[ExI] EPOC EEG headset

[Bryan Bishop]

2010/1/22  <ablainey at aol.com>:
> I didn't reply to Bryans question as I am already exporing the typing speed
> question and wanted a good answer for him. However I must point out that the
> Epoc has a limited number of triggers, not enough for a full alphabet. So as
> such it is not designed for direct typing.
> I am looking at a way around this using mouse based, predictive typing
> software. I know the exact piece of software I need and as soon as I get my
> hands on it, I will begin testing and let you know how fast it is. If it
> works I will make a video so you can see how it works. If it does work as
> predicted, It could spell the death of the keyboard! replaced by thought
> activated typing.

Mouse-based typing sucks immensely. Most EEG headsets get you up to
15wpm. I can do 120wpm on a QWERTY, and 150wpm on a (very) good day.
But if you insist on a mouse, try this on for size:

http://www.inference.phy.cam.ac.uk/dasher/
downloads: http://www.inference.phy.cam.ac.uk/dasher/Download.html

> For those that do. This is (I am told by Emotiv) the most cognitiv actions
> used for an application so far. the control of the arm uses 8 congnitiv
> (discrete triggering thoughts) and 2 facial movements to control 10 seperate
> movements of the arm.

What's the scaling factor? How many commands can you squeeze out per
channel that you add to an EEG device? I know you can do more with
better signal processing, but I haven't been able to find scholarly
literature on this.

- Bryan
http://heybryan.org/
1 512 203 0507