Short answer: it’s not one thing — it’s a stack of boring, very real problems that don’t show up in lab demos 

Longer Reddit-style take:

If you look at papers, acoustic sensing in wearables looks insanely promising. Heart rate without optics, respiration through fabric, gesture recognition, even material sensing.
So why don’t we see it everywhere?

IMO, it’s a combo of these 

1) Signal processing is a nightmare (outside the lab)

In labs:

• controlled environments

• known body positions

• clean signals

In real life:

• people move

• clothes rub

• buses vibrate

• microphones pick up everything

Separating “useful body vibrations” from random noise in real time, on-device, with low latency and low power is hard.
ML helps, but then you hit compute + battery limits.

2) Power consumption kills the business case

Acoustic sensing usually means:

• always-on microphones or piezo sensors

• continuous sampling

• DSP / ML inference

That’s fine for:

• prototypes

• research wearables

• short demos

But for a commercial product that needs days of battery life?
Optical sensors still win on energy-per-signal.

3) Comfort & industrial design are underrated problems

A lot of acoustic sensing works best when:

• sensors are tightly coupled to the body

• placement is very specific

• pressure/contact matters

That clashes with:

• “I forgot I’m wearing it” comfort

• fashion expectations

• one-size-fits-most products

Chest straps work.
Smart rings and loose bands? Much harder.

4) Privacy is a real red flag (even if it’s overblown)

Yes, engineers know:

“We’re not recording speech, just vibrations.”

Users hear:

“There’s a microphone on my body. Always on.”

That’s enough to:

• scare consumers

• scare regulators

• scare legal teams

Optical sensors don’t trigger that gut reaction.

5) Lack of demand… for now

This is the quiet one.

Most consumers are satisfied with:

• steps

• heart rate

• sleep score

• SpO₂ (sometimes)

Acoustic sensing shines when:

• you want clinical-grade insights

• or entirely new interaction modes

That market exists — but it’s still niche.

TL;DR

Acoustic sensing isn’t failing — it’s just early.

It’ll likely show up first in:

• medical wearables

• niche pro devices

• hybrid sensors (acoustic + optical/IMU)

Not because it’s bad —
but because shipping hardware is way less sexy than publishing papers.