When it comes to computer generated voices we are all pretty use to seeing something like what Stephen Hawking uses that produces that very identifiable sound ore even the more up to date type that Roger Ebert uses that is more experimental . However for children who for whatever reason are unable to vocalize these computer generated voices are typical more adult sounding which can be disconcerting for both them and the listener.
Well the folks at software company Lingit and Media LT, a company that develops devices for assisted living, have been working on a new type of voice synthesizer that is able to make the words being ‘spoken’ by the child sound – well, more youthful.
The Norwegian researchers at the companies first created a ‘master voice’ from combining the recordings of multiple adult speakers who recited several thousand different phrases that was used to create a library of words and sounds. Once that was done they recorded a single child saying a fewer number of phrases then they had the computer compare the master voice to the child’s voice, using the recorded phrases as a reference point, and then altered the master voice to sound like that of a child.
“The result sounds rather like a child with unusual elocution skills, but it’s still much better than the voice of an adult,” said Lingit’s Dr. Torbjørn Nordgård.
It is hard enough for children but add to that not being able to speak and their life is only made more difficult – especially on their peer level as well as socially – so efforts like this are great to see.
via GizMag
