Pages in topic:   [1 2] >
New to Cafe Tran
Thread poster: Samuel Keating

Samuel Keating  Identity Verified
Canada
Local time: 18:25
English to French
+ ...
Nov 7, 2018

I can't find how to get to populate the translated text instead of simply copying the original EN.



 

Samuel Keating  Identity Verified
Canada
Local time: 18:25
English to French
+ ...
TOPIC STARTER
A couple more things ... Nov 7, 2018

Also, I'm having trouble figuring out how to get CafeTran to fix spelling issues in real time (for example, automatically add a non-breaking space before "?" and "!" in French or switch from straight quotes ("") to guillemets (« »))

 

Jean Dimitriadis  Identity Verified
France
Member
English to French
+ ...
Some answers Nov 7, 2018

Hello Samuel,

I can’t find how to get to populate the translated text instead of simply copying the original EN.


To prevent source text from being copied:

Edit > Preferences (Options) > Workflow pane: Make sure “Copy source to target during segmentation” and “Transfer source to empty target during translation” are not enabled.

Then, the MT setting you show in your screenshot should work. MT suggestions do not automatically replace already populated segments.

Also, I’m having trouble figuring out how to get CafeTran to fix spelling issues in real time (for example, automatically add a non-breaking space before “?” and “!” in French or switch from straight quotes ("") to guillemets (« »))


CafeTran does not have an AutoCorrect feature as you describe it (to automatically add a non-breaking space when you type the “?” sign, etc.). These need to be typed manually.

You can define double punctuation signs, such as guillemets for insertion, though. For a example, you can have a non-breaking space inserted after and before the guillemets when you insert such pairs:

Edit > Preferences > General > Surround with characters. Example for guillemets: « S » and insert one of the three available punctuation pairs. If you select text, this adds surrounds the selected characters with the defined punctuation signs.

There are dedicated shortcuts for surrounding selected characters with these punctuation signs as well as for bringing the target text context menu for insertion: Edit > Preferences > Keyboard Shortcuts > Show alternative spelling / Show surrounding characters. You can see the shortcut in Edit > Show alternative spelling.

On the other hand, CafeTran has an excellent feature that allows you to quickly correct a mistyped word without returning to it (in most cases). Say you have typed “reguard” instead of “regard”. The word will be underlined in red. You have already typed one or two words and notice that this previous word is mistyped. Just use the “Replace with alternative spelling” shortcut to cycle through the alternative spelling offered by the spellchecker. Shortcut found in Edit > Replace with alternative spelling. Otherwise, you can bring the “Show alternative spelling” while having the cursor on the mistyped word, then select the correct spelling.

You can find various answers in the official knowledge base: https://cafetran.freshdesk.com/support/solutions

Jean

[Edited at 2018-11-08 06:21 GMT]


 

Roy Williams  Identity Verified
Austria
Local time: 02:25
German to English
Experimienting with CafeTrans Nov 20, 2018

I'm also new to Cafe trans and have been playing around with it recently. The problem I'm having is that I can't seem to find the pretranslation function to leverage existing matches from a TM. How does this work in CT?

 

Jean Dimitriadis  Identity Verified
France
Member
English to French
+ ...
*CafeTrans = CafeTran Nov 20, 2018

Hello,

With the translation memories loaded, you can use:

Translate > Insert all exact matches. This action searches all source segments for exact translation matches in your resources. If any exist, CafeTran copies them into the target segments.

For non-exact matches, first check the Memory settings you wish to use in Edit > Preferences > Memory. Check the Auto-assembling insert threshold (%) and the Fuzzy match insert threshold (%). [Auto-assembling is a
... See more
Hello,

With the translation memories loaded, you can use:

Translate > Insert all exact matches. This action searches all source segments for exact translation matches in your resources. If any exist, CafeTran copies them into the target segments.

For non-exact matches, first check the Memory settings you wish to use in Edit > Preferences > Memory. Check the Auto-assembling insert threshold (%) and the Fuzzy match insert threshold (%). [Auto-assembling is a special CafeTran feature that uses and “repairs” various enabled resources to create a suggested translation.] You can also assign Priorities to your TMs and/or keep them out of Auto-assembling.

Fuzzy matches above that threshold are automatically inserted as soon as you open a specific segment. So, unless mistaken, there is no pretranslation for fuzzy segments per se, only for exact segments, but in effect, your segments come pretranslated with your fuzzy matches the moment you access each segment.

If the automatic matching is taking more time than expected, you can perform Preliminary memory matching: https://cafetran.freshdesk.com/support/solutions/articles/6000059860-pre-translation

Finally, you can check the match statistics in Project > Statistics > Memory statistics or Total statistics.

Here is the official CafeTran Espresso knowledge base for more information: https://cafetran.freshdesk.com/support/solutions The above information can be found in the Reference documents listed there.


