OP here -- this article isn't a recommendation to ditch whatever you use and get at throat mic! I just think they're interesting and finally scratched my own itch so figured I'd my experience (and leave some hints if anyone wanted to go make a bluetooth version).
The tl;dr:
> tl;dr - Don’t chuck the Sony 1000XMs/AirPods just yet. Throat mics are an improvement over a standalone wired microphone (if that’s what you are using), and have better background noise reduction but in a quiet room your usual bluetooth headset is still way more convenient.
We had them in the military and we preferred the boom mics. Throat mics, the exact ones pictured, did not pick up low-talking or whispering, which is a large amount of pre-contact traffic.
The new thing the DOD has spent an UNGODLY amount of time and money on is called TCAPS. They were even worse. They're noise-canceling earbuds with a bone-conducting microphone that sits inside of your ear.
Sounds great - but when you start shooting and moving (and pouring sweat) the earbuds slip out slightly, breaking contact with your ear. Every. Damn. Time. And you can't feel it happen. The way you find out is you key up to speak and no one responds to you. All anyone hears is dead air. After a while someone figures it out and says "last station, check your mic". They went straight into the pile of useless "good idea fairy" nonsense we keep in the arms room, only pulling it out for inventories. And there is a lot if that. We have monocular eye pieces in there that have a heads-up display. Some nerd thought they would be a good idea. Put it on an actual fighter and it is of no added benefit.
The only headset I have used that consistently works are Peltor COMTACs and their recent competitors. Over-the-ear noise cancelling ear muffs with a boom mike. Never had a problem with wind noise. Biggest issue was the mike being pulled away from your mouth when it gets snagged, or people who think they need to yell into the mike. They also had a compatibility issue with PRC-119Fs/ASIPs/RT-1523s where you need to keep the volume setting at 8 or below otherwise you get a high pitched squeal when you key up and no traffic goes out.
I'm curious as to why the author even thought that a throat mic would have good audio quality. Throat mics are designed for loud, high intensity situations. For the use case of a throat mic, "good audio quality" means the receiver can actually understand what you're saying, something that is pretty easy to achieve in a home office environment.
And I'm sure that it's possible to make a throat mic that has great audio quality. But I think the overlap of "people who want throat mics" and "people who want good audio quality" consists solely of the author of this article.
> I'm curious as to why the author even thought that a throat mic would have good audio quality. Throat mics are designed for loud, high intensity situations. For the use case of a throat mic, "good audio quality" means the receiver can actually understand what you're saying, something that is pretty easy to achieve in a home office environment.
The good audio quality I thought would come from it being viable in a "loud, high intensity situation". I thought quite simplistically that there must be something there if people are depending on throat mics in warzones.
To be clear though, the throat mic certainly has "good audio quality" -- I use it on most of my calls now, and I don't get any complaints -- tested with a friend beforehand and he didn't know the difference until I asked.
> And I'm sure that it's possible to make a throat mic that has great audio quality. But I think the overlap of "people who want throat mics" and "people who want good audio quality" consists solely of the author of this article.
Would sure like to find some more people (maybe ones that like to do kickstarters or have manufacturing connections) because I'd sure like to expand this niche!
> I use it on most of my calls now, and I don't get any complaints
It's probably because people you are talking to are nice. It's a bit awkward to ask someone to invest a little in a better mic.
I did complain however to a colleague that his room is somehow generating a low-end feedback or standing wave. I measured the peak frequency and asked him to measure his room, but I couldn't confirm 100% if that's the reason :) Probably nobody else noticed it, it just annoyed me because I have studio monitors, I had to unplug them to understand him better.
If you are curious if it really has good audio quality, render a spectrogram of the throat mic and a professional podcasting mic and compare the image. The richer the image, the better it is.
I enjoyed your article. I think it is very in the spirit of hacking and science that you asked a question and then meticulously chopped away at it until you had a solution-shaped object.
I have also been fascinated by throat-mics from e.g. cyberpunk novels.
Have you also thought about bone conduction to replace headphones? That would be very cyberpunk :)
> I thought would come from it being viable in a "loud, high intensity situation".
