1 reply to this topic

[javaftper]

Hey,

Have any high level tetras with limited hand function tried the OCZ NIA neural impulse actuator? What do you think?

From what i've read it could be a good replacement for mouse support which is not great in voice recognition systems and i'm not a fan of the IR head or mouth controls.

Website says - "The headband uses carbon nanofiber-based sensors to provide the highest possible dynamic range for the recording of bioelectrical signals that are amplified and digitized and further de-convoluted into computer commands."

Link here if you have n't seen it- http://www.amazon.co..._neural_impulse

youtube-

btw- after using dragon naturally speaking 9 & 10 on PC/windows vista for about 18 months i have recently moved to macspeech 1.5.2 on mac osx 10.6 which is amazing! very accurate like dragon naturally speaking 10 but a better interface and very fast translation rate. It is expensive - RRP £199 but play have it on sale at the moment for £120.

thanks,
// j

Attached Thumbnails

• 

[Mike Ellison]

javaftper, on Oct 2 2009, 10:46 PM, said:

Hey,

Have any high level tetras with limited hand function tried the OCZ NIA neural impulse actuator? What do you think?

I have one I've played with some. So far, I haven't been able to train the brain waves well enough to use, although I've probably only spent 4-8 hours working on it at this point. It has a 'motor' sensor (best I can tell is galvanic response?) that I can trigger reliably by clenching my jaw or raising an eyebrow.

With mine, I have to have the black box touching skin for anything to work. There's some kind of grounding issue with it otherwise that's been reported in forums quite a bit - if it's not grounded, the inputs all tend to float. And those edges are pretty but sharp! You might be able to clip a grounding strap to it and yourself.

The software is strongly geared towards games. Not sure how easy it would be to use it general-purpose. They didn't provide particularly useful information for developers last time I checked, but there's some people hacking away at it and publishing their results (http://code.google.c.../TechnicalSpecs). I kinda got the feeling that OCZ was specifically avoiding competing with http://www.brainfingers.com/, whether by contract or marketing desire.

My expectation is that some people would have a much easier time using the brain waves than me (it's reading 'em, but mine seem to go 'burst mode' a lot even after recalibrating a gazillion times and rechecking the electrodes... end result is either it reads in the low range or is off the charts), but it'll take time to train regardless. I'd view it as a toy worth exploring if you have the time to dedicate to it, but at best a beta version.