Posts: 220
Threads: 37
Joined: Apr 2002
Reputation:
0
Can anyone recommend some good cyrillic OCR software. I have a number of books written in Russian. The OCR software I use with my scanner does a pretty good job and saves a hell of a lot of typing should you need to have an electronic copy of something. With a decent OCR program that supports the Russian language, you could scan a page of a Russian book, and then use a machine translator (eg PROMT) to translate the text to English. Any suggestions?
Kind Regards,
Shane Cook.
Posts: 2,150
Threads: 482
Joined: Oct 2003
Reputation:
0
06-14-2004, 02:17 PM
(This post was last modified: 06-14-2004, 02:22 PM by Taz.)
Everybody's equal, But some more than others!
"Those who come to us with the sword - will be killed by the sword" - Alexander Nevski
Posts: 2,293
Threads: 78
Joined: May 2002
Shane,
I am not sure this is such a good idea. Even the best OCRs can achieve 95% accuracy max, and the resulting document needs to be proof-read. If you feed it in the PROMPT without proof-reading you may get total nonsence.
Alexei
Posts: 153
Threads: 9
Joined: Dec 2002
Reputation:
0
I have used prompt for translation, it needs to be heavily edited after, unless you already completley understand Russian, I would not recommend it.
Ron
Posts: 130
Threads: 23
Joined: Aug 2002
Reputation:
0
06-17-2004, 12:56 AM
(This post was last modified: 06-17-2004, 12:59 AM by d-riemer.)
Shane,
for several years, I am using the Russian software ABBYY FineReader 4 for English and German optical character recognition. The character recognition is outstanding, for printed masters usually 100%. For licensing reasons, Russian is not included in my Version 4, while Bulgarian and Ukrainian are. The software has been upgraded and is now at Version 7, which includes Russian.
Search on the Web for "ABBYY FineReader" and you find lots of offers, even on ebay, with dramatic price differences. Since it was developed in Russia, I assume it would do the Russian or Cyrilic character recognition at least as well as it has worked in my experience with the Latin characters and the German umlauts (the software contains dictionaries for the main languages and checks the spelling of each word after recognition, it should, therefore, not be difficult for software to do a word-for-word translation into English that might provide some basic understanding of a text; that is the way I try to translate text now, which requires a lot of time).
I would love to translate Russian or Bulgarian texts but have not advanced to that point yet. Your question reminded me to investigate that field again.
Dietrich
Posts: 2,150
Threads: 482
Joined: Oct 2003
Reputation:
0
The top link I posted is also for this Russian software ABBYY and many other items.
Eddie.
Everybody's equal, But some more than others!
"Those who come to us with the sword - will be killed by the sword" - Alexander Nevski
Posts: 2,293
Threads: 78
Joined: May 2002
Guys,
I have used the ABBYY, and could not get more then 95% accuracy even from a computer print out. It often confuses Russian script b for 6, soft sign for b, etc. Sometimes it would start recognizing Russian characters as English for no apparent reason, and this is the worst when trying to search through the text or use a machine translation software.
Alexei
Posts: 501
Threads: 59
Joined: Nov 2007
Reputation:
0
As far as I can tell, the Russian term for research is "punching" (Пробить, Пробивкa). What is the story behind this word choice?
Alex