Looking for a CAT tool, which has heart for Uralic-Altaic language family!
Thread poster: Dagdelen

Dagdelen  Identity Verified
Turkey
Local time: 19:49
Member (2010)
German to Turkish
+ ...
Sep 11, 2010

I'm a native Turk, do translations between German/sometimes English to Turkish & vice versa. I have used WF pro & Across v.5.0 previously.
I'd like to have a cat tool, which can provide me consistent solutions between those language families, stays durable, has (at least input/output) compatibility with Trados and doesn't cost 3.000 € (I mean star transit).
The specialty/deviation of this language group is agglutinating of the words/verbs with suffix; words are getting bigger and bigger, it never ends. I suppose that my across (earlier WF) cannot match my TM's and terminologies I've saved with enlarged Turkish sentences/terminologies in the target and the opposite.
So the question is, which one has the most powerful TM engine regarding executing vigorously translations between those two different language families (Indo/European - Uralic/Altaic)?


Direct link Reply with quote
 

Didier Briel  Identity Verified
France
Local time: 18:49
Member (2007)
English to French
+ ...
OmegaT has a Turkish tokenizer Sep 11, 2010

Rasim Dagdelen wrote:
I'm a native Turk, do translations between German/sometimes English to Turkish & vice versa. I have used WF pro & Across v.5.0 previously.
I'd like to have a cat tool, which can provide me consistent solutions between those language families, stays durable, has (at least input/output) compatibility with Trados and doesn't cost 3.000 € (I mean star transit).
The specialty/deviation of this language group is agglutinating of the words/verbs with suffix; words are getting bigger and bigger, it never ends. I suppose that my across (earlier WF) cannot match my TM's and terminologies I've saved with enlarged Turkish sentences/terminologies in the target and the opposite.
So the question is, which one has the most powerful TM engine regarding executing vigorously translations between those two different language families (Indo/European - Uralic/Altaic)?

Omegat can use a Turkish tokenizer (stemmer), which should enhance segment and glossary matches when the source text is Turkish. (There are also German and English ones.)

In a ProZ message, you can read the feedback of a Turkish user:
it seems to work very well with Turkish. It even detects non-trivial inflexions.

But the only direct compatibility with Trados is at the TMX (translation memory) level.

Didier


Direct link Reply with quote
 

Dagdelen  Identity Verified
Turkey
Local time: 19:49
Member (2010)
German to Turkish
+ ...
TOPIC STARTER
OmegaT Sep 11, 2010

Hi Didier!
That was the word/function I really needed "tokenizer or lexical analyzer"! A Cat tool should be able to analyse the words/sentences lexical, find the root word/verb of swelled up sentences of Turkish/Hungarian/similar languages out and show me my fuzzies/matches in order to give some clues!
WF & Across have problems finding segment and glossary matches when the source text is Turkish.
Direct compatibility with Trados via TMX files is O.K.
Do you mean OmegaT is stable/reliable enough, to earn my baguette with it? Does it work ergonomic?


Direct link Reply with quote
 

Piotr Bienkowski  Identity Verified
Poland
Local time: 18:49
Member (2005)
English to Polish
+ ...
You could try Swordfish Sep 11, 2010

It is not free software but works with all features for a month.

It is compatible with Trados formats and if you want better matching of very long words, you should optimize the TM for quality when you first set it up.

You can find out more about Swordfish at: www.maxprograms.com

Regards,

Piotr


Direct link Reply with quote
 

Grzegorz Gryc  Identity Verified
Local time: 18:49
French to Polish
+ ...
memoQ Sep 11, 2010

Rasim Dagdelen wrote:

I'm a native Turk, do translations between German/sometimes English to Turkish & vice versa. I have used WF pro & Across v.5.0 previously.
I'd like to have a cat tool, which can provide me consistent solutions between those language families, stays durable, has (at least input/output) compatibility with Trados and doesn't cost 3.000 € (I mean star transit).
The specialty/deviation of this language group is agglutinating of the words/verbs with suffix; words are getting bigger and bigger, it never ends. I suppose that my across (earlier WF) cannot match my TM's and terminologies I've saved with enlarged Turkish sentences/terminologies in the target and the opposite.
So the question is, which one has the most powerful TM engine regarding executing vigorously translations between those two different language families (Indo/European - Uralic/Altaic)?


