Since I heard of Stephen Hawking's condition I have been thinking about  
ways for him to "speak in real-time".

The first method I thought of is the following:
Using the same methods as speech recognition but by monitoring brainwaves  
(cognitive science).
It might be possible to create a hat or a wig with electrodes connected to  
a computer that monitor brainwaves.
The computer would have to be taught each word by recording the patterns  
of the brainwaves while thinking the word. This would have to be done many  
times for each word to be plausible.

The second method builds on the first, but requires imagining the sound of  
what you want to say. If you use your imagination to imagine a specific  
sound there might just be a specific part of the brain that gives off  
brainwaves at those frequencies. For instance if you listen to the note  
middle C (261.626Hz) being played on a piano. Afterwards you use your  
imagination to imagine the same sound, would an eeg device be able to pick  
up that frequency? If this is works, it should be possible to identify the  
specific place in the brain that uses "sound imagination" and to use this  
directly (no computer required - only an amplifier!).

I don't know if any of this has been tried yet, but I just think it might  
work.


-- 
-Nekar Xenos-

"The spoon is not real"

Post a reply to this message