Sound: AR's Unsung Modality
arinsider.co
arinsider.co
So I've been working really hard on audio interactions for people with low- or no-vision. I'm finding that, for simple stuff, it's completely possible to build 3-DOF, maybe even 6-DOF, interfaces, completely audio-based.
One of the first things I did was create a system to make sure that every clickable thing in the scene had interaction sounds. When you hover over something that is interactable, it makes a "fwip" sound (in addition to animating the graphic). When you leave the object, it makes a "fwup" sound. The leep in usability from this simple feedback of "this is interactable" was pretty huge.
I've integrated good speech-to-text and text-to-speech (thanks, Microsoft Azure Cognitive Services! Easy to use, fast, and extremely high quality!). That has gone a long way to really improving the experience, even while I use it as a person with full-vision.
But I've also been experimenting with some things like making the currently selected object hum a little bit. The further away from the center of your view it is, the more it hums, thereby anchoring it in your spatial reference, even when you can't see it. With the limited FOV of VR and AR headsets, this is helping to overcome some of the object permanence issues and helping find lost things.
I'm also considering making "the closest non-selected object in view" hum in a different tone, growing louder the closer you get to it. So when you are scanning around an area, you'd hear objects come into and out of view before actually needing to hover over them.
Anyone, just some of my thoughts.