Hmm, that's a shame, as I find will Google Translate and Microsoft Translator very good sometimes in my pair/direction, and am happily using them especially with MT AutoSuggest.

Are there any plans to also allow adaptive MT to also work with other MT providers, or is there a specific technical/commercial reason why this is not possible?

Michael

Hi Álvaro,

I think you are probably misunderstanding what this is... I think? AdaptiveMT is machine translation that starts of as a basis by using SDL Language Cloud which you have today. So it doesn't matter what other resources you have as this works independently of those.

All you would do is correct the machine translation and then two things would happen when you confirm it:

1. The corrected TU would be sent to your Translation Memory as normal
2. The machine translation engine would take note of your correction so it offered the corrected translation next time

Now, you may well ask why this is useful given you already have this in your TM now? The reason it's useful is because the correction would be applied to other similar segments that are not translated yet but would benefit from the correction you have made. The machine translation will start to adapt to the way you are translating with improved translations for segments that you do not find in your TM.

So your TM/Termbases/Dictionaries have nothing to do with this. It's not the same as training an Engine with thousands or TUs that you have translated in the past. This is much more dynamic.

Hope I managed to explain that?

On the trial version... yes of course, there will be a trial version at some point and all users will be able to test it if they wish. It's much too soon at the moment as we have still not reached the point where we have the production candidate.

Regards

Paul

Unknown said:

Are there any plans to also allow adaptive MT to also work with other MT providers, or is there a specific technical/commercial reason why this is not possible?

Not commercial at all.  Entirely technical.  If you think about it the MT engine has to be able to do a few things (at least!):

1. Accept corrected translations

2. Use the corrections to adapt the translation algorithms

3. Do this in realtime

All of these things are outside of our control.  So if other MT providers developed a solution that could handle something like this then it would be perfectly feasible.  Otherwise, it won't happen.

Regards

Paul

I see. Hmm, well, I'll just have too wait and see how good your SDL Language Cloud system is. my options seem to be the following:

(1) stick with Google Translate/Microsoft Translator, which has better quality (and which is already great inside Studio together with MT AutoSuggest)

(2) use SDL Language Cloud + AdaptiveMT (maybe AdaptiveMT will make up for the engine's lower quality) 

(3) try Lilt (which, although developed by ex-Google people, says they don't use the Google Translate engine, and have developed their own)

(4) use Slate Desktop (which I own a licence for), to create my own engines and use them via the plugin inside SDL Studio. however, so far, I have never been able to create one that is even remotely better than Google Translate

Michael

Unknown said:

Well, you do say that the basis is SDL Language Cloud, which is another source of MT, so I don't understand (from my newb point of view, of course) why another source of MT can be used as a basis, and then grow that MT individually with the adaptive MT function.

Because Google, Microsoft and anyone else would have to correct their MT based on what you change.  They don't do that!!  This is out of our hands altogether.  SDL Language Cloud is ours and we take the correction and then tweak the SDL Language Cloud MT based on your personal changes so that it's better next time for you.

The SDL Language Cloud limits will probably be similar to what they are now.  So you get reasonable use for free and for any more you pay.

Regards

Paul