[Edited at 2018-11-20 15:59 GMT]
Collapse


 

Samuel Keating  Identity Verified
Canada
Local time: 18:25
English to French
+ ...
TOPIC STARTER
Spellcheck Nov 22, 2018

Jean Dimitriadis wrote:

On the other hand, CafeTran has an excellent feature that allows you to quickly correct a mistyped word without returning to it (in most cases). Say you have typed “reguard” instead of “regard”. The word will be underlined in red. You have already typed one or two words and notice that this previous word is mistyped. Just use the “Replace with alternative spelling” shortcut to cycle through the alternative spelling offered by the spellchecker. Shortcut found in Edit > Replace with alternative spelling. Otherwise, you can bring the “Show alternative spelling” while having the cursor on the mistyped word, then select the correct spelling.



Unfortunately, I can't get the spellcheck function to work. It doesn't seem to recognize the target language (FR), so every word is underlined and the suggestions are all English.

Mind you, I wasn't able to find a FR-CA dictionary at the link CafeTran gives (ftp://ftp5.gwdg.de/pub/openoffice/contrib/dictionaries/), but I did install the FR-FR one, and I'd be fine if it used that.



 

Jean Dimitriadis  Identity Verified
France
Member
English to French
+ ...
Spell checker Nov 22, 2018

Hello Samuel,

CafeTran provides two links for installing spell check dictionaries. For French, I won’t use the first OpenOffice link, as LibreOffice is a much more updated alternative.

French can be found in the second link CafeTran provides.

More specifically, the French dictionaries are available here: https://extensions.libreoffi
... See more
Hello Samuel,

CafeTran provides two links for installing spell check dictionaries. For French, I won’t use the first OpenOffice link, as LibreOffice is a much more updated alternative.

French can be found in the second link CafeTran provides.

More specifically, the French dictionaries are available here: https://extensions.libreoffice.org/extensions/dictionnaires-francais

I don’t see French Canadian indeed, but you can use the French one.

Just download the .oxt file. Then, in Edit > Spell check > Install dictionaries > Select your language variant in the drop-down menu. A file selector opens. Point it to the .oxt file. CafeTran will do the rest. Restart CafeTran as you’re being asked, and you are good to go.

NOTE: For the latest version of the Hunspell French dictionaries, instead of using the .oxt file (which dates back to 2017), go to the Dicolecte link included in the French Dictionaries Extension page listed above.

To make things faster, here’s the page: Download the latest Hunspell dictionaries (currently 6.3 - August 2018 version) for French: https://grammalecte.net/download.php?prj=fr and install that .zip instead, as explained above or in the official instructions: https://cafetran.freshdesk.com/support/solutions/articles/6000110210-installing-a-spell-checking-dictionary

NOTE 2:

If you unzip the compressed file, you will notice that Hunspell offers multiple versions of French spell check dictionaries:

- fr-classique
- fr-moderne
- fr-reforme1990
- fr-toutesvariantes

These are explained in the README file inside the ZIP.

I think CafeTran uses the fr-toutesvariantes, but I’m not sure.

If you want to point CafeTran to a specific version, just rename the .dic and .aff files as "fr_CA", copy them into a separate folder, ZIP them and install the ZIP via CafeTran.

To update a spellcheck dictionary, I think you just need to repeat the install steps.

Alternatively, you can do so by going in the directory where CafeTran installs its spellcheck dictionaries. It can be found in the CafeTran Espresso installation folder, then /cafetran/resources/spellchecker. There, replace the already installed .dic and .aff files with the wants you want to use for your language variant. Do not remove the user dictionaries also found there. CafeTran uses those when you add unknown words to your dictionary for that language variant. It also lets you edit it via the Edit menu.

Let me know if that works.

Jean

[Edited at 2018-11-22 06:16 GMT]
Collapse


 

Jean Dimitriadis  Identity Verified
France
Member
English to French
+ ...
[Duplicate post] Nov 22, 2018

[Duplicate post]

[Edited at 2018-11-22 06:11 GMT]


 

Samuel Keating  Identity Verified
Canada
Local time: 18:25
English to French
+ ...
TOPIC STARTER
I wish it were that simple Nov 26, 2018

Jean Dimitriadis wrote:

To prevent source text from being copied:

Edit > Preferences (Options) > Workflow pane: Make sure “Copy source to target during segmentation” and “Transfer source to empty target during translation” are not enabled.

Then, the MT setting you show in your screenshot should work. MT suggestions do not automatically replace already populated segments.



This doesn't work. Even though both of those options are unchecked, I still get the EN copied to the target segment.

