"Stephen's voice is very distinctive, but you say there might be a problem retaining it?
'I guess the most interesting thing in my office is a little grey box, which contains the only copy we have of Stephen's hardware voice synthesizer. The card inside dates back to the 1980s and this particular one contains Stephen's voice. There's a processor on it which has a unique program that turns text into speech that sounds like Stephen's, and we have only two of these cards. The company that made them went bankrupt and nobody knows how it works any more. I am trying to reverse engineer it, which is quite tricky.'
Can't you update it with a new synthesizer?
'No. It has to sound exactly the same. The voice is one of the unique things that defines Stephen in my opinion. He could easily change to a voice that was clearer, perhaps more soothing to listen to – less robotic sounding – but it wouldn't be Stephen's voice any more.'"
"The DECtalk DTC01 voice synthesiser he uses, which has an American English accent, is no longer being produced. Asked why he has still kept it after so many years, Hawking mentioned that he has not heard a voice he likes better and that he identifies with it. Hawking is said to be looking for a replacement since, aside from being obsolete, the synthesiser is both large and fragile by current standards. As of mid 2009, he was said to be using NeoSpeech's VoiceText speech synthesiser.[37]
In Hawking's many media appearances, he appears to speak fluently through his synthesiser, but in reality, it is a tedious drawn-out process. Hawking's setup uses a predictive text entry system, which requires only the first few characters in order to auto-complete the word, but as he is only able to use his cheek for data entry, constructing complete sentences takes time. His speeches are prepared in advance, but having a live conversation with him provides insight as to the complexity and work involved. During a TED Conference talk, it took him seven minutes to answer a question."
Also see Speech Synthesis on Wikipedia.
Note: I created a new Speech Synthesis label for this post. I realized I didn't have one and couldn't get myself to use the Speak n' Spell label. :)
