Translation memory matches Thread poster: Tai Fu
|
|---|
Tai Fu 
United States
Local time: 17:16
English to Chinese
+ ...
CAT tools are great for software localization or other related translation assignments when you have a ton of repetitions. It really helps with consistency. It works great for 100% matches where you don't need to translate them more than once (leading to consistency issues because there's more than one way to translate something). However the problem comes up when the match is less than 100%. MemoQ like all other CAT tools fills in the blank for you even when there's a 70% match, the problem is ... See more CAT tools are great for software localization or other related translation assignments when you have a ton of repetitions. It really helps with consistency. It works great for 100% matches where you don't need to translate them more than once (leading to consistency issues because there's more than one way to translate something). However the problem comes up when the match is less than 100%. MemoQ like all other CAT tools fills in the blank for you even when there's a 70% match, the problem is sometimes a small mismatch require you to write the whole statement in a different way for the statement to fit the original, so I question any assistance I get when the program just automatically fills in the blank for me when the match is less than 100%.
Is there any way to tell the program to NOT fill in the blank unless there's a 100% match?
I'm using MemoQ 6.2 by the way. ▲ Collapse
| | | | TheAlphaSens (X)
United Kingdom
Local time: 01:16
English to French
> Translation
> Automatic lookup and insertion
> Raise the "Minimum match rate" of section 1. to 95% or so
| | | | | Pre-translate option | Jun 19, 2013 |
Hello,
When you have the source text is open, you should click on Operations/Pre-translate.
Here choose exact match (=100% match) or exact match with context (101% match).
Click OK. This should only fill in the segments with 100%+ matches.
I hope this answers your question.
Best,
Edina
| | | | Tai Fu
United States
Local time: 17:16
English to Chinese
+ ...
TOPIC STARTER
I did that, and all it really did was run through "pre-translate" but not change how fields get automatically filled in by the software from what's in the translation memory (I would ctrl-enter anytime I finish a segment to get it into the TM). What I mean is I don't want the software auto-filling anything unless it's a 100% match or better...
I must be missing something.
| | |
|
|
|
Tai Fu
United States
Local time: 17:16
English to Chinese
+ ...
TOPIC STARTER
TheAlphaSense wrote:
> Translation
> Automatic lookup and insertion
> Raise the "Minimum match rate" of section 1. to 95% or so
Thanks!
| | | | Premium✍️
United States
Local time: 17:16
Member (2006)
English to French
+ ...
| How do we get all segments automatically filled? | Jun 19, 2013 |
Edina Gugyella wrote:
Hello,
When you have the source text is open, you should click on Operations/Pre-translate.
Here choose exact match (=100% match) or exact match with context (101% match).
Click OK. This should only fill in the segments with 100%+ matches.
I hope this answers your question.
Best,
Edina
Hi everybody,
I am a happy user of memoQ 6.2 and 6.5 now. However I always wonder why all segments in the translation pane are not filled automatically after the Pre-translate option is selected. After all, that's what PRE-translate suggests. For example, I am working on two documents, I did what Edina and TheAlphaSense suggested, but I still have to confirm each translation result manually. The results from my TMs and aligned documents do appear instantly, but I have to confirm each and every one EVERY time.
In Fluency 2013, once you run the MT and save your TM for a project, the translations are generated and propagated for ALL segments! Which is a true marvel and a fantastic time-saving trick!!
| | | | Tai Fu
United States
Local time: 17:16
English to Chinese
+ ...
TOPIC STARTER
I don't really use the MT function, just never realized any sort of time saving I'd get from it. The danger I see is missing crucial terms the MT doesn't pick up or not included in the translation memory. Oftentimes I would see an automatically filled in segment with a 70% (or so) match from that entry in the translation memory but without examining the segment you wouldn't know which 70% matched or didn't match, and if you don't manually confirm you run the risk of providing incorrect translati... See more I don't really use the MT function, just never realized any sort of time saving I'd get from it. The danger I see is missing crucial terms the MT doesn't pick up or not included in the translation memory. Oftentimes I would see an automatically filled in segment with a 70% (or so) match from that entry in the translation memory but without examining the segment you wouldn't know which 70% matched or didn't match, and if you don't manually confirm you run the risk of providing incorrect translation.
You should always manually confirm any TM auto-fills anyways to make sure it's not putting things in the wrong context. Especially in languages such as Chinese meanings can change dramatically depending on context, so a 70% match might as well be no match at all.
On an unrelated note I really wish all Chinese-English translation assignments are in traditional... but then the Red China has the bigger market... ▲ Collapse
| | | | Premium✍️
United States
Local time: 17:16
Member (2006)
English to French
+ ...
| What did you do? | Jun 19, 2013 |
Tai Fu wrote:
However the problem comes up when the match is less than 100%. MemoQ like all other CAT tools fills in the blank for you even when there's a 70% match, .....when the program just automatically fills in the blank for me when the match is less than 100%.
Hi Tai Fu,
How do you get the blanks automatically filled? That's what I want, but they are staring BLANK at me.
| | |
|
|
|
Tai Fu
United States
Local time: 17:16
English to Chinese
+ ...
TOPIC STARTER
MultiPro wrote: Tai Fu wrote:
However the problem comes up when the match is less than 100%. MemoQ like all other CAT tools fills in the blank for you even when there's a 70% match, .....when the program just automatically fills in the blank for me when the match is less than 100%. Hi Tai Fu, How do you get the blanks automatically filled? That's what I want, but they are staring BLANK at me.
