I have used enough expensive wireless earbuds to know that they can make music sound polished, give podcasts a cozy little studio gloss, and cancel enough train noise that I briefly forget I am sharing public space with humanity. Then someone on a Teams call tells me I sound like I’m broadcasting from a bathroom in a moving vehicle, and the illusion collapses completely.

That’s not a fluke, and it’s not always fixed with a firmware update. There are real, structural reasons this keeps happening, and most of them have very little to do with price.

Your earbuds have two different jobs

And calls expose the weaker one.

When your earbuds play music, they use a Bluetooth profile called the Advanced Audio Distribution Profile, better known as A2DP, which is purpose built for high-quality, one-way audio streaming. A2DP streams high-quality audio in one direction only and supports higher-quality codecs such as AAC and the Qualcomm aptX codec family. That’s the pipeline your expensive earbuds use when they impress you at the gym.

The moment you take a call, the whole setup changes. Your earbuds need to enable two-way communication, which usually means switching to the Hands-Free Profile, or HFP, so the microphone can send your voice back to the device. Older headset implementations may also involve HSP, but HFP is the main one people run into today.

That is where the drop in quality begins. In common Bluetooth call setups, HFP is built around low-bit-rate voice audio, often using narrowband CVSD at 8kHz or wideband mSBC at 16kHz. For comparison, CD audio runs at 44.1kHz. That does not mean every call is automatically dreadful, but it does mean the ceiling is much lower than the music mode you were using five seconds earlier.

The most common scenario for older Bluetooth call audio is 16kHz mono, which takes your immersive stereo music experience and collapses it into something much closer to a flat, center-channel telephone feed. Some devices still fall back to narrower 8kHz audio, which gives speech that familiar dull, muffled character because the useful frequency range is cut off so aggressively.

The premium codec your earbuds use for music is mostly irrelevant here. If you understand what Bluetooth codecs are and how they work, you know that aptX Adaptive, LDAC, and AAC are all A2DP technologies for playback from a device to headphones. They play no role in the call microphone path. Upgrading to earbuds with a fancier music codec does not automatically change how you sound to the person on the other end of your Zoom call. These are separate pipelines with separate constraints.

On Windows, this is often where the problem becomes most obvious. Unless your PC, Bluetooth hardware, drivers, earbuds, Windows build, and calling app all properly support the newer LE Audio path, activating the microphone can still push the headset into a lower-quality communications mode. That is why the same earbuds can sound better on a phone call than on a laptop in Teams. The earbuds are not necessarily changing; the audio chain is.

The microphone is tiny, distant, and surrounded by problems

Your voice is not where the microphones are

Credit: Rob Webb / MakeUseOf

Even if Bluetooth call audio were not a quality ceiling, earbuds would still be fighting physics every time you speak. The hardware picking up your voice is not a studio condenser microphone or even a decent dynamic mic. It is a tiny MEMS capsule sitting near your ear canal, pointed roughly toward your face, competing with everything in your environment.

A standard phone call, with your phone held near your mouth, puts the microphone roughly 15 to 30 centimeters from your lips. A desk headset with a boom arm positions the capsule even closer, often just a few centimeters from the corner of your mouth, with a direct line of sight to the source of the sound. An in-ear earbud sits near your ear, farther from your mouth than the phone you just put down, and often faces outward into the room rather than directly toward your voice.

Stem-style earbuds, such as AirPods, house microphones lower down the stem, which positions them slightly closer to your mouth. That can give them a real voice-pickup advantage over more compact round buds. Compact round designs from Sony, Bose, Samsung, and others often have to place their microphones on the outer shell, which is a worse starting point for voice capture than a stem pointed closer to your mouth.

Walking outside compounds everything. Wind turbulence hitting a small earbud microphone can create low-frequency noise that overwhelms your voice. Typing on a keyboard during a meeting creates a repetitive pattern of sharp, percussive clicks that can confuse noise-suppression systems. A reverberant room, the kind with hard floors and bare walls that now passes for a home office, wraps your voice in reflections that a tiny microphone array is ill-equipped to untangle.

Beamforming helps, but it has limits that show up in your voice

You can call it AI processing, too

Modern earbuds try to address proximity and environmental issues using beamforming. That means using signals from multiple microphones to mathematically focus the pickup toward your mouth while rejecting sound from other directions. You can call it AI processing if you want, and the marketing departments certainly do.

