Skip to content
HearingTracker
Log in

Android LE Audio users: does the pocket your phone is in decide which aid drops? No dropout counts too

50 replies981 viewsLast reply

Jump
Link to this topic
Reply

Someone on Reddit told me to bring this here, said there are a lot of Android users over here. Fair warning first: I work factory side on aids, so I’ve got a stake in the answer.

What I’d claimed was this. Phone in one front pocket, and the aid on that same side drops first. Swap pockets and you swap the ear. My reasoning was 2.4GHz and a body in the way, since pocket to the opposite ear is the long path.

Then he ran it. Pixel 9, Oticon Intent, start to finish, no extra hardware. Left pocket, right pocket, shirt, shorts. Four combinations, not one dropout. I conceded over there and I’m not walking that back.

Someone else ran the same swap and got exactly what I’d described. Move the phone, the bad ear moves with it. I don’t have their phone or aid model though, which is part of why that one is harder to weigh.

So it’s one result each way, both n=1, and I can’t tell which is the odd one out.

Fair question why I don’t just test it myself. A bench can sweep parameters all week but there’s no torso on it and it doesn’t go for a walk, and I think the walking is most of what’s going on here.

If you’ve got LE Audio aids and a Pixel or Samsung that does LEA:

Put something on through the aids first. Music, a podcast, a call, whatever you’d normally stream. If nothing’s playing there’s nothing to drop. Phone in one front pocket. Walk five minutes. Note which ear cuts out first, if either does. Then move the phone to the other pocket and walk it again.

Two walks. That’s it.

When you post, say which aids, which phone, and whether it’s actually LE Audio rather than classic Bluetooth. Tell me what you observed, not whether it confirms anything.

Both directions count the same. Each one has exactly one data point right now, so a second clean run and a second dropout are worth exactly as much to me. Whatever comes back I’ll post in this thread, including if it goes against me again. It isn’t going anywhere else.

Thanks either way.

15 people in this conversation36 min readJump to latest

Pixel 10; Philips 9050 hearing aids.

When the phone is toggled to LEA on, I get no drop outs at all. The connection is rock solid. I can walk into the next room and I get no drop outs with the phone in the other room.

Let’s make this clear, there is a toggle switch for LEA in the phone menu for my Pixel 10.

Pixel 10. Starkey Edge .

Phone front pocket three feet from my aids.

Long walks, no drops. Inside, outside.

Pixel 9, Sound gear Amp (Starkey edge) no problems when walking with phone in left or right pockets. Using . As other have stated I can leave my phone in another room in the house and walk around without losing the call signal. 30 ft plus.

I’ll play, with a possible bonus data point. I use Oticon Intent 1s with my Samsung Galaxy S26 (currently running beta OS version). I had no cut-outs with left or right front pockets. I then moved my phone to the side pocket of my shorts designed for that use, thinking that will definitely put my entire torso between the phone and my left ear. I did not disconnect while walking around listening to music. Then sound diminished, and I checked my phone to discover a missed incoming scam call. When the phone app released audio, music resumed in my left ear followed several seconds later by my right ear. Volume was lower than I had set it, so I had to turn it back up. In other words, perhaps it’s a multi-factor issue with connections and apps fighting for signal. I also just confirmed I am still in mode (I’ve had several instances of my connection switching to after changing apps and devices).

Cochlear Nucleus 8 sound processor on left ear. Bluetooth protocol. Still waiting for firmware upgrade to activate Bluetooth .

When keeping Samsung Galaxy S23 Ultra in the jeans pocket on the same side as sound processor - frequent disconnection. It is more often when I change pocket.

On right ear: Phonak Audeo Infinio Sphere 90. Bluetooth Classic (probably 5.2). No problem with connection, rock solid regardless of pocket.

Six reports in under a day. Thank you, genuinely.

Reading them back so nobody has to scroll.

user990, Pixel 10 with Philips 9050, LEA toggled on. Rock solid. Phone in another room and it still doesn’t care.

freezerman404, Pixel 10 with Starkey Edge . Long walks, inside and out. Nothing.

Jamesdjs, Pixel 9 with Sound Gear Amp. Either pocket, thirty feet, nothing.

requa, Galaxy S26 with Oticon Intent 1. Both front pockets, then the side pocket with a whole torso in the way. No dropout. The one interruption was a scam call grabbing the audio, which isn’t the same animal at all.

That’s four for four. Phone in a pocket, on , nobody drops. My claim is dead in LEA and I’m not going to try to rescue it.

Then Bimodal_user posted, and that one is the whole thread.

One person. One phone. One walk. Two ears running two different protocols. Left is a Nucleus 8 on , and the phone in the same side jeans pocket disconnects often. Right is a Phonak Infinio Sphere 90 on Classic, and it doesn’t care where the phone sits.

I couldn’t have designed that control. Same body, same gait, same room, same second. The only thing that changed is the protocol, and only one side breaks.

