Best way to create a termbase from OCR input?

On 2003-01-26 11:50, fvg wrote:

Which of the CAT progs is best suited for creating a termbase from OCR input?

(...)I\'m thinking of scanning it, and then getting it into a termbase format. (I\'ve done a lot of OCR, so I don\'t anticipate problems in identifying the fields)

But I\'m a beginner with the CAT progs, and I\'d be most grateful for information on the following:

which of the CAT progs would be best suited for generating such a termbase? (I haven\'t bought one yet, since the availability of my most important dictionary is really what it all hinges on.)



Ideas anyone? I\'d be most grateful.



regards,

Dr. Frederik van Gelder





Wordfast will accept any 3 columns tab delimited text file or xls as glossary. So if you work with Wordfast, realize that, by the time you have a table, you have a fully searchable glossary as well.



If you end up working with Trados, know that you can create a multiterm file by importing the table you get.



All major CATs (DV, Wordfast, Trados, SDLX, Transit ...) have some sort of a termbase, and all of these can import tab/comma delimited text - which is what you get after OCR, I guess.



To my knowledge, the easiest way to work with and to handle terminology in is Wordfast. Totally open formats, and you can use your glossaries to check your translations in real time.



But in all cases, all major CATs can use your OCR output - just a matter of how much manipulations you will have to do to fit it in. (Trados Multiterm import function is a pain, imo).



Beside CATs, there are a number of dictionary applications out there, and most will handle tab delimited text as well.



Really, the point is more to get a feel of the different existing CATs, and see which one you are most comfortable with. They can all do the stuff you need.



Hope this helps ▲ Collapse