In a restaurant or party, background noise can make it difficult to hear people talking, even up close. But soon we could be wearing headphones that use artificial intelligence to filter out noise that’s more than a few feet away, creating a “sound bubble” that lets you focus on your own conversation.

Developed by engineers at the University of Washington, the device is essentially a pair of noise-canceling headphones equipped with six additional microphones along the headband. A small on-board computer runs a neural network trained to analyze the distance of different sound sources, eliminating noise coming from a distance and amplifying sounds closer to the user.

The end result is a sort of sound bubble, as the team describes it, that can be customized with a radius of 1 to 2 m (3.3 to 6.6 ft). The idea is that you can clearly hear people talking inside that bubble, while noises outside it are suppressed by an average of 49 decibels. If someone else enters the bubble, they can also join the conversation.

“Our abilities to focus on people in our vicinity can be limited in places like noisy restaurants, so creating sound bubbles on audible material has not been possible until now,” said Shyam Gollakota, lead author of the study. “Our AI system can learn the distance for each sound source in a room and process it in real time within 8 milliseconds on the hearing device itself.”

The neural network was initially trained on data collected in 22 different indoor environments, such as offices and living spaces. The headphones were placed on a mannequin head and rotated while the noises were played from different distances. The algorithm appears to compare the different phases of each sound frequency to determine how far away the source is and blocks the sounds accordingly.

This is just the latest version of the technology, which the team has been developing for some time. An iteration last year used a swarms of small robots which would move on its own in a room, taking measurements to create separate audio streams for different sources, allowing the user to mute certain areas on demand. Just a few months ago, researchers demonstrated a version that could pick out a single person’s voice Watching them.

This sound bubble version could end up being the most practical iteration of the technology, allowing you to have a clear conversation in a bar with people at your table. It could work even better if it can be integrated into smaller equipment , such as hearing aids or headphones, and luckily the team is already working on that, as well as setting up a startup to commercialize the technology.

The research was published in the journal The electronics of nature. The team demonstrates the sound bubble technology in the video below.

Source: University of Washington