Honest caveat on it. A Nucleus 8 is a sound processor, not an aid, and I don’t know how its antenna sits next to a . So it isn’t protocol and nothing else. It’s still the closest thing to a control anybody here has produced, and it came from one post.

So the pocket effect looks real. It just isn’t mine, because I never said which protocol I meant. Four LEA reports say no. Classic on the right ear says no. ASHA says yes.

That points at link margin rather than geometry. LE Audio’s isochronous streams carry a retransmission budget by design. I’m not aware of anything equivalent in ASHA. A torso costs you something like ten to twenty at 2.4GHz, and that’s survivable if you’ve got margin to spend, and it isn’t if you don’t. That’s my read. I’d like someone to shoot it down.

One more thing from requa’s post that I nearly scrolled past, and it might be the most useful sentence in here. He says his connection has silently fallen back to ASHA after changing apps and devices. He had to go and check to confirm he was still on LEA.

If that happens to other people, then the guy on Reddit who reproduced my original claim may not have been on LE Audio at the moment it dropped. He’d have had no way to know. Two results pointing opposite directions, and it’s possible neither person measured wrong. They just weren’t measuring the same thing.

Which changes what I should be asking for. Checking the protocol before a walk doesn’t prove much. What I want is somebody who can check it right after a dropout, while it’s still fresh. If the connection turns out to be sitting on ASHA at that moment, that’s the answer.

And I’ll say the obvious part. I work factory side, so I’d love this to land on “LE Audio fixes it.” Right now the only ear that saw the effect was on ASHA, and Bimodal_user handed me that for free while I was busy being wrong.

@luogangan said:

One person. One phone. One walk. Two ears running two different protocols. Left is a Nucleus 8 on ASHA, and the phone in the same side jeans pocket disconnects often. Right is a Phonak Infinio Sphere 90 on Classic, and it doesn’t care where the phone sits.

I couldn’t have designed that control. Same body, same gait, same room, same second. The only thing that changed is the protocol, and only one side breaks.

I forgot to mention that I have never use simultaneous streaming to sound processor and Phonak Audeo Infinio Sphere. It’s usually Bluetooth for Nucleus 8 + Bluetooth low energy (not Audio) for myPhonak app.

I’m glad I was able to contribute something to the discussion.

That’s the most useful thing anyone’s said in this thread, and it costs me the best part of my argument.

I had your two ears down as a controlled comparison. One body, one walk, two protocols, only one side breaking. What you’re telling me is that the right side was never carrying a stream. It was holding a low energy link for the app while the Nucleus 8 did the actual listening on .

So there was no comparison. An idle app connection and a live audio stream fail in completely different ways, and a link with nothing on it has nothing to drop. I read what I wanted to read.

What’s left standing is the half you actually measured. Your ASHA side drops, and it drops on the pocket side. That’s real and I’m keeping it. What I don’t have any more is a Classic data point, from you or from anyone. That row is empty again.

I’ve already put this on Reddit with your two ears as the centrepiece, so I’ll go and correct it there as well. Better it comes from me a few hours in than from somebody spotting it next week.

Don’t apologise for the timing either. You flagged it before anyone built anything on top of it, which is the only version of this that’s cheap.

In reply to @luogangan

@luogangan said:

That’s four for four. Phone in a pocket, on LE Audio, nobody drops. My claim is dead in LEA and I’m not going to try to rescue it.

I think that theoretically there may be connection problems even with Bluetooth LEA, if the person is holding the phone in their pocket and is in an open space. Then the radio waves have nothing to bounce off to reach the /sound processor.

I don’t use Bluetooth LEA yet.

Samsung S25 Plus and Jabra Enhance Max. No drops while during a phone call. I carry my phone in my back pocket when walking around while on a call due to too many items in both front pockets. I have also left my phone on a table and walk all over the house without any drops. I have gotten around 60 feet (18.2m), still no drops.

Two things landed at once and they’re connected, so I’ll take them together.

Bimodal_user, the open space point is better than you’re giving it credit for. Indoors, when your body blocks the straight line to the far aid, the signal still gets there off a wall or a ceiling or the floor. Out in a field there’s nothing to bounce off, so the only path left is the one going through you. That’s the worst case, and it isn’t the same experiment as walking around a house.

Which matters, because look at what I’ve actually collected.

Turtley, Samsung S25 Plus with Jabra Enhance Max. Back pocket on a call, no drops. Phone left on a table, sixty feet, all over the house, no drops. That’s the strongest range result in this thread and I’m counting it.

But it’s indoors. So is “phone in another room”, which two other people here reported. Meanwhile the dropouts that started all this were people out on walks. I’ve been sorting my table by protocol and I never once separated indoors from outdoors. Those aren’t the same radio problem.