Take a look on memoQ.
It is developped by Hungarians and the terminology handling is based mainly on radicals and stemming.
The default matching setting is 50% of the word length but you can also select absolute (always detect the match independently of the length) and custom level (the pipe is used to separate the variable and invariable part of the word).
You can also force the exact recognition.
For words with internal alternances, you can group different radicals in one entry.
Of course, it works also with multiword entries (the custom mode is ideal for heavily inflected indoeuropean languages as Polish).

The recognition mode is defined at the word level, so you can tune in your termbases in a very precise way, including for very short words.
E.g. the the Polish forms pies, psa, psu etc. (Nom., Gen. and Dat. sg for "dog") may be easily recognized using variants (pies, exact match) and patterns (e.g. ps|a, custom).
In completely impossible in Trados (Multiterm) unless you enter practically all the variants of this word (psu, psie, psy, psom etc.) and use a very low matching level in order to detect some forms (psach, psami).

Cheers
GG


Direct link Reply with quote
 

Didier Briel  Identity Verified
France
Local time: 18:49
Member (2007)
English to French
+ ...
Some people work with it Sep 11, 2010

Rasim Dagdelen wrote:
Do you mean OmegaT is stable/reliable enough,

OmegaT is usually very stable.
Even the "beta" version is quite reliable

to earn my baguette with it?

A number of professional translators are using it exclusively.
There are 1400 members in the Yahoo support group, so you may want to ask opinions there.

Does it work ergonomic?

As usual with tastes, your mileage may vary.
Some dislike it, some like it.

As it is free software, you can evaluate and test it as long as you want.

Didier


Direct link Reply with quote
 
xxxOlaf
Local time: 18:49
English to German
OmegaT is only for hard core users Sep 12, 2010

Didier Briel wrote:
There are 1400 members in the Yahoo support group, so you may want to ask opinions there.

That doesn't mean that all of them actually use the software.

Didier Briel wrote:
Does it work ergonomic?

As usual with tastes, your mileage may vary.
Some dislike it, some like it.


In my book, OmegaT would easily win the price for the least intuitive CAT tool out there. Since the OmegaT developers most likely never did any usability studies, they should at least have a close look at the GUIs of leading commercial CAT tools and implement the best features into the OmegaT GUI, instead of forcing the user to accept their view of what a GUI should look like. As it is right now, it is at best a proof of concept CAT tool that only hardcore users will want to use. And that's actually a pity, because under the hood most of the needed CAT functionality is already there.


Direct link Reply with quote
 

Dagdelen  Identity Verified
Turkey
Local time: 19:49
Member (2010)
German to Turkish
+ ...
TOPIC STARTER
Biting the hand that feed, bad CATs (from Grzegorz Gryc) Sep 12, 2010

Thank U all for the response!

I 'm gonna try MemoQ+/Swordfish for a while and take a look at OmegaT; I hope I can feed them with my TMX files to check out, if they match my never ending Turkish words in dependence of radicals and stemming.
A cat tool must work efficiently, user friendly (it means I must exclude trados) & should find the matches without entering all variants of a single word, till it gets the right thing (I'm almost getting crazy because of this subject with WF, Across … & U say it is the same with Trados). They have immense problems to find something, so I use Across, as if it was a kind of MS Word and organizing tool for translations.
I don't think that it is a big deal for a cat tool, to realize the matches, if there are only neighbored languages like EN-DEU-FR&….
MemoQ sounds plausible. Maybe I should go to memoQ, it is written by Hungarians, and they should have the same weird problem like us. To choose the right product of memoQ can be also another problem; memoQ translator pro via proz + annual upgrade fee or memoQ translator standard for a year.
I hope it is ergonomic and compatible with Trados.
Let's see what happens!
Ciao
Rasim


Direct link Reply with quote
 


To report site rules violations or get help, contact a site moderator:


You can also contact site staff by submitting a support request »

Looking for a CAT tool, which has heart for Uralic-Altaic language family!

Advanced search







SDL MultiTerm 2017
Guarantee a unified, consistent and high-quality translation with terminology software by the industry leaders.

SDL MultiTerm 2017 allows translators to create one central location to store and manage multilingual terminology, and with SDL MultiTerm Extract 2017 you can automatically create term lists from your existing documentation to save time.

More info »
LSP.expert
You’re a freelance translator? LSP.expert helps you manage your daily translation jobs. It’s easy, fast and secure.

How about you start tracking translation jobs and sending invoices in minutes? You can also manage your clients and generate reports about your business activities. So you always keep a clear view on your planning, AND you get a free 30 day trial period!

More info »



Forums
  • All of ProZ.com
  • Term search
  • Jobs
  • Forums
  • Multiple search