Microsoft developing virtual headphone algorithm

Considering that Vista isn't exactly getting perfectly positive praise, it looks like Microsoft is looking into other ways to bring cashflow to Redmond. While we already know the firm is diving head first into the VoIP handset market, it looks like it'll also be involved with a new virtual headphone. In an effort to allow VoIP / computer-based communications to be handled without tethering yourself to your PC and also allow you to hear ambient conversations around the office, researchers at Microsoft have "developed an algorithm that adjusts the timing of sound waves emitted from each speaker in an array, creating a focused beam of sound that acts as virtual headphones." Essentially, the speakers would create a "sweet spot" so that computer users could hear the audio perfectly, while individuals just inches away from the sound zone wouldn't hear much more than a peep. Furthermore, the focused wave technology could even bleed over into the tracking realm, which could actually allow the tones to follow one around as they move. Ideally, the creators want to conjure up a beamforming system that is easy to configure and relatively inexpensive, but we're hearing (ahem) that it'll be at least three years before these aural luxuries break into the corporate sphere.

[Olivier @ Mar 26th 2007 7:47AM]

What would be great is if you could "inverse" this technology; i.e. I would like my home stereo to mute the outside noise coming into my appartment.
All the honking, the shouting etc.. Can They get that to work?

[NHAnimator @ Mar 26th 2007 7:55AM]

...or make it so that only people under the age of 18 could hear Justin Timberlake.

[Vinny @ Mar 29th 2007 9:23PM]

Or better yet, just make it so no one can hear Justin Timberlake... ever.

[bloodangel @ Mar 26th 2007 7:59AM]

@Oliver

Get a sound proof room? probably be cheaper

[jeremiads @ Mar 26th 2007 8:38AM]

If Vista was selling beyond Microsoft's expectations, they would still be looking into other technology. Thanks for letting the bias show through more, Engadget. I registered to comment just because of how disgusting I've been at the lack of journalism here.

[Vilppi @ Mar 26th 2007 9:57AM]

Jeremiads:
> If Vista was selling beyond Microsoft's expectations, they would
> still be looking into other technology. Thanks for letting the
> bias show through more, Engadget. I registered to comment just
> because of how disgusting I've been at the lack of journalism
> here.

Yes, you're right; you've been disgusting. You registered to comment on an insignificant Blog-entry in a largely insignificant blog. "Blog" being the operative word here, you really should realize that there can be no "lack of journalism" in a blog, as journalism has nothing to do with blogging and you really shouldn't let anyone tell you otherwise.

[Derek @ Mar 26th 2007 9:03AM]

@Olivier - try a search for "noise cancellation technology" and you'll find it exists already.

@Microsoft (not that I think they'd read this, but) - This isn't a new idea - it's just a variation on the LRAD (Noisemaker) concept. I think the army want to be digging out their patents and suing Microsoft.

[nate w @ Mar 26th 2007 9:22AM]

microsoft? audio? ha. seriously, they should stick to what they do best. hold on a second and something will come to mind...

but really, good luck with that. the physics of what they "want" to do is pretty ridiculous. its not that hard to create a "sweet spot", but making it so that you theoretically only hear it in that one spot and not anywhere else is pretty out there. as far as i know, sound isn't really inclined to have that small of a path. it really wants to go out. i mean we're talking about air pressure. can you really control that to the extent that's required for this? i'm guessing not from microsoft. try meyer or a company that's actually in the music industry.

[Sylvr @ Mar 26th 2007 10:21AM]

So they developed their own version of:
http://www.engadget.com/2007/02/15/yamahas-ysp-900-digital-sound-projector-a-5-1-sound-bouncer-th/

[max andrews @ Mar 26th 2007 10:45AM]

So what's wrong with headphones?

There is a more novel application of this technology that's related to holographic sound reproduction. Basicially, they alter the timing of certain signals so that they interfere with each other at various points, creating a 3D sound field that you can walk through and experience. Way cooler and more novel than just focusing sound on a tiny spot. For that, why not just stick with headphones?

On an unrelated note, I'm really craving the lingonberry juice from the ikea cafeteria.

[max andrews @ Mar 26th 2007 10:53AM]

So what's wrong with headphones?

[Ian Harris @ Mar 26th 2007 11:48AM]

Directed audio has already been done using ultrasound. Holosonics are already shipping 'em too - check out http://www.holosonics.com.

[kn0w1 @ Mar 26th 2007 12:04PM]

so ... like these ultrasonic beam things (http://www.atcsd.com/), but not ultrasonic?

[rp @ Mar 26th 2007 1:54PM]

I'd like to see the frequency response of those holosonics. If Microsoft really does intend to create a headphone-killing array, they will probably have to sacrifice quite a bit of frequency response. I really doubt the frequency response will be anywhere near what you get with even the free headphones you get with a generic portable cd player.
But, as the article points out, Microsoft wants this to be used for VoIP/computer-based communication technology, so as long as they can provide adequate coverage of the vocal frequency range, things should be alright. After all, traditional phone service does the same thing, only over a handset. Don't expect to get any drum and bass out of this thing though.

[Zak @ Mar 26th 2007 2:14PM]

I totally had this idea first. Where's my money, Microsoft?

[rob @ Mar 26th 2007 10:38PM]

Huh. Talk about lighting a stack of $100 bills on fire. Maybe Microsoft is going to buy Holosonics.

[Fran Taylor @ Mar 26th 2007 11:35PM]

It's been done. The MIT Media Lab did this a long time ago.

See the exhibit at the DeCordova Museum.

I'm sure Microsoft will patent it and claim that there's no prior art.

[DM78 @ Mar 27th 2007 2:22AM]

Wow... they're about 10 years too late. Aureal had a generalized consumer version of this in 1997.

http://en.wikipedia.org/wiki/A3D

[Eon @ Mar 27th 2007 5:54AM]

The technology to do this is quite simple. Even though I prefer my headphones I think it's great to be able to have a consumer version of this technology. I thought of building this myself but you probably need good testing equipment.

nate w my guess is that they will achieve this by controlled use of constructive and destructive interference of sound waves so they will probably use multiple sound sources (drivers) to create different wave patterns simultaneously accross the same space.

The simplest form of such an audio beam forming device would be an array of tiny drivers on a flat square or disk and they all get the same audio signal. This will naturally form good on-axis response and a falloff to the sides. To compensate for the fact that this beaming works better on higher frequencies and to focus on differently shaped regions (other than a conical beam) all sorts of weightings and delays can be added to the audio signals for the different drivers and at different frequencies.

rp I agree that they will probably have to filter out low frequencies (where the wavelength is close to the size of the entire device) because you need a very big device to effectively focus the sound. And even then I doubt if the device will be perfect but I'd be interested to see how well this works.

[Eric @ Mar 27th 2007 8:58AM]

1. It's not simple
2. It's not based on LRAD or ultrasound
3. It's called Wavefield Sysnthesis

http://en.wikipedia.org/wiki/Wavefield_synthesis

oh, and microsoft are obviously not the first

[Loonie @ Mar 27th 2007 9:19AM]

Yes folks, Microsoft is reinventing the sound lens.

[kelly @ Apr 3rd 2007 5:41PM]

Yeah someone already beat microsoft to it and bought the domain name www.virtualheadphone.com

lol