Turtley, one thing I need from you, and I need it from everyone including the people who already agree with me. Can you confirm the phone was actually on and not Classic? The Enhance Max does both. I spent this week arguing that people who think they’re on LEA sometimes aren’t, so I should point that at my own data before I get comfortable with it.

Bimodal_user, you’ve cost me two things in two days. First the control, now the venue. I’d take both again.

Sure, I can confirm the phone is on .

Screenshot_20260815_080208_Settings

I also did a rudimentary range test during the call with the phone in the house and walked outside and got about 100 feet (328m) before the left aid drop out the I walked a little further and both aids dropped. Pretty decent range from such small devices if I say so myself! Not meaning to sound rude but, I can’t prove that my aids were receiving an actual LE Audio signal from my phone. You’ll need to have specialized test equipment for that.

That’s the screenshot I asked for. Counted.

I’m also counting the thing you did that I didn’t ask for, which turns out to matter a good deal more than the confirmation does, because you walked outside and at around a hundred feet the left aid went first and then both of them went.

That’s the first dropout in this thread. Five setups said never. This one says otherwise.

Which is where I eat something.

Yesterday I was arguing that indoors and outdoors aren’t the same experiment, and I was using that to explain why the good news might be too good to hold, but you’ve handed me the other half of it: your sixty feet indoors held and your hundred feet outdoors didn’t, on the same aids, the same phone, the same afternoon. So the venue counts.

Then the bit I like least. Your phone was sitting in a house rather than in a pocket. No body in the path. None. And the left one still went first, so whatever decided it should be the left ear rather than the right, it wasn’t my torso story. I’ve spent two threads explaining which ear drops by where your body is. That explanation just failed on a case with no body in it.

One number, because I’d rather ask than guess. You wrote about a hundred feet and then 328m. A hundred feet is thirty metres and 328 feet is a hundred metres, so which one did you actually walk? Your sixty foot conversion earlier was exactly right, so I think the metres is a slip. But I can’t round a range.

Your last line is right and it isn’t rude. Nobody here can prove what their aids are actually receiving without test equipment, and that includes the two minute test I’ve been asking people to run all week. The settings screen tells you what the phone thinks it negotiated. That’s all any of us have got.

Sorry, but the 328m is a typo on my part. 100 feet is 30.48 meters according to Google. Leaving the phone in the house while outside for the range test does induce a left aid drop out when I turned my head to the right blocks or interferes with the internal hearing aids antenna. Now, having the phone on my person all the time I’ve never had the signal drop.

That settles the number, and then it undoes the thing I built on top of it.

Thirty metres, not a hundred. Fixed in my table.

Now the part where I was wrong inside a day of writing it. I said your phone was in the house and there was no body anywhere in the path, and I made a meal of that. You’ve just told me you turned your head to the right. That is a body in the path. It’s a skull rather than a torso, and it sits between the phone and your left aid in exactly the way a leg sits between a pocket and the far ear.

So the thing I announced had failed hasn’t failed. It got a cleaner demonstration than I had before, and this time it came with a direction: the far side from where you turned. Same side everyone else has reported.

Your last sentence is the one I’d underline though. Phone on you, never a drop.

Put that next to the range test and the thread finally lines up. On your kit at pocket distance, has margin to spare and your head doesn’t matter. At thirty metres outdoors it’s already at the edge, and then your head is the last few it didn’t have. Same body, same physics, different amount of headroom to spend.

Which is what I should have been saying a fortnight ago instead of the pocket decides. The pocket decides when there’s nothing left in the tank. On there never was. On LE Audio you have to walk to the end of the garden first.

Second time in two days that someone hands me data and then corrects my reading of it before I can. I’d rather it kept happening.

Why only a battery metre, couldn’t we have a gain/ range metre in our phone or App as well, some handy statistics to see each Bluetooth etc

In reply to @tenkan

From ..

Cell phones do not show separate battery usage stats for Wi-Fi or Bluetooth because idle power use is too small to measure, and active data use is already counted inside the apps using them.

Well, mine’s not idle! Even a fractional amount would be interesting to see.

In reply to @Bimodal_user

That’s the first Classic number anyone has given me, and I’d emptied that row yesterday after you told me your Phonak side had never been carrying audio.

Twenty five metres through a load bearing wall, no line of sight. That’s further than the failure I logged this morning, which went at thirty metres in the open with nothing between the phone and the man at all.

So before I write it down I have to ask you the question I should have asked the first time, and I’m asking it precisely because of what happened the first time.

Was the Phonak actually streaming audio during those twenty five metres? Or was it the myPhonak app link again, sitting there with nothing on it?

I’m not being difficult. A link with no audio on it has nothing to drop, and I built a whole argument on not knowing that difference once already this week. If it was streaming, this is the strongest Classic result in either thread and I’ll say so. If it was the app link, it goes in the notes rather than the table and neither of us has lost anything.

Approximate is fine, by the way. I’d rather have your rough twenty five metres with the wall described than a tidy number with the conditions missing.

Related discussions