A decade ago, authors in the hearing literature were celebrating advancements that had produced the first digital signal processing applications in hearing aids, and predicting that cognitive function, auditory scene analysis, and similar individualized factors would be at the center of the next wave of progress (Trends Amplif. 2007;11:31). Those predictions turned out to be on target. In recent months, studies have announced promising findings about the potential application of cognitive control, visual guidance, and binaural beamformer technology in hearing aids.
These findings are generating widespread enthusiasm about the possibility of a new generation of assistive hearing devices. On the front lines of hearing health care, what can we expect to see in hearing aid technology in the next five to 10 years? Researchers and clinicians suggest that several new choices will make dramatic strides toward availability.
COGNITIVE LISTENING TECHNOLOGY
Building on research showing that it was possible to decode the attended target of a listener using neural responses in the listener's brain, researchers at Columbia University recently developed and tested an end-to-end system that receives a single audio channel containing a mixture of speakers heard by a listener, along with the listener's neural signals (J Neural Eng. 2017;14:056001). The system then automatically separates the individual speakers in the mixture and determines which speaker the listener wants to hear. It then amplifies the attended speaker's voice to assist the listener in that hearing. This entire process occurs in less than 10 seconds.
“This work combines the state-of-the-art from two disciplines: speech processing technology and auditory attention decoding,” said Nima Mesgarani, PhD, an associate professor in Columbia's neurobiology and behavior program within the College of Electrical Engineering. “Several other groups have been working on this idea, but our study is a breakthrough in removing a major obstacle toward real-world implementation of this idea.”
Mesgarani told The Hearing Journal that from a theoretical perspective, nothing should prohibit the application of this technology in an actual hearing aid. In fact, “several hearing aid companies have already started researching this idea and expressed interest in our approach.” However, he acknowledged that at least three technical barriers remain to be overcome.
“We need a robust and unintrusive way to measure brain signals, we need powerful algorithms to process sounds in the environment and synthesize what is ideal for the listener based on the task they are engaged in, and we need to have enough computational power to implement all of this in a small device,” Mesgarani explained. He emphasized that all three of these needs are active areas of research and significant progress is being made in each. He predicts that a cognitively controlled hearing aid could become available in five years.
BINAURAL BEAMFORMER TECHNOLOGY
Another application of brain interpretation of sound is being studied in Australia, at the HEARing Cooperative Research Center in Melbourne. Jorge Mejia, PhD, is a project leader and key researcher at the center. His focus is on the psycho-acoustics of hearing and signal processing coding for binaural hearing devices, and the development of bilateral beamformer algorithms for hearing devices.
“Unfortunately, there are many listening situations that remain challenging to people with hearing impairments, such as talking to a friend in a busy restaurant or asking for a beer at the local pub,” Mejia said. “I aim to address this issue by making hearing aids smarter and better. With my colleague, Richard Van Housel, PhD, we have developed technology based on binaural beamformers, sometimes called super-directional microphone features. However, the binaural beamformer we developed is different in that it stimulates how the brain handles sound to restore and enhance it.”
Mejia and Van Housel's initial findings were described in an article in The Hearing Journal earlier this year (Hearing Journal. 2017;70:34). Their pilot study showed that a commercially available biosensor system worn on the head could measure changes in brain activity that were consistent with perceived listening effort when people with normal hearing listened to a brief story in differing single number rating systems. In that article, Mejia wrote that with improvements in the measurement analysis method, it may be feasible to obtain brain activity measures from hearing aid users and use this information to control the noise management features of hearing aids.
Since then, Mejia and colleagues have performed further testing on this model. Finding the balance between the listener and technology continues to be challenging. “It was not until recently that we could relate mental capacity to the ability of the listener to adapt and fully take advantage of newer technologies,” he explained. “It is not just about the technology; it is also about how you deliver the technology to the listener. Therefore, we need to reconcile the development of newer technology with actual application or integration of the technology … and to achieve this we may also need to update the service delivery models.”
Mejia said he expects binaural beamformer technology to be a common feature in hearing aids of the future. “In addition, hearing aids will be smarter—perhaps learning about the individual's preferences as well as predicting those preferences of usage in autonomous ways,” he said.
VISUALLY GUIDED DEVICE
A third frontier of hearing aid research involves a visually guided prototype comprised of a beamforming microphone array and an eye tracker. The microphone array consists of 16 microphones arranged in four rows oriented front to back and mounted on a flexible band worn on the head. The eye tracker senses eye gaze and steers the direction of the beam.
“The studies we have conducted to date have focused on the performance of these two functional components working separately or in tandem,” said Gerald Kidd, PhD, a professor of speech, language and hearing sciences at Boston University's Sargent College. The team's research about the visually guided hearing aid, or VGHA, has attempted to determine “how well the beamformer can help the listener ‘solve the cocktail party problem’ by measuring speech recognition under masked conditions, such as when the target speech is masked by competing speech or competing noises.”
The research is the first large-scale undertaking to understand whether visual guidance of a highly tuned beamformer can provide benefit to listeners with hearing loss in multiple source environments (Trends Hear. 2017;21:2331216517722304). However, other groups have become interested in the concept and are pursuing similar approaches, Kidd said. “My feeling is that this is an area that holds considerable promise and may result in practical advancements in hearing aid technology in the future,” he added.
Although research interest and understanding are growing, more time is needed to fine-tune this application. “We are a long way from producing a clinical device that can be offered to patients, but so far the results—the basic scientific premises upon which this research is based—have been very encouraging,” Kidd said.
A BRIGHT OUTLOOK
Clearly, the hearing aid market is heading toward some dramatic improvements over the next decade. The implications for improved hearing are exciting, but they should be balanced with practicality. Todd A. Ricketts, PhD, a professor and vice chair of graduate studies in the department of hearing and speech sciences at Vanderbilt University Medical Center, noted that the need for cost-effective products and services also warrants attention and resources.
“Research is approaching current needs from a variety of different angles, but one that isn't necessarily on the cutting edge is how to make new devices cost-effective,” Ricketts said. “We need a lot more research into how to deliver cost-effective products to patients.”
He also pointed out that no matter how advanced hearing aid technologies become, patients still need professionals to guide their use. “One of the things that people pay for when they buy high-end hearing aid technologies are the technologies. But what they pay for regardless of technology is professional expertise,” he said. Now that Congress has passed the Over-the-Counter (OTC) Hearing Aid Act to create a new nonprescription category for simple hearing aids, hearing professionals and their patients face many unknowns.
“Ten years from now, I think the hearing aid market will look very different,” Ricketts said. “The OTC category will reshape things. I hope that it will be done right. I hope we will be able to provide really efficient services and hearing health care delivery to patients with mild to moderate hearing loss. I hope that we can individualize technology for people so that more complicated problems have high levels of individualization that target each person's needs. I can see hearing health care delivery having a very wide range, from heavily automated to very advanced automation with professional services.”
Mejia predicts even more far-reaching changes. “I am excited about what will happen over the next 10 years,” he said. “For the most part, I believe the negative stigma related to usage of hearing aids will change. Perhaps as we move into a noisier and more demanding society, hearing aids will become indispensable everyday devices, much like mobile phones are right now.”
Copyright © 2017 Wolters Kluwer Health, Inc. All rights reserved.