Mathematicians Solve the 'Cocktail Party Problem'

August 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

Explore further: Physicists develop key component for terahertz wireless

Related Stories

Physicists develop key component for terahertz wireless

September 14, 2015

Terahertz radiation could one day provide the backbone for wireless systems that can deliver data up to one hundred times faster than today's cellular or Wi-Fi networks. But there remain many technical challenges to be solved ...

A humanoid robot to liaise between space station crews

September 7, 2015

A team of French researchers from the Institut cellule souche et cerveau (Inserm/Université Claude Bernard Lyon 1), led by CNRS senior researcher Peter Ford Dominey, has developed "an autobiographical memory" for the robot ...

Apple event expected to focus on iPhones, TV

September 6, 2015

Apple is expected to unveil updated iPhones Wednesday along with an Apple TV revamp that may signal a push into online television streaming dominated by Netflix.

Economist discusses turbulent world of aviation reform

August 20, 2015

John Strong knows aviation. For roughly 30 years, the CSX Professor of Finance and Economics at William & Mary's Raymond A. Mason School of Business has immersed himself in the economics, safety and politics of the airline ...

Recommended for you

Mexican site yields new details of sacrifice of Spaniards

October 9, 2015

Excavations at the site of one of the Spanish conquistadors' worst defeats in Mexico are yielding new evidence about what happened when the two cultures clashed—and a native people, at least temporarily, was in control.

Ancient genome from Africa sequenced for the first time

October 8, 2015

The first ancient human genome from Africa to be sequenced has revealed that a wave of migration back into Africa from Western Eurasia around 3,000 years ago was up to twice as significant as previously thought, and affected ...

Rare braincase provides insight into dinosaur brain

October 8, 2015

Experts have described one of the most complete sauropod dinosaur braincases ever found in Europe. The find could help scientists uncover some of the mysteries of how dinosaur brains operated, including their intellectual ...


Please sign in to add a comment. Registration is free, and takes less than a minute. Read more

Click here to reset your password.
Sign in to get notified via email when new comments are made.