Why it matters: For those tired of incessant keyboard clacking from an office mate or the cries of a child during a flight, a pioneering headphone prototype developed by the University of Washington researchers may be the solution you’ve been seeking. This device goes beyond mere noise suppression, employing AI to selectively silence unwanted sounds while isolating specific voices for a clearer listening experience.
Traditional active noise-canceling headphones mainly generate sound waves to neutralize lower-frequency background noises, such as the hum of an airplane engine, albeit at the cost of also muting sounds you might wish to hear. This prototype, however, is designed to offer users a finer level of control over the sounds around them.
The headphones are equipped with microphones that capture ambient sounds, which are then processed by a neural network trained to identify various sound types, including the barking of dogs, telephone rings, and birdsong. Through a connected app, users can select which sound categories to block or allow, customizing their auditory environment.
A particularly innovative feature is the headphones’ ability to focus on a single voice within a noisy backdrop. With a simple button press, the device will prioritize the voice of a person directly in front of the user, enhancing it above all other background noise.
Shyam Gollakota and his research team introduced this technology on May 16 at a conference organized by the Acoustical Society of America and the Canadian Acoustical Association, where a functional prototype was demonstrated, as reported by New Scientist.
Audio is funneled through AI processing within the device, filtering out unwanted noise nearly instantaneously with just an 8-millisecond delay, virtually eliminating any awkward audio delays. The current model utilizes an OrangePi board for on-device AI processing, avoiding the need for cloud-based processing.
While this prototype is not yet available for purchase, its commercial production would likely involve miniaturizing its components for integration into future wireless headphone designs.
Interestingly, AI is finding its way into consumer audio devices, with advances in noise cancellation for microphones. This technology, however, takes a novel approach by enhancing what users can hear rather than merely filtering out what microphones pick up. Theoretically, any device equipped with a capable AI processor and microphone could offer this selective sound filtering capability.
The thought of such technology brings to mind scenes from the “Black Mirror” series, particularly the “White Christmas” episode, where characters could “block” others from their sensory perception. While enhancing certain voices and muting others might seem like science fiction, it’s an intriguing possibility that this technology could one day make real.