The reason they're used is because they're particularly good at rejecting background noise - so that you can hear someone over the sound of nearby gunfire. Throat mics are obviously right at the source of audio vibrations and aren't picking up vibrations from the air (at least not on purpose) so they have high ratio between signal (voice) and noise (unwanted background).
Could you provide samples? Maybe it works for you in particular, but all the throat mic tests I've seen on youtube have absolutely abysmal sound quality.
I don't have much to contribute besides to say that I also run Arch and I also would love to use a high-quality throat mike. My use case is mostly for comms while riding eBikes, not office work. I have a Blue Snowball that I use for voice chat in my quiet, personal home office.
> Throat mics are designed for loud, high intensity situations. For the use case of a throat mic, "good audio quality" means the receiver can actually understand what you're saying, something that is pretty easy to achieve in a home office environment.
I've only configured a handful of throat mics, but they actually are designed to be used when you're speaking at a whisper. That's one of the main features, you're not grabbing a handset and shouting into a directional microphone off your shoulder.
Honestly, your sample doesn't sound very good. It's a bit better than the other samples that have been linked in this thread, and you're perfectly audible, but it's very muffled and there are noises that sound like the throat-mic equivalent of cable noise. If I head this on a call I'd assume you were using a particularly bad laptop mic or gaming headset.
Not bad enough that I'd comment on it, but if I knew you'd gone to the trouble of a custom setup with elaborate extra adapters etc, I'd expect it to sound excellent (like a podcast mic or similar).
Yeah it's definitely not very good -- there are tons of good solutions in this thread now though.
> Not bad enough that I'd comment on it, but if I knew you'd gone to the trouble of a custom setup with elaborate extra adapters etc, I'd expect it to sound excellent (like a podcast mic or similar).
That (used to be) the best part! No one knew how much trouble I went through for this subpar setup!
To me it just sounds bad, like he’s coming in over a poor cellphone connection. If I was in a call I’d find that I’d have to put some effort in to listen.
It's bad compared to normal microphones, but it's much better than I expected (especially if you turn the volume up a bit). If it actually avoids capturing background noise regardless of level, it's great as soon as you can't get a quiet environment.
I have a pretty serious audio setup for WFH. I assessed throat mics and bone conduction headphones and IEMs, I then ultimately decided against these options and instead went for a front-address microphone on a boom and open circumaural headphones.
In my testing, all of these technologies intended for stealth (primarily) were uncomfortable for long periods of wear/use, and traditional options were massively more comfortable. I wear headphones all day, for instance, as I listen to music when I’m not in meetings.
I love bone conduction headphone. I am currently using the Aftershokz Opencomm. The good thing is that they don't obstruct your ears, and you maintain full awareness of the outside world. Great when you are on the streets.
If that't not what you want, and when you are working from home, it is usually the case, then bone conduction headphones are not for you. When working from home, I am using a more traditional headset (Sennheiser Game One), which has much better sound quality.
For the microphone, I use the built-in headset mic, which is acceptable but not great. Best result I have heard on the receiving end is with a dynamic microphone, namely the Shure SM58, which is by far the most popular microphone for live vocals.
Static mics are great in a studio, but most people don't have studios. Dynamic mics do a great job at rejecting background noise, that's why they are so popular in live performances, and that's why they are also great in a noisy room. One logical constraint that goes with it is that you need to put it very close to your mouth, if you are doing video calls, you can't have it out of the frame, you will look like a rock singer ;)
For what it's worth I have bone conduction headphones (Aftershokz Aeropex) and I find them very comfortable to wear for long periods (a few hours at a time; I have never used them for a full day), more so than other types of headphones.
They don't cover my ears at all, so obviously they are only good in situations where I don't want to block out sounds from the environment. I tend to use them on a low volume, partly also because when I use them I don't want to block out other sounds, and partly because I don't like loud music most of the time. Within those parameters, I'm very happy with them.
Yes. I have a fairly serious setup too, and chose something similar.