You need to press CTRL-ENTER after you finish each segment. This puts that entry into the translation memory, so if the exact same (or similar) segment comes up, it automatically appears in the box. You can then modify them or whatever. Now if its multiple documents within the same project then they all carry.
Be careful with TM's however because terms do not always carry across fields! TM's for medical texts would look silly in a software localization project, unless it's a game that deals with medical issues.
| | | | Premium✍️
United States
Local time: 17:16
Member (2006)
English to French
+ ...
| MTs for Blanks | Jun 19, 2013 |
Tai Fu wrote:
I don't really use the MT function, just never realized any sort of time saving I'd get from it. The danger I see is missing crucial terms the MT doesn't pick up or not included in the translation memory. Oftentimes I would see an automatically filled in segment with a 70% (or so) match from that entry in the translation memory but without examining the segment you wouldn't know which 70% matched or didn't match, and if you don't manually confirm you run the risk of providing incorrect translation.
Agree, I use MT very sparingly to get the gist, a first draft and save some keystrokes. But it is always questionable, and I end up rewording, changing whole sentence structures and make spelling corrections, verb agreements, tense agreements and oh yes, GENDER agreements corrections, especially in French!!:) I do appreciate however the speed and time saved. I'm extremely cautious terminology-wise. It goes without saying that I constantly check the context. In the end, I get my blanks filled already. That is why I use MT actually to get a preliminary translated document that I use for my aligned docs from Fluency to MemoQ. I just overwrite any nonsense spewed by the MT, fine tune the writings and finalize the translation.
You should always manually confirm any TM auto-fills anyways to make sure it's not putting things in the wrong context.
This is not a new translation. I originally confirmed each entry manually in the previous translations. Actually just now, for testing purposes, I am pretending this is a new translation. I used my aligned previously translated documents so I didn't have to reconfirm the entries/translation results again. I chose the Pre-translate option. But the translation pane is ALL blank. All entries have already been committed to this project TM. That's why I am puzzled as to why I have to RE-RE-RE-confirm the results for the SAME source document, SAME words, SAME sentences.
My question is in that case, is there a way to have the blanks FILLED automatically since the TM has confirmed entries already and previously translated docs are being used as LIVE docs?
Thanks in advance to any soul who will have the answer.
Have a great week!
[Edited at 2013-06-19 21:56 GMT]
| | | | TheAlphaSens (X)
United Kingdom
Local time: 01:16
English to French
My take is that you have selected "Exact match" while the matches in your TM/Livedocs are actually 95%-ish or less.
Select "Good match" rather than Exact match.
>Operations
>Pretranslate
>Good match
If you select "Exact match" and the matches in your TM are not exact 100%, they won't be inserted. Edina's tips and my tips were specifically directed at Tai fu's request, you should not have followed the procedure I have described above, what you wa... See more My take is that you have selected "Exact match" while the matches in your TM/Livedocs are actually 95%-ish or less.
Select "Good match" rather than Exact match.
>Operations
>Pretranslate
>Good match
If you select "Exact match" and the matches in your TM are not exact 100%, they won't be inserted. Edina's tips and my tips were specifically directed at Tai fu's request, you should not have followed the procedure I have described above, what you want to do is completely different (if you did, revert back to the default settings).
Note: you can use MT directly from memoQ, you don't have to use Fluency first to MT your files.
[Edited at 2013-06-19 20:05 GMT]
[Edited at 2013-06-20 01:28 GMT] ▲ Collapse
| | | | Tai Fu
United States
Local time: 17:16
English to Chinese
+ ...
TOPIC STARTER
By the way is there any way in MemoQ to change the simplified chinese in the source into traditional chinese characters? Not really translation but just changing the characters itself so I can read it better...
| | |
|
|
|
I chose the Pre-translate option. But the translation pane is ALL blank. All entries have already been committed to this project TM. That's why I am puzzled as to why I have to RE-RE-RE-confirm the results for the SAME source document, SAME words, SAME sentences.
Again in Operations/Pre-translate, note that there is a second tab called confirm/lock.
If you select exact match with context(101%) or exact match (100%), the pre-translate option will confirm these se... See more I chose the Pre-translate option. But the translation pane is ALL blank. All entries have already been committed to this project TM. That's why I am puzzled as to why I have to RE-RE-RE-confirm the results for the SAME source document, SAME words, SAME sentences.
Again in Operations/Pre-translate, note that there is a second tab called confirm/lock.
If you select exact match with context(101%) or exact match (100%), the pre-translate option will confirm these segments, i.e. make them jolly green.
So you don't have to confirm these again. However, it doesn't work for matches that are 99% or less, obviously.
Hope this saves you time!
Best,
Edina ▲ Collapse
| | | | To report site rules violations or get help, contact a site moderator: You can also contact site staff by submitting a support request » Translation memory matches | Trados Studio 2022 Freelance | The leading translation software used by over 270,000 translators.
Designed with your feedback in mind, Trados Studio 2022 delivers an unrivalled, powerful desktop
and cloud solution, empowering you to work in the most efficient and cost-effective way.
More info » |
| | Trados Business Manager Lite | Create customer quotes and invoices from within Trados Studio
Trados Business Manager Lite helps to simplify and speed up some of the daily tasks, such as invoicing and reporting, associated with running your freelance translation business.
More info » |
|
| | | | X Sign in to your ProZ.com account... | | | | | |
Login to reply/comment 
