What plugins is recomended to install (OmegaT) ? Thread poster: Vitor Pinheiro
|
|---|
Its being hard to fully configure OmegaT. There are so many technical steps to follow, and there are no tutorials on how to fully configure OmegaT, only pieces of tutorials, wich is making the configuration process very long for me.
So far i did installed apertium plugin from https://sourceforge.net/p/omegat/wiki/Plugins/ , but there are several other plugins, and i dont know w... See more Its being hard to fully configure OmegaT. There are so many technical steps to follow, and there are no tutorials on how to fully configure OmegaT, only pieces of tutorials, wich is making the configuration process very long for me.
So far i did installed apertium plugin from https://sourceforge.net/p/omegat/wiki/Plugins/ , but there are several other plugins, and i dont know wich ones i should use.
I'm doing translation, transcreation and transcription. Sometimes the text are in some image format. And i only work with 1 language peer (portuguese-english).
What plugins (or even other tools, since we're alredy here ...) do you recommend to me ? ▲ Collapse
| | | | Susan Welsh 
United States
Local time: 17:38
Russian to English
+ ...
You don't need any plug-ins to install OmegaT. It works "out of the box." Plug-ins are all optional, depending on your needs. I'm afraid that's the only thing I can say, since I don't use them much.
| | | | Vitor Pinheiro
Brazil
Local time: 18:38
English to Portuguese
+ ...
TOPIC STARTER
Susan Welsh wrote:
You don't need any plug-ins to install OmegaT. It works "out of the box." Plug-ins are all optional, depending on your needs. I'm afraid that's the only thing I can say, since I don't use them much.
True, but whithout plugins, aparently, there arent some usefull features, like automatic translation, autocompletion, dictionaries. So i'm searching and trying those plugins one by one, wich is being kinda a long process for me. Like, just now i had to remove the automatic translation apertium plugin, because it didnt have the portuguese-english peer. So any tips would be very welcomed.
| | | | Samuel Murray
Netherlands
Local time: 23:38
Member (2006)
English to Afrikaans
+ ...
Vitor Pinheiro wrote:
I'm doing translation, transcreation and transcription. Sometimes the text are in some image format. And I only work with 1 language pair (Portuguese-English). ... What plugins do you recommend to me?
For what you want, no additional plugins are needed. Many years ago, many of the features needed to be installed separately as plugins. There are still plugins, but you don't need many of them.
If you're going to translate XLIFF files, install the Okapi plugin:
https://okapiframework.org/wiki/index.php/Okapi_Filters_Plugin_for_OmegaT
and disable OmegaT's built-in XLIFF filter in Options > File Filters.
In addition to plugins, there are also scripts that you can install, but OmegaT does come bundled with some scripts already.
Make sure you have the latest version of OmegaT: 5.7.0
https://sourceforge.net/projects/omegat/files/OmegaT%20-%20Latest/OmegaT%205.7.0/
OmegaT won't be of any use with transcription and image translation.
| | |
|
|
|
| Plugin for local MT: OPUS-CAT | Mar 8, 2022 |
Vitor Pinheiro wrote: So far i did installed apertium plugin from https://sourceforge.net/p/omegat/wiki/Plugins/ , but there are several other plugins, and i dont know wich ones i should use. What plugins (or even other tools, since we're alredy here ...) do you recommend to me ?
Plugin for local MT for OmegaT: OPUS-CAT (OPUS-MT)
Portuguese models (all models are only in one direction; for reverse pair you can download another models)
PT-EN, PT-CA, PT-EO, PT-Galician, PT-Tagalong, PT-Ukrainian.
I am using the models with Czech.
No limits and no fees. All is running in my PC.
I very often use a script "write_untranslated2file.groovy" after automated translation from TMXs and for translation in eTranslation.
Note to Apertium: I do not use it, there no good JAR with Czech language. I have got PL-CS engine in a year 2014 but developing is not continuing. The engine is bi-directional and work also without OmegaT.
Milan
| | | | Samuel Murray
Netherlands
Local time: 23:38
Member (2006)
English to Afrikaans
+ ...
Vitor Pinheiro wrote:
True, but without plugins, apparently, there aren't some useful features, like automatic translation, autocompletion, dictionaries.
Tell us which version of OmegaT you're using, and which user guide you're using (some user guides are old).