There are tons of different mic options which are designed for different applications--lavalier mics, headset mics, shotgun mics, omni mics, throat mics, boundary mics, and ordinary cardioid mics.
An ordinary cardioid mic is cheap, durable, and gives high-quality audio for one person. It's hard to beat. If you're trying to hide the microphone so it's less visible on-camera, you end up spending more money for worse audio.
The humble headset microphone is also a good option--a bit more utilitarian.
I tend to agree with you in the main - I use MDR-7506's and a sE Dynacaster as my normal setup - but you absolutely can get comfortable IEMs. But they really aren't cheap. If I need to do something that looks "cans free" at my desk I keep a set of Shure SE215s, which aren't super comfortable but they're tolerable for an hour or two.
To me, reasonably affordable throat mics sound terrible, which is a big factor.
I've worn Shure SE215s for 4+ hours at a time countless times for professional broadcasts. I found them to be very comfortable, and actually do use them at home for conferencing. It's definitely a personal preference thing, although I'm sure the more expensive ones are even more comfortable.
The headphones thing is very much a matter of morphology and personal preference. I know for instance anything other than IEMs is intolerable over any significant period of time for myself.
And choice of headphone! Open versus closed headphones lie much different on your head; clamp strength and weight vary hugely as well. I'm on my third set of AKG k702s as the only headphones I can wear for hours; finding the right pair for anyone takes time.
I use my Sennheiser PC38X for every work and personal call/conference I do. They have an excellent mic and great sound. I'd use them for music daily if I didn't WFH and have excellent speakers and even better headphones for just music.
I should write up my solution using a parabolic dish that focuses sound from a relatively small zone where my head (usually) is, into a good cheap mic. You get good isolation (even from keyboard sounds) requires nothing on your head or throat, and ... physics! (You just need to accept a big dish in your field of view, and you also can't pace, walk around, as with bluetooth solutions.)
You know what is crazy, those microphone guns[0] mics might actually be amazing.
I know that people (youtubers) actually have problems with being able to turn their A/Cs on because of the background noise... Rather than trying to remove that noise with AI/post processing or turning the A/C off maybe people should be looking into microphone guns.
Marketing has pushed a lot of Youtubers towards LDCs like the Blue Yeti, which are great in a studio setting but pick up lots of background noise. For some reason everyone considers the SM7B as the way around that (despite being overkill and overpriced for their applications), which is a broadcast dynamic. Dynamics can handle fan noise and busy streets no problem.
There are less expensive handheld dynamics which are suitable for Youtubers, like the SM58 (very few of them use that, but it's ubiquitious in live sound and with processing can sound very decent). Sennheiser and AKG also make good stuff. If those don't provide enough isolation there are ENG dynamics that are designed to handle interviews in the middle of loud protests and such.
A smaller form factor solution would be an array of mems mics and some software for beam forming. It's pretty popular in enterprise teleconference equipment these days. You can also solve the pacing problem by steering the beams but that's a difficult software problem to solve
I used the crappiest $5 mic I had with a large steel salad bowl to listen to normal conversations 100 feet away. It's incredible what you can pull up. A normal sized bowl mounted on a little tripod on your desk would work great. Add some head tracking with a webcam and you can go mobile. Or just have someone point it at your mouth the whole time like in the NFL: https://imgur.com/a/hyN9V
So, to go even more nerdy on this idea, create a microphone array and steer the input precisely at your mouth. Bonus, you can program the FPGAs to do this yourself!
This is what some "conference phones" do -- I have an Anker powerconf, which has an array of six microphones and is very good at beamforming. It sits on my desk for general use, and I also use it for meetings with multiple local participants. It's better than the much-more-expensive conference room mic setups we have in the office.
Even mics with passive rejection (basically an omnidirectional mic facing outwards that passively cancels the input from the cardioid mic facing your mouth) do pretty well at these tasks.
Antlion and Sennheiser make good passive rejection mics.
I went down this path once. For normal life everything about throat mikes is awful: the sound quality, discomfort, difficult to get positioned correctly, and get to transmit your swallowing sounds to the world.
