Auto-suggest on Wordfast 5
Автор темы: Lucia Messuti
Lucia Messuti
Lucia Messuti  Identity Verified
Италия
Local time: 17:38
английский => итальянский
+ ...
May 26, 2018

Hello Collegues,

I would like to use the auto-suggest feature, but it only works with glossary and capitalized words/acronyms (and only if the are not attached to another word or a dot).

I selected also TM, but I don't get any suggestion from the TM, sub-segmetns as well, although there are many repetitions.

Do you know how I can manage this?

Thanks
Lucia


 
Edward Potter
Edward Potter  Identity Verified
США
Local time: 11:38
Член ProZ.com c 2003
испанский => английский
+ ...
Glossary May 28, 2018

Hi Lucia.

Set up a few words using the Glossary function. Play around with that a bit.

Let us know how it goes.

Edward


 
Lucia Messuti
Lucia Messuti  Identity Verified
Италия
Local time: 17:38
английский => итальянский
+ ...
Автор темы
TM subsegments, not Glossary May 28, 2018

Hi Edward,

Thanks for your tip, I commonly use the autosuggest feature with the glossary, but I refer to a different function of this feature, that is the "sub-segment" suggestion.

The system suggests you smaller segments from the TM, in the same way (and view) of the glossary.

I had this feature in WF4, but not in WF5. And this is an interesting feature, especially if you have large TMs.

Lucia


Samuel Keating
 
Samuel Keating
Samuel Keating  Identity Verified
Канада
Local time: 15:38
английский => французский
+ ...
Same here Aug 26, 2018

I updated from WF3 to 5 this month, and the only TM / TM sub-segments it suggests are ridiculous things like after I'm done typing "the" up pop suggestions for "The", and after I'm done typing "wine" up pop suggestions for "Wine", "ready" gives "Ready", etc.

-- I have no idea where those come from (it does that to a good number of words, though not by any means all of them), or why the first letter is invariably capped, or why it thinks I'd want to change from a non-capped word to
... See more
I updated from WF3 to 5 this month, and the only TM / TM sub-segments it suggests are ridiculous things like after I'm done typing "the" up pop suggestions for "The", and after I'm done typing "wine" up pop suggestions for "Wine", "ready" gives "Ready", etc.

-- I have no idea where those come from (it does that to a good number of words, though not by any means all of them), or why the first letter is invariably capped, or why it thinks I'd want to change from a non-capped word to a capped word in the middle of a sentence. Since I usually use at least 2 TMs, there are at least 2 identical irrelevant suggestions.

Even worse, if there's actually a glossary entry for the word I'm trying to type, that appears underneath any of the other, useless, ones.

I imported TMs that have 10s of thousands of TUs, so I wondered at first if WF somehow hadn't parsed them to extract suggestions that would actually be useful, and maybe it would just be a matter of time.

Then I had a segment that was exactly identical to half of a previous segment, and it offered nothing ... so as far as I can tell, the feature is a joke, and I've unchecked the "TM" and "TM sub-segment" boxes.

AFAICT, the auto-suggest feature also doesn't work for numbers, although it did in 3, and the documentation suggests it should in 5 as well.

***
FWIW, I did a really large project a few months ago using Studio for the first time (the client gave me a 1-month license). Working from a blank TM, within the first hour or so, the suggestions became useful, and within a few days, they were a huge help.

Studio would even correct simple issues on its own. EG, if the TM match wasn't bold, instead of suggesting text in a regular font, or giving you a 99% match that you then had to bold, it would go ahead and fix it to bold on its own.

I haven't ditched WF yet ... but I was very happy when I thought that they'd have something similar. Unfortunately ...
Collapse


Nathalie Lahaie-Feldberg
 
Samuel Keating
Samuel Keating  Identity Verified
Канада
Local time: 15:38
английский => французский
+ ...
Screenshots Aug 26, 2018







Why, why, why?

[Edited at 2018-08-26 19:52 GMT]


Stefan Stefanov
 


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


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

Auto-suggest on Wordfast 5







LinguaCore
AI Translation at Your Fingertips

The underlying LLM technology of LinguaCore offers AI translations of unprecedented quality. Quick and simple. Add a human linguistic review at the end for expert-level quality at a fraction of the cost and time.

More info »
Pastey
Your smart companion app

Pastey is an innovative desktop application that bridges the gap between human expertise and artificial intelligence. With intuitive keyboard shortcuts, Pastey transforms your source text into AI-powered draft translations.

Find out more »