Spelling dictionaries must be installed separately, that is true. And since some of the stemming features use spelling dictionaries, those specific features won't work without the spelling dictionary installed.
And yes, getting hold of a spelling dictionary is a nightmare. I just spent 15 minutes (!) googling in search of an English Hunspell spelling dictionary and finally got one that isn't very old or some strange file extension (but that's because I knew exactly what I was looking for, which isn't something new users would be able to say).
For English, go here:
https://github.com/marcoagpinto/aoo-mozilla-en-dict
Click the green Code button and select Download ZIP. It's 17 MB, unfortunately, even though the dictionaries themselves are just over a megabyte. Then copy the relevant DIC and AFF files to the OmegaT spellchecking dictionary folder. Let us know if you don't know where that folder is.
For Portuguese, the only one I could find is this one:
https://github.com/wooorm/dictionaries
Again, click the green Code button and select Download ZIP. This time, it's 50 MB. There are some English dictionaries in there but they are quite old. Again, copy the relevant DIC and AFF files to the OmegaT spellchecking dictionary folder -- I assume you'll need both PT (which in this file in BR) and PT-PT. I have no idea how old the PT-PT one is, but the PT-BR one is from 2013, so fairly recent.
Autocompletion is built-in to the latest version of OmegaT.
Machine translation is also built-in with OmegaT these days. But you need a key from each of the providers, and those are usually not free. I believe MyMemory is free if you register an account with them: https://mymemory.translated.net/doc/keygen.php
Dictionaries (i.e. dictionaries that tell you the meaning or translation of words) must be created or found separately, that's true. And not all of them work with OmegaT, so it's a matter of trial and error (and asking people on the OmegaT forums if they have specific recommendations).
So I'm searching and trying those plugins one by one, which is being kinda a long process for me.
I understand, but you don't need to do that. Apart from the Okapi plugin and the relevant spelling dictionaries, you don't really NEED anything else.
Like, just now I had to remove the automatic translation Apertium plugin, because it didn't have the Portuguese-English pair.
You could have found that out by visiting the Apertium page and going to their documentation:
https://wiki.apertium.org/wiki/Main_Page
| | | | | MyMemory and MateCat | Mar 9, 2022 |
Samuel Murray wrote: Machine translation is also built-in with OmegaT these days. But you need a key from each of the providers, and those are usually not free. I believe MyMemory is free if you register an account with them: https://mymemory.translated.net/doc/keygen.php
I think that generated key is for translator who will working for agencies using MateCat. Registering form is asking resume and I was offered by a default rate 0,035 €/word for EN-CS pair.
I was not completed the registration and I tried generated KEY in OmegaT. The number of translated segments is the same if I work without KEY, only with registered email address.
Milan
| | | | frytex
Local time: 23:38
English to Polish
+ ...
| extensions and external features | Mar 10, 2022 |
Vitor Pinheiro wrote:
True, but whithout plugins, aparently, there arent some usefull features, like automatic translation, autocompletion, dictionaries. So i'm searching and trying those plugins one by one, wich is being kinda a long process for me. Like, just now i had to remove the automatic translation apertium plugin, because it didnt have the portuguese-english peer. So any tips would be very welcomed.
automatic translation: you need api from key from google, microsoft, etc.
i believe google api is free for some limited (quite big, though) amount of characters per day. microsoft wants money. i know nothing about other machine translation API-s because i don't use them. i can help you if you want.
autocompletion: i don't know what you mean. the only autocompletion i come accross is inserting glossary entries with Enter while typying a few first letters. but you need a glossary - again, it's an external thing.
spelling: the link to download spelling dicts from openoffice (in the spell config window) doesn't work but all you have to do is to find webpages with files .dic and .aff and place them manually in the dictionaries folder. basically, you can download them from here: https://extensions.openoffice.org/en/search?f[0]=field_project_tags:188 (oxt files) change the file extension to zip and extract dic and aff from inside.
dictionaries: it's similar. you need files with dictionaries. some are here: http://download.huzheng.org/misc2/ - you have to look for something similar for your language.
| | |
|
|
|
Vitor Pinheiro
Brazil
Local time: 18:38
English to Portuguese
+ ...
