Mathematicians Solve the 'Cocktail Party Problem'

Aug 22, 2006

Officials at the CIA and scientists around the world have pondered the "cocktail party problem" for decades. How could they separate one sound - perhaps a voice - from a group of other recorded sounds, perhaps a multitude of voices at a cocktail party? Now, two researchers at the University of Missouri-Columbia have found a mathematical solution to this problem.

"Theoretically, our solution says you should be able to pick up voices on a squeaky old microphone and then separate them all out so that you can hear what each person is saying in his or her own voice," said Peter Casazza, professor of mathematics in MU's College of Arts and Science. "This is a very old problem, and we have the first mathematical solution to it."

Casazza and Dan Edidin, also a professor of mathematics at MU, worked with Radu Balan of Siemens Corporate Research to solve the problem. Their solution shows that it is possible to separate voices and still retain vocal characteristics. Researchers had previously found a solution for separating and reconstructing voices, but they were only able to reconstruct the words spoken, not the characteristics of the voice itself.

"Our solution is called 'signal reconstruction without noisy phase,'" Edidin said. "In speech recognition technology, a 'signal' could be a recording of 25 people in a room talking at the same time. Our solution shows that we can pull out each voice individually, not just with the words, but with the voice characteristics of each individual. We showed that this 'cocktail party problem' is mathematically solvable."

Although Casazza, Edidin and Balan do not have a computer program that can do this automatically, they hope to find a way to develop one. Currently, their solution runs on a computer, but the process cannot be easily replicated or distributed.

"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," Casazza 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."

Casazza said that there are already programs that can separate and reconstruct voices, but they are not completely reliable. For example, such programs have difficulty separating voices with similar pitch characteristics. A program using the researchers' solution would be more exact.

Source: University of Missouri

Related Stories

Lawmakers visit Yucca Mountain, consider nuke waste dumping (Update)

Apr 09, 2015

Several members of Congress are heading to the mothballed site of a proposed radioactive waste dump in the Nevada desert amid new talk about a decades-old problem—where to dispose of spent nuclear fuel stored at commercial ...

Successful demonstration of DARPA's Persistent Close Air Support (PCAS) system

Apr 07, 2015

Close air support (CAS)—delivery of airborne munitions to support ground forces—is difficult and dangerous because it requires intricate coordination between combat aircrews and dismounted ground forces ...

Can phone companies do more to block robocalls?

Apr 07, 2015

Jeri Vargas put her elderly mother on the "Do Not Call" list years ago. So why is the 88-year-old woman with Alzheimer's disease still getting several recorded phone calls a day pitching her everything from ...

Co-founder feuds at LA tech startups show how handshake deals can blow up

Apr 02, 2015

Two Stanford University undergraduates shook hands in their dormitory in early 2011, agreeing to partner on what they hoped would be the next big technology startup.

Researchers aim to safeguard privacy on social networks

Mar 31, 2015

At the end of 2014, Facebook reported 1.39 billion monthly active users. In the meantime, 500 million tweets were sent each day on Twitter. Indeed, social networks have come to dominate aspects of our lives. ...

Traffic court goes digital: Startup fosters settlements

Mar 31, 2015

Traffic court is going digital. Michigan startup Court Innovations has developed a software solution that allows drivers to settle traffic violations by negotiating in a virtual environment instead of showing up to court ...

Recommended for you

Devices or divisive: Mobile technology in the classroom

17 hours ago

Little is known about how new mobile technologies affect students' development of non-cognitive skills such as empathy, self-control, problem solving, and teamwork. Two Boston College researchers say it's ...

Same-sex couples arranging civil partnerships encounter hostility from families and the public, research says

23 hours ago

Many same-sex couples arranging civil partnerships and marriage ceremonies encounter hostility and disrespect from families, colleagues and the public, research shows.

Americans with illegal Iraq War souvenirs go unprosecuted

23 hours ago

As the elected Iraqi government seeks diplomatic respect and struggles to save its ancient sites from the rampages of the Islamic State group, American military members, contractors and others caught with ...

Forming school networks to educate 'the new mainstream'

23 hours ago

As immigration increases the number of non-English speaking "culturally and linguistically diverse" students, schools will need to band together in networks focused on the challenges of educating what has been called "the ...

Rare tidal movements expose Kimberley dinosaur tracks

23 hours ago

While audiences in Perth attend Walking with Dinosaurs this weekend palaeontologists working near Broome will be documenting the extinct vertebrates' extensive fossilised footsteps using laser scanning technology.

Why everything you've heard about women and negotiation might be wrong

Apr 17, 2015

University of Florida student Samantha Miller was listening to a lecture on a commonly held trope about negotiation—that women are bad at it—but the conventional wisdom didn't fit with her experience at all.