Unfortunately, if I try to take a screenshot with the workflow pane open, the screenshot only takes the pane and not the background, so you have to take my word for it ... but I assure you those options are unchecked

I've also rebooted CafeTran several times and moved around segments to make sure that it wasn't just "stuck" at one.

I did notice that the "Automatic transfer to target segments" option in the screenshot I posted above unchecks itself for each new segment. I.e., checking it does nothing for the current segment, and when I move to another segment, it's unchecked again. Could that be the problem?

If so, I haven't found a way to keep it checked.


 

Jean Dimitriadis  Identity Verified
France
Member
English to French
+ ...
It should be Nov 27, 2018

I just did a test and the feature works as expected for me.

First, make sure you have CafeTran's latest version installed.

Also check that Edit > Preferences > Auto-assembling: "Automatic insertion of matches" is NOT enabled.

I then suggest you make this test:

- Create a new native project: In the Dashboard, select your language pair and one or more MT engines. For now, do not attach any TMs. Drag and drop a supported document.
- Press St
... See more
I just did a test and the feature works as expected for me.

First, make sure you have CafeTran's latest version installed.

Also check that Edit > Preferences > Auto-assembling: "Automatic insertion of matches" is NOT enabled.

I then suggest you make this test:

- Create a new native project: In the Dashboard, select your language pair and one or more MT engines. For now, do not attach any TMs. Drag and drop a supported document.
- Press Start. Are the target segments already populated? If yes, there is probably something wrong with the settings. If nothing else works, reset all settings (reset button found in Edit > Preferences). If the target segments are empty, as expected, proceed to the next step.
- Enable "Automatic transfer to target segments" on only one MT engine. If needed, make sure the rest don't have the option enabled. Also, make sure the MT engine you set up with this option is not set as "Stop automatic MT engine". If it is, uncheck this option.
- Proceed to the next few segments. Does the MT engine output populate the target segments? It should. If it does work in the test but not once you attach TMs and other resources, you might need to tweak the Auto-assembling and fuzzy match threshold settings in Edit > Preferences > Memory.

If the above still does not work, you might need to bring this to the developer by creating a support ticket: https://cafetran.freshdesk.com/support/tickets/new (Requires free registration).


[Edited at 2018-11-27 05:51 GMT]
Collapse


 

Roy Williams  Identity Verified
Austria
Local time: 02:25
German to English
Thanks! Nov 27, 2018

Jean Dimitriadis wrote:

Hello,

With the translation memories loaded, you can use:

Translate > Insert all exact matches. This action searches all source segments for exact translation matches in your resources. If any exist, CafeTran copies them into the target segments.

For non-exact matches, first check the Memory settings you wish to use in Edit > Preferences > Memory. Check the Auto-assembling insert threshold (%) and the Fuzzy match insert threshold (%). [Auto-assembling is a special CafeTran feature that uses and “repairs” various enabled resources to create a suggested translation.] You can also assign Priorities to your TMs and/or keep them out of Auto-assembling.

Fuzzy matches above that threshold are automatically inserted as soon as you open a specific segment. So, unless mistaken, there is no pretranslation for fuzzy segments per se, only for exact segments, but in effect, your segments come pretranslated with your fuzzy matches the moment you access each segment.

If the automatic matching is taking more time than expected, you can perform Preliminary memory matching: https://cafetran.freshdesk.com/support/solutions/articles/6000059860-pre-translation

Finally, you can check the match statistics in Project > Statistics > Memory statistics or Total statistics.

Here is the official CafeTran Espresso knowledge base for more information: https://cafetran.freshdesk.com/support/solutions The above information can be found in the Reference documents listed there.


[Edited at 2018-11-20 15:59 GMT]


Thanks for the info!


 

Samuel Keating  Identity Verified
Canada
Local time: 18:25
English to French
+ ...
TOPIC STARTER
Solved! Dec 9, 2018

Jean Dimitriadis wrote:

Also check that Edit > Preferences > Auto-assembling: "Automatic insertion of matches" is NOT enabled.


[Edited at 2018-11-27 05:51 GMT]


I haven't been using CafeTran all that much (except for that one client who insists on being sent back sdlrpx packages, heh) ... but it turned out this was the issue.

For some reason, I missed this advice in your list of things to try, but I noticed that after a while, what CafeTran was inserting wasn't exactly the EN source, but that occasionally one or two words were being translated, and I eventually figured out the culprit was the auto-assembling feature.

I don't remember checking the box for that when I first installed, but I may have ...

Thanks for your help!

[Edited at 2018-12-09 03:44 GMT]


 

Samuel Keating  Identity Verified
Canada
Local time: 18:25
English to French
+ ...
TOPIC STARTER
No grammatical spellcheck? Dec 15, 2018

Jean Dimitriadis wrote:

Hello Samuel,

CafeTran provides two links for installing spell check dictionaries. For French, I won’t use the first OpenOffice link, as LibreOffice is a much more updated alternative.