TOPIC STARTER | Thanks a lot, i just have some more doubts | Mar 10, 2022 |
Samuel Murray wrote:
Tell us which version of OmegaT you're using, and which user guide you're using (some user guides are old).
I just downloaded the 5.7, as you suggested.
Samuel Murray wrote:
And yes, getting hold of a spelling dictionary is a nightmare. I just spent 15 minutes (!) googling in search of an English Hunspell spelling dictionary and finally got one that isn't very old or some strange file extension (but that's because I knew exactly what I was looking for, which isn't something new users would be able to say).
I'm a very new user then, because i would never figure this out. Thanks, both spelling dicts are working perfectly !
frytex wrote:
autocompletion: i don't know what you mean. the only autocompletion i come accross is inserting glossary entries with Enter while typing a few first letters. but you need a glossary - again, it's an external thing.
So the built-in auto-completion will use the glossary as a database for predicting words ? Would it use the dictionaries as well ? For me the auto-completion is predicting all words i already typed, at least in the present project. Would be interesting if it predicts all words typed in all projects.
Milan Condak wrote: Samuel Murray wrote: Machine translation is also built-in with OmegaT these days. But you need a key from each of the providers, and those are usually not free. I believe MyMemory is free if you register an account with them: https://mymemory.translated.net/doc/keygen.php I think that generated key is for translator who will working for agencies using MateCat. Registering form is asking resume and I was offered by a default rate 0,035 €/word for EN-CS pair. I was not completed the registration and I tried generated KEY in OmegaT. The number of translated segments is the same if I work without KEY, only with registered email address. Milan
Thanks, MyMemory seems to be indeed free (up to 10.000 words/month https://rapidapi.com/translated/api/mymemory-translation-memory/pricing). And apparently there is a plugin called google-tanslate-whitout-api-key ... (https://sourceforge.net/projects/omegat-gt-without-api-key/files/) Don't know how it works, but its working.
frytex wrote: dictionaries: it's similar. you need files with dictionaries. some are here: http://download.huzheng.org/misc2/ - you have to look for something similar for your language.
Great, i downloaded 3 portuguese-english dicts. If i use diferent dicts for the same peer, there would be some kinda of incompatibility ? Or they would just merge into more word definitions ?
Milan Condak wrote:
Plugin for local MT for OmegaT: OPUS-CAT (OPUS-MT)
Is it better then the built-in MT for predicting sentences ? Because it's platform is windows, and mine is linux. I am searching for other MT plugins now that you mention this one, but is it really that much better then the built-in MT ? Because, although i just started using OmegaT, the built-in MT is predicting some sentences sometimes.
[Edited at 2022-03-10 15:05 GMT]
| | | | | Congratulations | Mar 10, 2022 |
MyMemory max. 100 segments, next:
429: Too Many Requests
google-tanslate-whitout-api-key = 5 segments; if I was new plugin user it worked fine.
Milan
| | | | Samuel Murray
Netherlands
Local time: 23:38
Member (2006)
English to Afrikaans
+ ...
frytex wrote:
I believe Google API is free for some limited (quite big, though) amount of characters per day.
You need to sign up with your billing details, and you do pay, but it is very, very cheap.
Vitor Pinheiro wrote:
So the built-in auto-completion will use the glossary as a database for predicting words ?
No. There are a number of different types of autocompletion. They are either activated automatically (Options > Autocompletion), or you can activate it manually by pressing Ctrl+space (and if you press Ctrl+space several times, it cycles through the various options). Glossary autocompletion is one of them. There is also "history completion" which I suspect is based on the TM, but I'm not 100% sure.
Milan Condak wrote:
google-tanslate-whitout-api-key = 5 segments; if I was new plugin user it worked fine.
I would be surprised if such a plugin works. Or, if it continues to work year in and year out, since Google may have fixed the bug that allows the plugin to work.
[Edited at 2022-03-11 09:19 GMT]
| | | | frytex
Local time: 23:38
English to Polish
+ ...
| More than one dictionary, simultaneously | Mar 13, 2022 |
frytex wrote:
autocompletion: i don't know what you mean. the only autocompletion i come accross is inserting glossary entries with Enter while typing a few first letters. but you need a glossary - again, it's an external thing.