I should have added that to my post - throat mikes are very uncomfortable. The band would also grab stubble so sometimes when I would turn my head I would get a nice sting as it tugged on a hair.
Bluetooth is a good idea, but you need to be wary of the emissions you're relying on for critical comms. If you have a whole formation using a piece of the spectrum, that piece of the spectrum is now a target.
On Bluetooth audio issues: try switching from PulseAudio to PipeWire; it’s better in every way now save perhaps for one or two remaining tiny niches, and pipewire-pulse lets it act as a drop-in replacement for pulseaudio. The usual installation experience is completely uneventful: you install it, restart your session, and everything works at least as well as before.
On the throat mic itself: what does it sound like? I need a sample. I’m having difficulty finding any not-obviously-bad audio samples (e.g. not recording the microphone, but rather the output of a speaker after radio transmission!).
> Good mic sound quality, switch was not detectable to people on the other end, and was clear.
From the impressions I’m getting, I’d be surprised if “not detectable” was accurate—rather, people are used to lousy audio and are not likely to comment on it unless you ask them or it’s just too bad. But yeah, I’d rather like a sample to judge myself.
> On Bluetooth audio issues: try switching from PulseAudio to PipeWire; it’s better in every way now save perhaps for one or two remaining tiny niches, and pipewire-pulse lets it act as a drop-in replacement for pulseaudio. The usual installation experience is completely uneventful: you install it, restart your session, and everything works at least as well as before.
I did! Pipewire is working awesome -- I'm using pipewire-pulseaudio actually, there's a package on arch that IIRC bridges the two so I can use pavucontrol etc.
> On the throat mic itself: what does it sound like? I need a sample. I’m having difficulty finding any not-obviously-bad audio samples (e.g. not recording the microphone, but rather the output of a speaker after radio transmission!).
Just put up a I'm going to put one up right now -- just updated the blog and it's deploying, should be up within 3 min or so.
> From the impressions I’m getting, I’d be surprised if “not detectable” was accurate—rather, people are used to lousy audio and are not likely to comment on it unless you ask them or it’s just too bad. But yeah, I’d rather like a sample to judge myself.
This definitely might have been the case, but the friend I was talking with I most certainly have run bad audio by before -- so he's seen worse and better.
That sounds very bad. Not quite as bad as any samples I had found, and still reasonably understandable, which some lousy device microphones (especially laptops’) aren’t—it’s bad in a somewhat different way. But I would strongly dislike having to talk to you with that as your microphone, especially if you were in a context where you could have used almost any other microphone. If I knew that you had put me through such a thing for no good reason (which could be context, physical disability, lack of possession of an alternative), I would be very likely to resent that fact.
Frankly, even if others don’t comment on it, you’re making talking with you far more stressful (whether consciously or subconsciously), because people are having to put a lot more effort into understanding you.
> Could Throat mics be the ultimate in high quality audio for cheap? (not really).
No. Absolutely not. Any podcast using such a microphone will bomb hard.
Also: pipewire-pulse doesn’t bridge the two in the most likely understanding of the word “bridge”—rather, it implements the PulseAudio protocol on top of PipeWire. In typical Linux package manager terms, it conflicts with and provides pulseaudio and pulseaudio-bluetooth.
Readit News
The tl;dr:
> tl;dr - Don’t chuck the Sony 1000XMs/AirPods just yet. Throat mics are an improvement over a standalone wired microphone (if that’s what you are using), and have better background noise reduction but in a quiet room your usual bluetooth headset is still way more convenient.
The new thing the DOD has spent an UNGODLY amount of time and money on is called TCAPS. They were even worse. They're noise-canceling earbuds with a bone-conducting microphone that sits inside of your ear.
Sounds great - but when you start shooting and moving (and pouring sweat) the earbuds slip out slightly, breaking contact with your ear. Every. Damn. Time. And you can't feel it happen. The way you find out is you key up to speak and no one responds to you. All anyone hears is dead air. After a while someone figures it out and says "last station, check your mic". They went straight into the pile of useless "good idea fairy" nonsense we keep in the arms room, only pulling it out for inventories. And there is a lot if that. We have monocular eye pieces in there that have a heads-up display. Some nerd thought they would be a good idea. Put it on an actual fighter and it is of no added benefit.
