I read that Stephen Hawking used some kind of infrared sensor on his glasses that sensed his cheek muscle moving which would then speak though a voice synthesizer, but how did that work? How would the computer know which word he was trying to type/speak if he could only move his cheek muscle?
In: 4
iirc, on screen keyboard hooked to a sensor following his eyes and prediction texts. Cheek sensor was the confirm button.
He was confined to a wheel chair, but that doesn’t mean he couldn’t move at all. He used his thumb to move a joystick and he blinked to select words.
He did a both of the things mentioned above.
Originally, he used his thumb and a blink switch. Then he lost use of his hands and they made they eye sensor and cheek sensor.
A lot of people have already covered the methods he used, but you should also know that pretty much any time he was interviewed he was given the questions well in advance, that way he could have time to prepare his answers and just have them queued up for the interview. Otherwise having a conversation with him was pretty slow as he had to type up his responses pretty slowly.
The sensor was probably an electrode like the ones used for electromyography (EMG) studies, which can detect the tiny electrical pulses that travel through the nerves to the muscles. It also could be some sort of tiny strain sensor which would detect the change in shape of his cheek as he flexed it.
In terms of software, he would probably be using something like the T9 auto-correct that used to be used on non-smartphones, which can predict and suggest words based on a few button presses (or cheek flexes). If he really was limited to a single action (such as flexing a cheek), the software might present each letter in the alphabet one at a time, and a flex would select the letter. If he had more than one muscle he could control, he might have been able to use a cursor or pointer to select letters from an on screen keyboard.
In any case, communicating would have been incredibly slow. When you see him in an interview or public speaking engagement, he would have prepared his statements or answers long before the event, and triggered the computer to speak the pre-composed words at the appropriate time.
Latest Answers