I am locked in a quiet, car­peted room, lis­tening to a robotic voice on my iPhone. When the voice pauses, I repeat after it: “Jo’s gen­tle­manly demeanour amused and set him at his ease.”

This rather odd sen­tence, along with a few hun­dred others, form part of a test for a new phone app that will allow people to donate their voices to help other people. VocaliD wants to use speech record­ings to create per­son­alised syn­thetic voices for those who are unable to speak on their own. One day, my voice may be cut up and cus­tomised for a person who needs a voice that sounds a bit like mine.

There are mil­lions of people with severe speech imped­i­ments because of a stroke, Parkinson’s or cere­bral palsy, for example. In the past, a lucky few have had syn­thetic voices built for them. When movie critic Roger Ebert lost his ability to speak due to cancer, Scot­tish text-​​to-​​speech com­pany Cere­Proc was able to build a sub­sti­tute that sounded close to his own.

But most people don’t have a rich supply of audio record­ings at their dis­posal to help stitch a new voice together. Gen­er­ally, they are stuck with generic com­put­erised voices – think Stephen Hawking, who uses an early syn­the­siser called DECtalk.

Read the article at New Scientist →