Forum Discussion
VOICE SDK
Hello, I am using the Meta - Voice SDK - Immersive Voice Commands in my Unity project. The package waits to listen for about 5 seconds when the user is not speaking. I want to extend the waiting time...
With dictation, you can call Dictation.Activate(). Where 'Dictation' is your Meta AppDictitationExperience game object. And Dictation.Deactivate() when you want it to stop listening.
I don’t want to completely disable the voice response feature. Let me explain with an example: When the user enters the game, I ask them a question like “Would you like to watch a video with me?” The user is expected to answer with either “yes” or “no.”
In my game, I want to wait for their response for at least 10 seconds or more even if the user doesn’t speak immediately. However, the SDK currently keeps the listening duration quite short (around 5 seconds), and stops listening if there is no input.
I would like to increase this silence timeout so the system keeps listening longer, even when the user remains silent for a while before responding.
I have done that. And it can be done. In the 'App Dictation Experience' Meta component, in the inspector, is a 'Keep Alive Time In Seconds', 'Min Transaction Keep Alive Time in Seconds', and 'Max Recording Time'.
As a test, I set them all to 300 seconds. Seems to work.
Look at their 'DictationSampleScene'. You can tweak the knobs to do that. I just tried and it works.
Hello again, I’m using the AppVoiceExperience script. Even if I update the values, the listening time when I stay silent doesn’t change.
Am I missing something? I tried opening an empty project, but it doesn't change either. I can upload my empty project to GitHub. Can you help me?I have not used that component. Its docs say it has a limited and fixed timeout (and the docs do seem outdated and do not match the components code.).
https://developers.meta.com/horizon/documentation/unity/voice-sdk-activation
I would step into the code in the debugger and see if there is a way to copy that component, then tweak the copy to increase the timeouts.
It does seem to call the same mid and lower level API's as App Dictation Experience. So tweaking seems possible.
Question/Thoughts:
Your code responds to the user pressing a button - Correct?
If it does, call dictation, get the string, then pass that via the App Voice Experience Activate(string...) method?
And I just noticed a 'Always Record' checkbox. I have not tweaked with that one.