that you’d buy such a product.

Share

16 thoughts on “Anagnostis – OCR for ancient Greek

  1. I have the demo. I am disappointed in the program, as we all are used to having the availability to check the spelling to to catch the problems within the program itself. This does not have a dictionary. Each word has to be looked at individually. I am using a rougher paper and have the brightness turned up as far as it will go. Still has problems.

    I cannot check that it will load multiple tiffs or scans as the demo requires that you save the first page, which is disabled.

    It does come with a manual. Though if they are calling the Help a manual, well then it really is not helpful.

    You have to know Ancient Greek to be able to work with the out put. It does have a training part, which I have not used yet.

  2. Thank you for these comments! Nice to get some info. I don’t tend to use the dictionary, I admit, and expect to check each word. But presumably it highlights when it isn’t sure about a word? What’s the accuracy like?

  3. I could put texts in Ancient Greek into OCR. My bid isn’t high. If you need it, send me a sample. And I will prove I can do it.

  4. Roger, I work for Ideatech, the maker of Anagnostis, and I would like to take the chance of your post to explain a few things:
    a. Anagnostis is a VERY niche product while it required many man years of development. After almost 20 years of operation, the development cost has not been matched with equal revenues. This explains the high price tag.
    b. Regarding Kathy’s comment about spell checking, well, I one knows greek, it is easy to understand that this is not possible: the Greek language texts span a period of 3000 years. In this period, syntax, grammar and meaning have largely changed, new words have been added and others deleted. For a spell checking program, to make meaningful suggestions, it has to take into account the era the text was written. A spelling checker for classical Greek has nothing to do with a spelling checker for Modern Greek in polytoniko. Yet, both are greek, same script, and both are polytonico. We have long discussed of undertaking this project but simply the market demand does not justify the cost.
    I hope this helps to better understand our situation.

  5. Of course, if they do an ‘ancient Greek’-only version for 78 euros, I will buy such a product. Me too !

  6. Has anyone used ABBYY Fine Reader? If so, what version? Some of their readers are supposed to be pretty good for ancient Greek, but they have cheap-ish engines and more expensive ones, and I’m not sure whether the cheap ones are any good.

    I note that OCR Opus/Tesseract is also supposed to be pretty good. It’s open source (i.e., free, and available on Google Code), but very alpha and only runs on a Linux platform. (Maybe in a year or two or three there will at least be a beta version, or something that runs on Windows.)

    There’s a very technical article about this all on Perseus at: http://www.perseus.tufts.edu/~ababeu/ecdl2009-preprint.pdf

  7. I use Finereader 10 Pro a lot, but I’m not aware that it supports Ancient Greek. I’m willing to learn different, of course!

  8. Is it possible to use a scanned immage of ancient greek with “Agnostic” software and print it with Basic-D (a Braille Printer for blind people)?

  9. I don’t understand why one should need an OCR for Ancient Greek. Just find the text on TLG and modify the text to meet the edition you have at hand. The number of modifications you will have to implement will always be less than the corrections you will have to make if you scan it, even with the best OCR program.

  10. I used FineReader to scan ancient Greek but some accents are not regognised. All the times I wanted – while correcting the Greek text – to introduce the needed accents for ancient greek I was told that such accents are not accepted.
    Anagnostis … is a good start. Only that is too expensive and the demo edition doesn’t allow to save the file.
    Hope …. for the future!

  11. Dear Sirs…….. Allow me to present to you some suggestions:

    1) The program is much too expensive.
    2) Allow to see the results of your OCR by permitting to see it in Word….. in doc……… at least a few lines.

    Thanks a lot for your wonderful work
    [Edited by RP to convert to lower case]

  12. I presume Rolando’s comment is for the manufacturers of Anagnostis.

    I have never been able to bring myself to purchase it, and I have rather a lot of OCR software. The reason has always been price.

    Export to Word is certainly essential, if it isn’t present.

Leave a Reply