02 / Explore · Guide05 / Menu · Guide

Guide · Safety1 min read ·

Safe listening

The master level on a schematic ten-second plot: it starts with a soft fade-in, rises by at most 6 dB per second and stops at the −1 dBFS ceiling, the line of the true-peak limiter. dBFS is the file's scale, not the sound pressure at the ear; a browser cannot know dBA. Not a measurement.
The master level on a schematic ten-second plot: it starts with a soft fade-in, rises by at most 6 dB per second and stops at the −1 dBFS ceiling, the line of the true-peak limiter. dBFS is the file's scale, not the sound pressure at the ear; a browser cannot know dBA. Not a measurement.

Level and duration

The safe-listening standard of the World Health Organization and the International Telecommunication Union (H.870) recognises that weekly sound exposure depends on level and duration together: when the level rises by 3 dB, the time for the same exposure halves.

A browser or app cannot know the actual sound pressure reaching your ears, so Frekans Studio never shows dBA or a "safe level" on screen. Your device's own headphone-safety features (e.g. "Headphone Safety" on iOS) can make this measurement.

Frekans Studio's protections

Sessions start with a fade-in, the level rises by at most 6 dB per second and the output is limited at −1 dBFS. During long headphone listening you are reminded to take a break every 2 hours. Gamma sessions show a note about ringing in the ears; if you feel discomfort, the "Report discomfort" button stops the sound immediately.

What does the evidence say?

    Context · not a claim

    The equal-loudness contours of pure tones vary with frequency; the 2004 review found differences from the older contours that often reach 14 dB below 500 Hz. The same numeric level is not heard as equally loud at every frequency.

    Evidence levels and sources come from the database; the article adds no new claims. Evidence methodology

    Studies

    Rows come from the evidence database's source records: finding, sample and certainty from the source itself; the level is that of the claim citing it. The table adds no new numbers.
    WorkTypenFindingLevel
    Suzuki 2004[1]Experiment12 studiesBrings together the results of 12 studies from the mid-1980s onwards to propose new equal-loudness-level contours for pure tones. Differences from the 1956 Robinson and Dadson contours are greatest below 500 Hz and often reach 14 dB.

    Frequently asked questions

    How loud is too loud?
    A practical check: you should be able to talk normally with someone next to you. Keep your device volume preferably below 60%.
    My ears are ringing. What should I do?
    Stop listening and rest somewhere quiet. If the ringing persists, see a health professional.

    Who is it for?

    For whom

    • Anyone looking for sound for relaxation, focus or a sleep routine; the sounds are generated on your device and promise no effect.
    • Anyone who wants to read the evidence level of what they listen to.
    • Everyone who listens for long periods or on headphones; the measure is level × duration, a browser cannot know the actual sound pressure, so the page gives rules rather than numbers.

    Not for whom

    • If you have epilepsy or a history of seizures, a psychiatric diagnosis, a medical device such as a pacemaker, tinnitus or any other health condition, or if you are pregnant, consult your doctor before use.
    • Do not listen while driving or operating machinery.
    • Keep your device volume comfortable, preferably below 60%. You can also use your device's headphone-safety settings.

    Frekans Studio is not a medical device; it does not diagnose, treat or prevent any disease.

    Listen and try

    Sources

    1. Suzuki Y, Takeshima H. Equal-loudness-level contours for pure tones. J Acoust Soc Am. 2004;116(2):918-933. doi:10.1121/1.1763601 PMID 15376658

    Citations come from the evidence database; every source is verified with Crossref or PubMed. Links go to the publication's own page (DOI).

    More on this topic

    Where to go next

    Method

    Sources
    This article is not linked to a claim in the evidence database; it explains the technique and its use.
    Editorial responsibility
    Murat Tunalı · Editorial responsibility lies with Murat Tunalı; no independent medical review has been carried out.
    How this article was prepared
    The first draft was prepared by an AI model (Claude); the article adds no new figures, citations or claims, and the evidence summary comes from the database. Murat Tunalı approved publication. No independent scientific adviser has reviewed it yet. Evidence methodology · Credits
    Limitations
    This article is for information only; it is not medical advice. Evidence levels are per claim; a session's level is the weakest of its linked claims.
    Updates
    Last updated: 18 October 2026
    Cite as
    Frekans Studio. “Safe listening”. Last updated: 18 October 2026. https://frekans.studio/en/guide/safe-listening

    No cookies, no tracking. Details: Privacy notice (KVKK) On-device storage