On television shows where I’ve seen him, his lips and tongue don’t appear to be moving at all, yet his thoughts are being transferred as a computerized voice.
How?
(And does anybody else think they could make it sound a lot more human, but he actually likes it in that computerized sound, as kind of his image?)
I believe the voice is Apple Text-To-Speech Software voice “Fred”, but that’s all I can help you with.
“Fred” sounds about right. In one of his books-on-tape he jokes that he has an American voice synthesizer and he hasn’t yet taught it to talk it proper English.
But what’s he doing to articulate his thoughts? He ain’t typing, and I don’t see his tongue or anything else manipulating any system.
I saw Steven Hawking in Boston a few months ago.
Most of the times you see him his computer is just playing back pre-canned speech. He controls when each sentence or paragraph starts. That’s how he gave his lecture.
When he’s composing real time, as in his answers to the questions after the lecture, it takes 3-5 minutes to compose a 10-15 second (spoken slowly) reply. The moderator had to make small talk until Hawking announced “I will reply now” (another canned message).
I’m not sure exactly how he controls the computer, but I think it may be as sophisticated as a camera reading the motions of his eyes.
I’ve seen movies made looking over his shoulder, and he’s paging through very dense lists of words, phrases, and letters.
He controls the computer by a joystick thing in his hands, with like 2 buttons on it. It is the same sytstem he has had for like ten years. It scrolls through all the words and he picks the word he wants by pressing one of the buttons. Thats why it takes so long.
I read somewhere that he now thinks of the computer voice as his normal voice. He said that even if someone offered him a voice synthesizer with a British accent, he probably wouldn’t take it.
[hijack]
Yes, coconuts migrate. They float. That’s why they’re native to every continent but Antarctica.
[/hijack]
I think he took a liking for the married woman assigned to care for him. & is now married to her, right?
I’m surprised that they can’t make a more natural sounding voice. Is there some non-technical reason, or is that really the limit of computer generated speech?
I’m pretty knowledgable in the computer area, but that’s one area I know nothing about.