Hearing aids amplify sounds in our surroundings. New research makes it possible to distinguish between wanted sounds, e.g. a voice, and unwanted noise.
People with hearing impairments often have difficulties detecting low-level sounds so amplification is an important function of the hearing aid. Today, all sounds and noises are amplified, so when hearing aid users listen to a person talking, both the speech and the reverberations from the room are amplified. This may work in a small living room, but in restaurants, churches, railway stations, or other large rooms, it can be challenging for hearing aid users to pick out the desired sound—the voice—when the reverberations of the voice are also amplified. This also causes trouble determining where the sound is coming from as the volume of voice and reverberations is equally high.
"Our method means that not only will it be easier for the user to understand what another person is saying. It will also be possible to assess where in the room the sound is coming from."
Tobias May, associate professor at DTU Electrical Engineering
A research group from DTU has now developed a new algorithm that makes it possible to amplify only the sound of a voice, and not the reverberations created by it. This requires fast-acting sound compression, making it possible to recreate the same sound images as those perceived by normal-hearing listeners. And the researchers have succeeded in establishing how this can be done.
“Our method means that not only will it be easier for the user to understand what another person is saying. It will also be possible to assess where in the room the sound is coming from,” says Assistant Professor Tobias May from Hearing Systems at DTU Electrical Engineering, one of the researchers behind the development of the new method.
Strong interest from hearing aid manufacturers
The new research has been presented both in The Journal of the Acoustical Society of America and at one of the world’s leading conferences on hearing aid technology.
“Several hearing aid manufacturers expressed a keen interest in our results, which we will now continue working on and try to refine even further. The next step will be testing the method in situations involving voices from several people talking at the same time. We hope that it will be possible to preserve the listener’s spatial impression, even if several different voices are amplified at the same time,” explains Tobias May.
If the next experiments produce just as good results, prospects are promising for future hearing aid users. The method developed by the DTU researchers is relatively easy to implement in existing hearing aids.
Photos: Audio Visual Immersion Lab (AVIL) at DTU Electrical Engineering. Here, it is possible to simulate acoustic signals in different environments and test the correlations with spatial hearing.