Researchers at a US university claim they have found a way to separate the sound of a single voice from a crowd.
Increasing research effort is going into speech recognition and other audio technology as scientists look for ways to search and use the rapidly expanding volume of digital voice data.
The University of Missouri researchers say they have solved a longstanding problem of how to separate one sound - perhaps a voice - from a group of other recorded sounds while still retaining the speakers vocal characteristics.
Researchers have previously found ways to separate and reconstruct voices, but were able only to reconstruct the words, rather than the characteristics of the voice itself.
Mathematics professor Peter Casazza said the Missouri researchers were now looking for ways to make their solution replicable. “The computer we use is doing the work without an algorithmic program. It uses a system called a neural net, which is designed for the computer to teach itself. Basically, it works on trial and error,” he said.
“This isn't consistent and cannot be duplicated easily. We need to find a way to design an implementable algorithm that could do this consistently and quickly.”
Vote for your IT greats
Who have been the most influential people in IT in the past 40 years? The greatest organisations? The best hardware and software technologies? As part of Computer Weekly’s 40th anniversary celebrations, we are asking our readers who and what has really made a difference?
Vote now at: www.computerweekly.com/ITgreats