Sony’s recent WF-1000X earbuds, for example, combine multiple microphones, beamforming, and the same sensor technology found in bone conduction headphones to improve voice pickup. That is a lot of engineering aimed at a problem that a boom microphone solves with distance and a foam windscreen.

The bone-conduction sensor picks up vibrations through your jaw and skull rather than through the air, giving the processing system another clue about when you are speaking. But bone conduction is not a replacement for acoustic proximity. It is a clever workaround for the fact that the microphone cannot get any closer to your face.

The other half of the processing problem is that aggressive noise suppression can make your voice sound unnatural on the receiving end. Unlike dedicated tools that denoise audio and remove background noise in post-production, a real-time algorithm working hard to strip out background noise can also strip harmonic detail from speech. This results in a metallic, compressed, or watery quality that makes the listener feel as if they are receiving a heavily processed phone call from 2005.

The noise cancellation you hear in your own ears, which filters incoming environmental sound, is also completely separate from the noise suppression applied to your outgoing microphone signal. Turning up your ANC does not make you sound clearer to the person on the other end. It only makes the world quieter for you.

Apple’s Voice Isolation on iPhone is one of the better examples of platform-level help. It prioritizes your voice and blocks ambient noise, while Wide Spectrum does the opposite, and Standard applies basic voice processing. When you wear compatible AirPods during calls on your iPhone, Voice Isolation can improve how you sound and reduce background noise, such as wind, for the listener. It works reasonably well when conditions are cooperative.

The catch is that this improvement comes from Apple’s platform-level audio processing. It is not a universal Bluetooth miracle. Someone calling from Teams on a Windows laptop or using a different phone-and-headset combination may not get the same result. The quality improvement is real, but it depends on the device, operating system, headset, and app all cooperating.

Better call audio is coming, but you need the whole chain

LE Audio changes the rules, on paper

Bluetooth LE Audio should improve this situation, but the fine print matters. LE Audio is not just a single new call mode. It is a newer Bluetooth audio architecture built around LC3 and a collection of profiles designed to handle modern media and voice use cases more cleanly than the legacy A2DP/HFP split.

In practical terms, that means the brutal quality downshift when a microphone activates does not have to be the default forever. It can support greater efficiency, lower power consumption, and improved handling of voice and modern media features, like how Auracast audio broadcasts, compared with older Bluetooth audio paths. But it only helps if every part of the chain supports it properly.

Qualcomm’s aptX Voice targets a similar problem from the other direction. It offers 32kHz sampled voice audio with a flat 16kHz frequency response, which means it can preserve more high-frequency speech detail than a narrowband Bluetooth call. The limitation, as with all codec improvements, is that both devices need to support it.

Windows 11 is also improving here. Newer builds support better Bluetooth LE Audio behavior when a microphone is active, including stereo playback quality in supported configurations, rather than the old collapse-into-mono audio. But that requires compatible headphones, compatible PC hardware, updated Bluetooth drivers, and the right Windows build. Miss one part of that chain, and you may still end up on the old lower-quality path.

Windows 11 is also improving here. Newer builds support better Bluetooth LE Audio behavior when a microphone is active, including stereo playback quality in supported configurations, rather than the old collapse-into-mono audio. But that requires compatible headphones, compatible PC hardware, updated Bluetooth drivers, and the right Windows build. Miss one part of that chain, and you may still end up on the old lower-quality path.

So, what can you actually do today?

When call quality really matters, picking up your phone and holding it to your face is still one of the most reliable options. The microphone is closer to your mouth, the platform processing is optimized for that exact use case, and you bypass the Bluetooth microphone chain entirely for the uplink.

For desk-based calls, buying a dedicated headset with a boom microphone will outperform almost any pair of wireless earbuds in terms of voice quality. The advantage is physical, and software cannot fully compensate for it. On the iPhone, enabling Voice Isolation during calls in Control Center costs nothing and can make a real difference in noisy environments.

Until the entire chain gets better at working together, expensive earbuds will remain brilliant for music and frustratingly average for meetings. They are not necessarily bad products. They are just optimized for the part you hear, not always the part everyone else hears.

Battery Life

Up to 8 hours

Charging Case Included?

Yes

Microphones

Yes, two microphones

Case battery

Up to 24 hours

Colors

Black, Silver