Hello everyone.

The Mixed Reality Forums here are no longer being used or maintained.

There are a few other places we would like to direct you to for support, both from Microsoft and from the community.

The first way we want to connect with you is our mixed reality developer program, which you can sign up for at https://aka.ms/IWantMR.

For technical questions, please use Stack Overflow, and tag your questions using either hololens or windows-mixed-reality.

If you want to join in discussions, please do so in the HoloDevelopers Slack, which you can join by going to https://aka.ms/holodevelopers, or in our Microsoft Tech Communities forums at https://techcommunity.microsoft.com/t5/mixed-reality/ct-p/MicrosoftMixedReality.

And always feel free to hit us up on Twitter @MxdRealityDev.

How to use GrammarRecognizer?

I'm trying to use GrammarRecognizer but I hadn't sucessed. The Microsoft examples at HoloAcadamy wasn't enough, anybody have some Unity simple example for me study?

Tagged:

Best Answer

Answers

  • Are you trying for KeywordRecognizer? Just want to make sure we are talking about the same thing before I start trying to confuse you :smile:

  • @Jesse_McCulloch said:
    Are you trying for KeywordRecognizer? Just want to make sure we are talking about the same thing before I start trying to confuse you :smile:

    Yes, I do. I used KeywordRecognizer and it's worked. But I didn't understand how to use GrammarRecognizer, especialy the part of SRGS file.

  • Thanks for the clarification, I hadn't seen the documentation for GrammarRecognizer. That's cool stuff.  I think my approach to this would be using LUIS from cognitive services, but that's it's own topic...

  • @utekai said:
    You have to provide your own srgs file, and it works off the rules in that file. The demo provides a sample of an srgs ruleset. The idea is it allows commanding with variable information (for instance, you want to change the color of an object and let the user say which color, but there are many objects and many colors). Compared to keyword recognizer which doesn't allow variable information in the command, with srgs ruleset, you can gather variable information. Srgs rules can be tricky though, but there are many examples available around the web for gathering dates, numbers and other variable information. Done well, srgs allows natural language processing with multiple tasks in one statement (for instance a user could say 'book a room for two adults and two children in a 4 star hotel in for two nights in Shanghai on january 12th, 2016'). Processing that with keyword recognizer would be not practical, but srgs makes it rather easy.

    The doc page explains it fairly nicely ...

    I'm from Brazil and want to use Brasilian Portugues language in my aplication, so I need to change the dictionary or some thing like that to recognizer the portugues pronunciation. I have a theory, I intend to use GrammarRecognizer to solve this problem. I saw that in the SRGS file you can set the languague, so I'm trying this right now. I'll return with the result of this experiment.

  • utekaiutekai ✭✭✭

    I've read that only only English is supported currently. Just because SRGS rules allow it doesn't mean the current software supports it. Then again, it might in particular if you can find language packs for the HoloLens (haven't seen any yet). Somehow the delays when using Cortana and other voice services makes their appeal and usefulness much less when attempting to provide real-life-like experiences to users. That Cortana delay with the unpredictability gets in the way, though provided speech recognition services work well enough at least in English.

Sign In or Register to comment.