Anyone thinking of using this for external mics research?

#1

I’d would love to see an open source project for external devices for hearing aids. I’ve had ideas such as building a table which is embedded with mics at each seat that would act act as external mics to hearing aids, and would connect to the hearing aids via bluetooth, with a mobile app to control the parameters.

These parameters include:
Which mics to focus on
Different modes for different environments
The ability to switch between the mics, such that the first person who begins talking is heard

With new hearing aids supporting bluetooth, I think something like this can be utilized and tested.

Currently, I use a mic from a major HA brand. It uses FM, the sound quality isn’t the best, it’s very expensive, but it works and it’s very useful in noisy situations. I’d like to have the flexibility to improve it, not only technically, but have the ability to embed mics like these mechanically (on a tabletop). Right now, I hold the mic when talking at a restaurant and this is inconvenient enough that I don’t always use it.

Unfortunately, I don’t have much experience with audio processing algorithms.
Has anyone else here thought of open source projects such as this?
Is anyone aware of any open source projects that are looking into external devices for hearing aids?
Do you have any pointers on getting started?

#2

The Tympan RevD has a Bluetooth module that supports audio, but I am not sure if there is too much latency in the signal to make it useful in a hearing aid application. That would need to be tested. The module we use is the BC127 module that you can get at SparkFun, for example, if you want to play with it.

#3

Hi @gera I agree with @biomurph that Bluetooth, while useful for integrating your phone with your hearing aid, probably adds too much latency to make it useful for streaming the audio from microphones, compared to the established technology of induction loops. I have not measured the latency of an induction loop, but I’d expect it to be low in an all-analogue system.

To make this avenue worth pursuing, you would need a context in which digital audio streaming added some benefits that induction loops could not provide. For example, if the hearing aid user wanted to rebalance the audio to focus between different microphones, it would be easier to do this via Bluetooth control of the audio mixer which feeds the induction loop, rather than using Bluetooth for the audio stream. My suggestion for getting started would be a multichannel USB soundcard with mic preamps that has ALSA support, so you can use it with a low-latency Linux kernel. Then you could add a control application which worked with your phone to that GNU/Linux device.

#4

Some existing major hearing aid brands provide external bluetooth mics. I’ve used one such bluetooth mic myself, mainly for one-on-one interaction using an external receiver so I’m not sure how latency compares to low energy bluetooth. The latency is slightly noticeable when I speak through the mic and there’s a mismatch in the time when I say a word and hear it, however it’s not a big deal when someone else speaks to it. I think the benefits from improvements in speech comprehension greatly outweigh any cons due to latency.

I have yet to see a bluetooth solution like the FM system I use. You can have multiple mics - up to 10 connected together. The way it works is one mic acts as a central connection point for the other mics, and it’s the central mic which sends the signal to an FM receiver. The system works by triggering the mic in which the first person talks. It works quite well with minimal latency. I’m not sure which wireless protocol is used to connect the other mics to the central mic. What if, instead of using an FM connection between the central mic and the FM receiver, we could use bluetooth? This way, I wouldn’t need an external receiver with newer hearing aids which have bluetooth built-in.

I’ve also tried induction loops, and while they are beneficial, I’d prefer something more convenient such that I don’t have to wear a wire around my neck. Induction loops also present another problem for me - I found that some movements can throw the angle off and position the receiver such that the signal received from the induction loop is attenuated.

Suppose we take this tabletop mic application and extend it - make it portable. I’d have multiple mics and have each person wear a microphone. To connect this system to a portable induction loop, ideally it would receive the audio wirelessly and then transmit the audio to the hearing aids via magnetic induction. Wouldn’t it be better to transmit the audio directly to the hearing aids without a middleman?

As a result, I think a new solution is worth looking into to accomplish something that was previously only possible with an external receiver - it’s very inconvenient to wear those things everywhere I go.

Given all this, I think having some measurements for how much latency there is when using low energy bluetooth would help in deciding whether this is worth pursuing. A little latency isn’t a dealbreaker. In a restaurant setting, I mute my hearing aid microphones when using external mics in order to alleviate the restaurant noise, thus I don’t hear any echo due to mismatch in latency.

Is it ok to name the brands I’m referring to above? The goal is not to endorse any brands, but for the sake of discussion of what I use. Perhaps it would help you understand what I’m looking for in an open source solution.

#5

Hi @gera the improvements in speech comprehension available can be processed at lower latency when the processor is local to the microphone and earpiece. See for example the research device from Bat and Cat https://www.youtube.com/watch?v=xJlAEyHkXD0 which uses a micro HDMI connector for multichannel audio.

The induction loops I was referring to are not worn by the user, they are built into the room and work with the ‘T’ switch on hearing aids. For example https://www.hearinglink.org/living/loops-equipment/hearing-loops/what-is-a-hearing-loop/