Over 54 years after the assassination of US president John F. Kennedy, the speech he was to deliver in Dallas on 22 November 1963 can be heard thanks to the work of a team of engineers at Scotland-based audio technology company.
CereProc teamed up with The Times newspaper (as part of its Unsilenced project) and Irish creative agency Rothco to build a database that has been used to deliver the speech in the late president’s voice.
Over 830 speeches and 116,777 phonetic sound units were researched for the project in the run up to the centenary of Kennedy’s birth.
Kennedy was shot as his motorcade travelled through Dallas en route to make a speech at the Trade Mart.
The text of that speech was kept and given to a local businessman by Kennedy’s vice-president Lyndon B. Johnson.
The Times released the new version of the speech on its website 16 March.
Cereproc built the monologue entirely out of data, using AI. It is a world first for the speech to be recreated in the president’s voice and was pieced together by analysing analogue recordings. The audio was then finessed at post production company, Screen Scene.
“As part of Accenture Interactive, we really believe that now is the time for creative ideas to become world class brand experiences. We all crave great experiences, so brands like The Times that embrace this consumer want, will reap the rewards,” said Patrick Hickey, CEO of Rothco.
‘JFK Unsilenced’ launches 16 March 2018, as a 22-minute video on the broadsheet’s website. To find out more about the project, visit The Times.
Edinburgh-based CereProc claims to have developed the world’s most advanced text to speech technology, where voices not only sound real, they have character, making them suitable for any application that requires speech output.
CereProc chief voice engineer, Chris Pidcock, said it was the first time that the company’s technology had been used in this way.