The only headset I have used that consistently works are Peltor COMTACs and their recent competitors. Over-the-ear noise cancelling ear muffs with a boom mike. Never had a problem with wind noise. Biggest issue was the mike being pulled away from your mouth when it gets snagged, or people who think they need to yell into the mike. They also had a compatibility issue with PRC-119Fs/ASIPs/RT-1523s where you need to keep the volume setting at 8 or below otherwise you get a high pitched squeal when you key up and no traffic goes out.
javascript: (location.hostname="teddit.net")
it takes a reddit url and loads it on teddit
Eternal September is eternal
And I'm sure that it's possible to make a throat mic that has great audio quality. But I think the overlap of "people who want throat mics" and "people who want good audio quality" consists solely of the author of this article.
The good audio quality I thought would come from it being viable in a "loud, high intensity situation". I thought quite simplistically that there must be something there if people are depending on throat mics in warzones.
To be clear though, the throat mic certainly has "good audio quality" -- I use it on most of my calls now, and I don't get any complaints -- tested with a friend beforehand and he didn't know the difference until I asked.
> And I'm sure that it's possible to make a throat mic that has great audio quality. But I think the overlap of "people who want throat mics" and "people who want good audio quality" consists solely of the author of this article.
Would sure like to find some more people (maybe ones that like to do kickstarters or have manufacturing connections) because I'd sure like to expand this niche!
It's probably because people you are talking to are nice. It's a bit awkward to ask someone to invest a little in a better mic.
I did complain however to a colleague that his room is somehow generating a low-end feedback or standing wave. I measured the peak frequency and asked him to measure his room, but I couldn't confirm 100% if that's the reason :) Probably nobody else noticed it, it just annoyed me because I have studio monitors, I had to unplug them to understand him better.
If you are curious if it really has good audio quality, render a spectrogram of the throat mic and a professional podcasting mic and compare the image. The richer the image, the better it is.
The reason they're used is because they're particularly good at rejecting background noise - so that you can hear someone over the sound of nearby gunfire. Throat mics are obviously right at the source of audio vibrations and aren't picking up vibrations from the air (at least not on purpose) so they have high ratio between signal (voice) and noise (unwanted background).
I've only configured a handful of throat mics, but they actually are designed to be used when you're speaking at a whisper. That's one of the main features, you're not grabbing a handset and shouting into a directional microphone off your shoulder.
The very first time I watched the movie "Speed." I've wanted one ever since. Keanu Reeves made it look so "tacticool" it hurt.
Not bad enough that I'd comment on it, but if I knew you'd gone to the trouble of a custom setup with elaborate extra adapters etc, I'd expect it to sound excellent (like a podcast mic or similar).
> Not bad enough that I'd comment on it, but if I knew you'd gone to the trouble of a custom setup with elaborate extra adapters etc, I'd expect it to sound excellent (like a podcast mic or similar).
That (used to be) the best part! No one knew how much trouble I went through for this subpar setup!
I'd definitely comment that you need to move your mic from your clothes or something if I was on a call with you and had to suffer that cable noise.
Deleted Comment
In my testing, all of these technologies intended for stealth (primarily) were uncomfortable for long periods of wear/use, and traditional options were massively more comfortable. I wear headphones all day, for instance, as I listen to music when I’m not in meetings.
If that't not what you want, and when you are working from home, it is usually the case, then bone conduction headphones are not for you. When working from home, I am using a more traditional headset (Sennheiser Game One), which has much better sound quality.
For the microphone, I use the built-in headset mic, which is acceptable but not great. Best result I have heard on the receiving end is with a dynamic microphone, namely the Shure SM58, which is by far the most popular microphone for live vocals.
Static mics are great in a studio, but most people don't have studios. Dynamic mics do a great job at rejecting background noise, that's why they are so popular in live performances, and that's why they are also great in a noisy room. One logical constraint that goes with it is that you need to put it very close to your mouth, if you are doing video calls, you can't have it out of the frame, you will look like a rock singer ;)
Also for anyone in an office who doesn’t want to put headphones on/off or in/out to interact with present people interspersed with remote people.
