A slow 15-minute walk; two bells to open, one at the end. Eyes open, no headphones, from a speaker; indoors, somewhere clear of obstacles. Notice your steps, the floor and the sounds around you. Don't listen in traffic, on stairs or anywhere that needs your attention.
Do not listen while driving or operating machinery.
01 / How it's made
Liner notes
Bed
Brown noise
Strikes
Singing bowl · 196 Hz
What do you hear?
swelling
2 layers, 2 sound engines; generated on this device every time it plays.
How is this sound made?
Brown noise−56 dB
Bell−28 dB
Singing bowl · 196 Hz, 3 strikes
In the room you can listen to each layer on its own, change its settings and save it as your own recipe.
Sound is generated on this device in real time; no files are downloaded from a server. There is a 10 s fade-in at the start and a 0 min fade-out at the end.
02 / Compare
Compare with another sound
Hear both at the same loudness; drag the slider to the middle and they blend.
AblendB
In the middle the two sounds truly blend; loudness is matched.
8 trials, half of them the same sound; get at least 7 right and the difference is audible. The result stays on this device.
03 / Evidence
What do we know, what don't we know?
◆B · Moderate evidenceLast reviewed 9 October 2026
A few controlled studies or meta-analyses; reasonable but limited generalisability.
Claim
Meditation, anxiety and stress
Meditation programs probably lead to a small reduction on some anxiety measures.
◆BEvidence level B: Moderate evidence
What do we know?
SupportIn a meta-analysis of 47 randomized trials with active controls (3,515 people), mindfulness meditation programs showed small positive effects on anxiety, depression and pain with moderate-strength evidence; evidence for stress and distress was low.Goyal et al. 2014
What don't we know?
Evidence gapThis evidence mostly comes from structured 8-week programs; whether listening to a sound session alone has the same effect is unknown.Goyal et al. 2014
LimitParticipants came from diverse clinical populations.Goyal et al. 2014
LimitThe programs were not shown to be better than drugs, exercise or other behavioral approaches, and NCCIH notes that much of the research is preliminary.Goyal et al. 2014 · NCCIH 2022
Studies
Goyal et al. 2014 · Meta-analysis · 47 studies · 3,515 participants · moderate certainty · Versus active controls: anxiety effect size 0.38 (95% CI 0.12-0.64; week 8); pain 0.33 (95% CI 0.03-0.62)