Are better vocal and instrument isolation techniques on the way? Math professors from the University of Missouri-Columbia claim to have solved the “cocktail party” problem, an auditory source separation problem that the brain has little trouble handling but is very difficult to perform with digital signal processing technology.
In the past, researchers were able to separate voices but not reproduce the characteristics of the voice itself, according to a statement from the university. Casazza said existing programs that can separate and reconstruct voices are not completely reliable because they have difficulty separating voices with similar pitch characteristics. The researchers claim that a program using their solution would be more exact.
While there are still issues to be resolved, it’s easy to imagine how this breakthrough may lead to new acapella extraction and unmixing techniques for remix and mash-up producers. For more information on the cocktail party problem and the proposed solution, check out the full article on the Dark Reading web site.
(Thanks to John LaCour for the tip!)
One Comment
Sounds very interesting doesn’t it? Seems that the Remix population may be able to extract the real voice from any track in future… real n1ce!
Post a Comment