Roger Ebert, the Pulitzer Prize-winning film critic and subject of Chris Jones’ prolific profile in Esquire, is on Oprah today.
In a preview video, Ebert’s wife hears Ebert’s voice for the first time since July 1, 2006, thanks to a computer program that splices together voice clips from his old movie commentaries.
“In first grade, they said I talked too much,” he typed. “And now I still can.”
His wife Chaz cries when she hears the new technology for the first time. “I actually think it’s incredible,” she said. She asks him what he feels. He typed/said: “Uncanny, a good feeling.”
NPR interviewed Dr. Matthew Aylett, chief technical Officer of CereProc, who explains how his Scotland-based company built Ebert’s synthetic voice.
Dr. AYLETT: So Roger’s lost his voice, but he has, of course, an awful lot of audio that he’s recorded in the past. So we’ve been able to mine this audio data and to produce a prototype of his voice for him.
ROBERT SIEGEL: He enters text, and the software finds those phonemes of Roger Ebert from things he actually said in recordings that you’ve assembled, and out comes a plausible Roger Ebert.
Business Insider Emails & Alerts
Site highlights each day to your inbox.