AI headphones let you listen to one friend in a crowd

AI headphones let you listen to one friend in a crowd

/ 10:04 AM June 03, 2024

You can consider Apple AirPods AI headphones because they use artificial intelligence for noise cancellation. However, what if we take that audiophile tech to the next level?

That is what the folks at the University of Washington (UW) achieved with its latest innovation. Its Target Speech Hearing blocks other sounds to help you hear a specific person.

READ: AirPods Pro might have “hearing aid mode”

Article continues after this advertisement

The technology lets you hear that person even if you’re not facing them or you’re in a noisy crowd. As a result, it could drastically improve future earbuds and hearing aids. 

FEATURED STORIES

How do these AI headphones work?

College news publication UW News reported on the Target Speech Hearing technology. Shyam Gollakota, a UW professor in the Paul G. Allen School of Computer Science & Engineering, is one of its main researchers. 

“We develop AI to modify the auditory perception of anyone wearing headphones,” he said. It works by “enrolling” a specific person into the system so that it can recognize it in a crowd.

Article continues after this advertisement

The program works with any off-the-shelf headphones with microphones, turning them into AI headphones. 

Article continues after this advertisement

The user starts by tapping a button while directing their head at someone talking. Then, the microphones will receive those sound waves signal an on-board embedded computer.

Article continues after this advertisement

The UW team’s machine learning software will learn the specific speaker’s vocal patterns. Eventually, the system’s ability to focus on the enrolled voice improves as the speaker keeps talking. 

Gollakota and his team tested the system on 21 subjects. The respondents rated the clarity of the enrolled speaker’s voice almost twice as high as unfiltered audio. 

Article continues after this advertisement

UW News admits that the TSH system has flaws. For example, it can only enroll one speaker at a time.

It cannot register one when no loud voices are coming from the same direction as the target speaker’s voice. Still, users can repeatedly enroll a speaker until voice clarity improves. 

Your subscription could not be saved. Please try again.
Your subscription has been successful.

Subscribe to our daily newsletter

By providing an email address. I agree to the Terms of Use and acknowledge that I have read the Privacy Policy.

The UW team plans to expand the system to earbuds and hearing aids. Soon, it might power your next pair of AI headphones!

TOPICS: AI, technology
TAGS: AI, technology

Your subscription could not be saved. Please try again.
Your subscription has been successful.

Subscribe to our newsletter!

By providing an email address. I agree to the Terms of Use and acknowledge that I have read the Privacy Policy.

© Copyright 1997-2024 INQUIRER.net | All Rights Reserved

This is an information message

We use cookies to enhance your experience. By continuing, you agree to our use of cookies. Learn more here.