Sound healing at home: what the research supports, and how to do it

The trials on sound healing are small, and they point the same way: steady sound lowers anxiety and shifts the body toward rest. Here is what has been measured, and how to get the same twenty minutes at home.

Sophie Kazandjian
July 2026 · 8 min read

Part of: The Science →

A sound bath in a studio near me costs thirty-five euros and lasts an hour. You lie on a mat under a blanket while someone plays gongs and singing bowls over you, and most people leave saying they feel as if they have slept for a week. I have been to a few. As a composer I notice what a sustained low note does to a room full of bodies, and I built an app on the idea that you can have a version of that at home for the price of a takeaway. So bear that in mind when I say the research is still young, and also when I say that what exists is more consistent than most people expect.

The oldest usable evidence is about music rather than bowls. A 2020 review pooled 104 randomised trials with 9,617 participants in which people listened to music, chosen for calm, in stressful situations: before surgery, in intensive care, during procedures. Physiological stress measures fell by a small-to-medium amount, heart rate most clearly; anxiety and restlessness fell by a medium amount. A Cochrane review from 2013 looked at pre-operative anxiety alone and found music listening lowered it by about six points on the standard scale, comparable in one large trial to a sedative. These are not sound healing studies, but they establish what sound healing rests on: the nervous system responds to steady, chosen sound, and the response is large enough to measure in a hospital.

Singing bowl and sound bath studies

The studies of the instruments themselves are fewer and smaller. The one most often cited is from San Diego in 2017, where 62 adults filled in mood questionnaires before and after a single Tibetan singing bowl meditation. Tension, anger, fatigue and low mood all dropped, and people who had never done it before dropped further than the regulars. There was no control group, so the study cannot separate the bowls from an hour lying down in a quiet room, and the authors say so.

A Chilean trial from 2023 did add controls. Fifty anxious adults were randomised to a fifty-minute live singing bowl session, a progressive muscle relaxation session, or a waiting list. The bowl group showed the largest fall in state anxiety, a shift in their EEG toward a calmer pattern, and a rise in heart rate variability, which is the measure of parasympathetic activity that runs through everything on this site. There were sixteen people in the bowl group and one session, so it is a pilot, and a promising one. Two systematic reviews, one in 2020 and one in 2025, come to the same place from opposite directions: the first found four studies of low quality and called the evidence insufficient; the second found nineteen and said the practice "may have potential" for anxiety, low mood and sleep, with small samples and mixed methods throughout.

So something happens. It has been measured a few dozen times in a few hundred people, and the next step is a large trial that separates how much is the sound and how much is fifty minutes lying down with your phone in another room. Either way, the person on the mat gets the benefit.

Binaural beats: the best-studied part of sound healing

Binaural beats are the part of sound healing that has been studied most, because they are easy to standardise: play one frequency in the left ear and a slightly different one in the right, and the brain perceives a slow pulse at the difference. A 2019 meta-analysis of 22 studies found a medium-sized effect across anxiety, attention, memory and pain, larger when the beats played for longer and when they started before the task rather than during it. Theta-range beats, the four-to-seven-per-second pulses used in the deeper stages of the app, were the ones most associated with reduced anxiety; the noise and beats article explains how a beat differs from the noise it sits on. A 2023 review of the EEG studies found the picture on brainwaves themselves is unsettled: five studies saw the brain's rhythm follow the beat, eight did not, one was mixed. The relaxation effect is better supported than the brainwave explanation usually given for it, and I try to say so whenever I describe what the app does.

How sound healing reaches the skin

No study of bowls, gongs, sound baths or beats has yet measured a skin outcome directly: barrier function, hydration, acne, redness, healing. The link from sound to skin runs through the stress physiology, which is well documented: a calmer nervous system runs lower cortisol and inflammation, and skin repairs its barrier and clears blemishes faster in that state. That is the claim I make, and the sound and skin article sets out how it works. Claims that sound "detoxes" or "regenerates" skin go further than anything measured, and I would rather you had the version that holds.

How to do sound healing at home

The studio version has three things going for it that the research supports: a fixed length, a body that is lying still, and sound that fills the ears without demanding attention. All three can be reproduced in a bedroom, and a bowl is optional.

Lie down; the effects in the trials came from people who were horizontal, and the parasympathetic shift is stronger that way. Use headphones rather than a speaker if you are using binaural beats, because the effect depends on each ear hearing a different tone. Choose a duration in advance and let the sound end itself; the meta-analysis found longer exposure worked better, and twenty minutes is a fair floor. Keep the volume low, lower than feels natural, because the studies used moderate levels and there is no benefit in loud. And do the same thing at the same time for a few weeks before judging it. Single sessions produce the effects above; repeating them is what changes how you feel when you wake up.

If you want a bowl, a mid-sized one struck once every minute or so, letting the note die away completely each time, gives you most of what the studio does. Without one, the app plays a low sustained tone with a brown-noise bed under it and a binaural pulse in the later stages, and the free ten-minute taster is the same thing cut short.

I still go to the studio sometimes. The thirty-five euros buys an hour in which nobody can reach me, and no study has measured that yet.

Try it free →

Twenty minutes was the floor above; ten is enough to feel what the routine does. This is the free taster, headphones on.

ON YOUTUBE · 10 MINUTES

Calm for stressed skin: the free 10-minute routine

The same routine as the free try page, on YouTube: 396 Hz to settle, 7.83 Hz Schumann, then 4 Hz theta. Headphones on for steps 2 and 3.

Watch on YouTube →