So the built-in auto-completion will use the glossary as a database for predicting words ? Would it use the dictionaries as well ? For me the auto-completion is predicting all words i already typed, at least in the present project. Would be interesting if it predicts all words typed in all projects.
I don't think all projects will be included in predictions. They are separate. And as the time goes by you would have more and more useless predictions if they were cumulative. But note that you can copy and paste an old tmx file into the new project (folders tm and tm\auto) so the fuzzy matches would show up and/or autopaste to the segment.
Great, i downloaded 3 portuguese-english dicts. If i use diferent dicts for the same peer, there would be some kinda of incompatibility ? Or they would just merge into more word definitions ?
I use just one, if any. I just chose the one that has a reasonable word base and a nice layout.
because, you see, some dictios have a lot of empty space beetween the lines (a lot of scrolling required) and they provide, for instance, a lot of useless information, like examples of how to use a given word in a sentence. I don't really need that.
and, as basically all words from a segment are searched by the dictio, you will get a very long useless list. at least I don't need examples.
also: when you connect a dictio to your project you will also notice that each of your segments will need extra time to load. it takes up to 3 - 5 seconds with big dicionaries. you will feel it even with only one dictio (if a big one) and it would probably get worse with more than one. so, choose wisely. I think that after some practice you will just stick to one.
and above all, in my case, the machine translation from google/microsoft is usually good enough. even if you don't decide to use the entire sentence (if poorly synthetized), you will probably find that it will be good enough to find the word that you don't know. thanks to that I don't use dictios, just machine translation. the only exception is a medical dictio if I translate some medical texts.
| | |
|
|
|
frytex
Local time: 23:38
English to Polish
+ ...
| auto completion, scripts and some more hints | Mar 13, 2022 |
So the built-in auto-completion will use the glossary as a database for predicting words ?
you have 3 options in the auto-completion menu. one of them is glossary. the other two are history completion and history prediction. just turn them on and check what they do.
one more remark: you have some built-in scripts that come with our OmegaT.
you might want to assign them to the keys Ctrl+Shift+ F2, F3 and so on because you will probably use them frequently.
the ones I found most interesting are:
open last project
open glossary file
open project folder
spellcheck (there's a built-in spell check and underlining but this alternate script does it in a nice window)
strip tags (if you paste a fuzzy match that matches all the words but has different tags - this script would remove tags and you can insert proper ones again with Ctrl+T)
show same segments (in which translation = source); that's for proofreading
write notes (I downloaded it; it makes an external file with all notes that I typed into the project)
open current file (in Word - if I want to see pictures, for instance)
tagwipe (wipe tags from poorly formatted docx documents)
by the way: if you experience to many tags in your segments, do the folowing: open your source file in word, mark the whole text and turn off the following things: automatic language detection (just set manually the proper language), turn off spell and grammar check, turn off automatic hyphenation.
than save the file and reload your project. you should see less tags. if this doesn't help, use the tagwipe script (I recommend level 1 or 2, not higher).
also, twice in my life I experienced the following problem: the source file wouldn't open or word reported it as broken. I thought my all project was damaged and gone.
in both cases the reason appeared to be the same: there were signs < and > in the text itself (like in this sentence: "the temperature should be < 10°C") - somehow OmegaT treated this symbol as part of the tag. and Word couldn't handle it.
so if you face this problem just eliminate the symbols from your translation and from your documents and generate them again. this will help.
| | | | | | Vitor Pinheiro
Brazil
Local time: 18:38
English to Portuguese
+ ...
TOPIC STARTER
Thanks you all, it really saved me tons of searching time !
| | | | There is no moderator assigned specifically to this forum. To report site rules violations or get help, please contact site staff » What plugins is recomended to install (OmegaT) ? | TM-Town | Manage your TMs and Terms ... and boost your translation business
Are you ready for something fresh in the industry? TM-Town is a unique new site for you -- the freelance translator -- to store, manage and share translation memories (TMs) and glossaries...and potentially meet new clients on the basis of your prior work.
More info » |
| | Anycount & Translation Office 3000 | Translation Office 3000
Translation Office 3000 is an advanced accounting tool for freelance translators and small agencies. TO3000 easily and seamlessly integrates with the business life of professional freelance translators.
More info » |
|
| | | | X Sign in to your ProZ.com account... | | | | | |
Login to reply/comment 
