Smartphone now lets you lend your voice

A new smartphone app may allow people to donate their voice to be customised into synthetic speech for those who are unable to speak, scientists, including one of Indian-origin, say.

The app is being developed under a programme called VocaliD which aims to give a voice to those with severe speech impediments because of a stroke, Parkinson's or cerebral palsy, for example.

Generally, such patients are stuck with generic computerised voices such as physicist Stephen Hawking.

"For these individuals, this is the only way that they interact with people around them," said Rupal Patel, a speech scientist at Northeastern University in Boston and VocaliD's co-director.

The researchers listen to the limited sounds that their patients are able to produce. These utterances shed light on what that person's speech might sound like - whether it's high-pitched, raspy or breathy.

A surrogate who is similar in age and the same sex is selected to donate their voice. That person reads through several thousand sample sentences, sourced from books like White Fang, The Wonderful Wizard of Oz and The Velveteen Rabbit.

Then, using a software tool called ModelTalker, the surrogate's voice is blended with the patient's and stripped down into the tiny units that make up speech.

"You probably wouldn't recognise it as having come from the donor any more," said Timothy Bunnell of the University of Delaware in Wilmington, who created ModelTalker and is also VocaliD's co-director.

Using this method, the group has built a handful of personalised voices, 'New Scientist' reported.

However, the process is slow, since surrogates must come to a studio to record for several hours. It takes at least 800 sentences to create a usable voice, and around 3,000 for one that sounds relatively natural.

VocaliD wants people all over the world to donate their voices so their "voice bank" will have a whole range of speaking styles on tap.

"If we were successful at being able to do this data collection via the iPhone, we'd really get to capture the variation of voices in the world," Patel said.