The all day “nothing on or in your ears” aspect is comfortable and healthy too.
They don't cover my ears at all, so obviously they are only good in situations where I don't want to block out sounds from the environment. I tend to use them on a low volume, partly also because when I use them I don't want to block out other sounds, and partly because I don't like loud music most of the time. Within those parameters, I'm very happy with them.
YMMV
There are tons of different mic options which are designed for different applications--lavalier mics, headset mics, shotgun mics, omni mics, throat mics, boundary mics, and ordinary cardioid mics.
An ordinary cardioid mic is cheap, durable, and gives high-quality audio for one person. It's hard to beat. If you're trying to hide the microphone so it's less visible on-camera, you end up spending more money for worse audio.
The humble headset microphone is also a good option--a bit more utilitarian.
To me, reasonably affordable throat mics sound terrible, which is a big factor.
https://drop.com/buy/drop-epos-pc38x-gaming-headset
A good wired microphone absolutely destroys the sound quality from bluetooth headsets.
I have, however, moved to a Jabra bluetooth. The wireless freedom is top-notch.
You know what is crazy, those microphone guns[0] mics might actually be amazing.
I know that people (youtubers) actually have problems with being able to turn their A/Cs on because of the background noise... Rather than trying to remove that noise with AI/post processing or turning the A/C off maybe people should be looking into microphone guns.
[0] https://www.amazon.com/Sony-ECMGZ1M-Zoom-Microphone-Black/dp...
There are less expensive handheld dynamics which are suitable for Youtubers, like the SM58 (very few of them use that, but it's ubiquitious in live sound and with processing can sound very decent). Sennheiser and AKG also make good stuff. If those don't provide enough isolation there are ENG dynamics that are designed to handle interviews in the middle of loud protests and such.
Antlion and Sennheiser make good passive rejection mics.
There's a kickstarter in here, I know it! I just need someone on HN that is crazy about hardware to see this post.
On the throat mic itself: what does it sound like? I need a sample. I’m having difficulty finding any not-obviously-bad audio samples (e.g. not recording the microphone, but rather the output of a speaker after radio transmission!).
> Good mic sound quality, switch was not detectable to people on the other end, and was clear.
From the impressions I’m getting, I’d be surprised if “not detectable” was accurate—rather, people are used to lousy audio and are not likely to comment on it unless you ask them or it’s just too bad. But yeah, I’d rather like a sample to judge myself.
I did! Pipewire is working awesome -- I'm using pipewire-pulseaudio actually, there's a package on arch that IIRC bridges the two so I can use pavucontrol etc.
> On the throat mic itself: what does it sound like? I need a sample. I’m having difficulty finding any not-obviously-bad audio samples (e.g. not recording the microphone, but rather the output of a speaker after radio transmission!).
Just put up a I'm going to put one up right now -- just updated the blog and it's deploying, should be up within 3 min or so.
> From the impressions I’m getting, I’d be surprised if “not detectable” was accurate—rather, people are used to lousy audio and are not likely to comment on it unless you ask them or it’s just too bad. But yeah, I’d rather like a sample to judge myself.
This definitely might have been the case, but the friend I was talking with I most certainly have run bad audio by before -- so he's seen worse and better.
That sounds very bad. Not quite as bad as any samples I had found, and still reasonably understandable, which some lousy device microphones (especially laptops’) aren’t—it’s bad in a somewhat different way. But I would strongly dislike having to talk to you with that as your microphone, especially if you were in a context where you could have used almost any other microphone. If I knew that you had put me through such a thing for no good reason (which could be context, physical disability, lack of possession of an alternative), I would be very likely to resent that fact.
Frankly, even if others don’t comment on it, you’re making talking with you far more stressful (whether consciously or subconsciously), because people are having to put a lot more effort into understanding you.
> Could Throat mics be the ultimate in high quality audio for cheap? (not really).
No. Absolutely not. Any podcast using such a microphone will bomb hard.