French can be found in the second link CafeTran provides.

More specifically, the French dictionaries are available here: https://extensions.libreoffice.org/extensions/dictionnaires-francais

I don’t see French Canadian indeed, but you can use the French one.

Just download the .oxt file. Then, in Edit > Spell check > Install dictionaries > Select your language variant in the drop-down menu. A file selector opens. Point it to the .oxt file. CafeTran will do the rest. Restart CafeTran as you’re being asked, and you are good to go.



I was able to get a French dictionary installed from Libre Office.

However, while it seems to flag the most basic misspellings, it completely misses any type of grammatical mistakes (verb forms, number agreement, etc.)




NOTE: For the latest version of the Hunspell French dictionaries, instead of using the .oxt file (which dates back to 2017), go to the Dicolecte link included in the French Dictionaries Extension page listed above.

To make things faster, here’s the page: Download the latest Hunspell dictionaries (currently 6.3 - August 2018 version) for French: https://grammalecte.net/download.php?prj=fr and install that .zip instead, as explained above or in the official instructions: https://cafetran.freshdesk.com/support/solutions/articles/6000110210-installing-a-spell-checking-dictionary



So that's what I did next.

Still no luck.

Here's an example of what I see:

grammar-spellcheck

There are 3 grammatical errors in this string, none of which are picked up.




Alternatively, you can do so by going in the directory where CafeTran installs its spellcheck dictionaries. It can be found in the CafeTran Espresso installation folder, then /cafetran/resources/spellchecker. There, replace the already installed .dic and .aff files with the wants you want to use for your language variant. Do not remove the user dictionaries also found there. CafeTran uses those when you add unknown words to your dictionary for that language variant. It also lets you edit it via the Edit menu.

Let me know if that works.

Jean

[Edited at 2018-11-22 06:16 GMT]


Here are all the dictionaries I see:

spellcheck-dictionaries

I'm sure I'm the one who's doing something wrong, but I don't know what, and the whole system of having to install your own dictionaries (and update them?) isn't really the most user-friendly

[Edited at 2018-12-15 03:31 GMT]


 

Jean Dimitriadis  Identity Verified
France
Member
English to French
+ ...
No built-in grammar check atm Dec 15, 2018

No, there is currently no grammatical check available in CafeTran.

Hunspell (which is a free-libre/open source software) is only a spell checker and a morphological analyzer.

LanguageTool, which could be an open source solution for introducing grammar checks, has not been integrated into the software, at least not yet. This has been requested several times in the past.

At the moment, various solutions exist for grammar checks:

- You can use Oka
... See more
No, there is currently no grammatical check available in CafeTran.

Hunspell (which is a free-libre/open source software) is only a spell checker and a morphological analyzer.

LanguageTool, which could be an open source solution for introducing grammar checks, has not been integrated into the software, at least not yet. This has been requested several times in the past.

At the moment, various solutions exist for grammar checks:

- You can use Okapi's CheckMate (free-libre/open source software) which integrates with LanguageTool. It offers QA checks on bilingual documents (XLIFF, etc.), so you can run a check in real-time.

- You can export a DOCX bilingual document (Project > Export > Export as bilingual document), then open it in Word or LibreOffice and use any additional spelling and grammar checks available in these tools.

LanguageTool offers add-ons for Word and LibreOffice (among other programs).

For French, a probably superior free-libre software solution is Grammalecte - https://grammalecte.net/ which can be integrated with LibreOffice. I suggest you try this one if you use LibreOffice (LanguageTool and Grammalecte cannot be enabled simultaneously).

If you are ready to invest in a complete solution, I would highly recommend the Druide Antidote software - http://www.antidote.info/ (corrector, dictionaries and language guides). If you buy the French and the English modules, you will also enjoy bilingual translations.

[Edited at 2018-12-15 06:23 GMT]
Collapse


 

Hans Lenting  Identity Verified
Netherlands
Member (2006)
German to Dutch
Grammarly Dec 15, 2018

Jean Dimitriadis wrote:

If you are ready to invest in a complete solution, I would highly recommend the Druide Antidote software - http://www.antidote.info/ (corrector, dictionaries and language guides). If you buy the French and the English modules, you will also enjoy bilingual translations.


For English there is also Grammarly: https://www.grammarly.com/native


 
Pages in topic:   [1 2] >


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

Moderator(s) of this forum
Natalie[Call to this topic]

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

New to Cafe Tran

Advanced search






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 »
SDL Trados Studio 2019 Freelance
The leading translation software used by over 250,000 translators.

SDL Trados Studio 2019 has evolved to bring translators a brand new experience. Designed with user experience at its core, Studio 2019 transforms how new users get up and running and helps experienced users make the most of the powerful features.

More info »



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