Blog

A new voice for Roger Ebert: his own

by Wendy M Grossman | posted on 13 September 2011


I have a piece up at Scientific American on the personalized, synthesized voice being built for the movie critic Roger Ebert by the Edinburgh company CereProc

Wendy M Grossman

The notion that Ebert, who appeared on many decades of TV shows reviewing movies, ought to be able to have a synthesized voice based on his own recordings was one that occurred to many people, as he blogged. I also know this from personal experience: at his 2009 film festival, a guy sitting next to me waxed excited about the idea, saying that a nearby lab was working on the necessary technology.

But this is the way it goes sometimes: the most obvious ideas are not necessarily the simplest ones to make happen. The most intriguing possibility in all this, though, is that one day soon it may become possible for people to bank their voices to create personalized synthesizers. In the BBC Radio 4 program made about this work (rush to listen - it will only be up for four more days as I type this), CereProc's Matthew Aylett talks about how much our identities are bound up in our voices. Or the voices we become accustomed to - he also notes that after several decades Stephen Hawking has become so identified with his synthesizer's stretched, metallic, 1980s sound that he has declined to upgrade to something newer and more natural-sounding. I find it intriguing that one of CereProc's near-term goals is to created an automated service that will allow people to create good-enough-for-most-purposes voices over the Web, making the whole business of synthesizers more personal and much more efficient and cheaper. Obviously, having a personalized synthesizer doesn't really solve the problem of being muted: the unavoidable typing lag means that no synthesizer can give a speechless person the ability to interact in public situations with speakers. But it does mean that when they *can* navigate interactions through speech the voice they use is theirs.

Meantime, enjoy CereProc's Bush-o-Matic: type in anything you like, and listen to a synthesized George W. Bush say it.

While I was working on the article, I also had a chance to read Ebert's autobiography, Life Itself, which is officially published today by Grand Central Publishing. Frequent readers of his journal entries over at his site will recognize some of the material, but the book is surprisingly different from the blog jottings - both more concise and more detailed. It is basically a no-holds-barred journey through his life, beginning with his childhood in Champaign-Urbana, Illinois, and moving through to the present day leaving little to the imagination: his weight, his alcoholism, his various relationships all get covered warts and all. As he writes several times in the book, he's only going to write his life once, and it isn't worth doing at all if he doesn't do it honestly.

wg


Technorati tags:       
Personalized speech synthesizers may be coming sooner than you think

Wendy M. Grossman’s Web site has an extensive archive of her books, articles, and music, and an archive of all the earlier columns in this series. Readers are welcome to post here, at net.wars home, follow on Twitter or send email to netwars(at) skeptic.demon.co.uk (but please turn off HTML).