Roger Ebert, the most well-known film reviewer in the US, lost the ability to speak four years ago after life-saving cancer surgery.
However, computer technology developed by Cereproc, a company spun out of Aberdeen University, has been able to reconstruct Ebert’s original voice by using recordings of television appearances and DVD commentaries.
To reconstruct Ebert’s voice, the company - formed in 2005 - mined the recordings to create a comprehensive database of words and sounds. As a result, when Ebert speaks, listeners who heard him in the past will still recognise his voice today.
Ebert - who is famous for his thumbs-up or thumbs-down film reviews, and is the only film critic ever to win a Pulitzer prize for journalism - can now communicate, using a laptop, by typing sentences that are converted by a synthesiser into the sound of his own voice.
In a televised interview with Oprah Winfrey, Ebert demonstrated the technology by predicting this year’s Oscar-winning films.
Unlike the types of speech systems developed in the 1980s - such as that used by Prof Stephen Hawking - Cereproc’s synthesised voices have character and emotion.
’We specialise in producing voices that have got a bit of character and don’t sound neutral or boring. This synthesis sounds very much like a natural voice,’ said Cereproc’s chief technical officer Dr Matthew Aylett.
英國鐵路公司如何推動凈零排放
It would be better if the trains had good coverage of the country. Large areas have no easy connection and so cars (or buses?) and lorries are still...