SpeakerEject
About

About SpeakerEject

SpeakerEject is an independent, free website that plays a water-eject sound for wet phone speakers and runs browser audio tests for speakers, headphones, bass and surround sound, with no app, account or microphone.

Reviewed 8 October 2026

It is built around a simple promise: make each tool easy to use and its limits easy to understand.

Why SpeakerEject exists

Search for a wet or muffled phone speaker and most results are confident promises with little checking behind them. SpeakerEject was made to offer one plain tool with its limits stated, and guides that send you to the manufacturer's own instructions first. The audio tests grew from the same idea: a reader who wants to know whether a speaker, headset or home-theater channel is working should be able to find out in a browser, without installing anything or being tracked.

What you will find here

The site has about 28 pages in four groups. Everything is free. There are no accounts, no page asks for the microphone, and nothing you play or type is uploaded.

What the site does

The water-eject tool generates a short low-frequency sweep using the browser's Web Audio API. It also provides device-specific preparation, drying guidance and a repeatable listening comparison. It does not use a microphone or diagnose phone hardware. The other tools play test signals so you can judge what you hear; they are listening references, not measurements or hearing tests.

Editorial approach

Manufacturer instructions and platform documentation take priority over competitor marketing claims. Pages identify what is documented, what is physically plausible and what remains untested. We avoid invented success rates, reviews and certifications.

How guidance is reviewed

  1. Start from the manufacturer. A guide begins with the current support page for the exact device family, or clearly limits its scope.
  2. Match every claim to a source. Each factual statement about a device, a menu path, a rating or a safety limit is compared with the official page or standard it relies on. The source is linked on the page.
  3. Record the check. The claim, the source, what the source says, the date and the verdict (confirmed, corrected or supported only by secondary sources) go into a source register kept with the site's files.
  4. Show the date. Every guide shows when it was last reviewed, and the date changes only after a real check.
  5. Check again on a schedule. Apple, Google and Samsung revise support pages, often around new releases in the autumn, so the next full recheck is due on 6 January 2027, and sooner if a source changes.

The same principles apply to other kinds of evidence:

  • Technical mechanism: distinguish documentation, patents and physical plausibility from controlled evidence. A patent describes an implementation; it is not proof that our browser tool reproduces it.
  • Search and competitor research: use it to discover reader questions, never as proof that a frequency, success rate or cleaning claim is true.
  • Product testing: label software checks separately from physical-device results. SpeakerEject has not completed controlled wet-device efficacy trials.

What the checks have found so far

The first full pass, on 6 October 2026, covered 113 claims on 16 pages. In that pass 85 claims matched their source, 24 were corrected or credited to the page that actually says them, and 4 rested on secondary sources only. On 16 October 2026 those 4 were checked again against Apple and other primary pages: 1 was confirmed and 3 were corrected. The register now holds 115 checked claims across 16 pages, most recently checked on 16 October 2026. Corrections included:

  • The iPhone guide named MagSafe chargers after a liquid alert. Apple's page names only Qi-certified chargers, so MagSafe was removed.
  • Samsung's humidity, restart and cache steps were credited to a US support page that does not contain them. They now point to the Samsung UK page that does.
  • A safe-listening limit read "below 60%" of maximum volume. The World Health Organization says "no more than 60%", and the page now says so.

We list them so readers can see that the checking changes pages.

What we measure ourselves

Where a page describes a file or tool of ours, we measure it instead of describing it from memory:

  • The water-eject WAV files are checked for frequency range, sweep period, level and format, and the page shows the measured sweep.
  • The surround test files publish SHA-256 checksums, a suggested citation and a free reuse licence (CC BY 4.0).
  • The headphone and phase test files publish sample-level measurements you can download and check.

What we do not claim

  • We do not claim that any sound removes water, or give a success rate. No controlled wet-device trials have been run, and the evidence page explains what is and is not documented.
  • We do not present the tools as measurements, hearing tests or hardware diagnosis.
  • We do not give repair or electrical-safety advice. The guides point to the manufacturer; see the terms of use.

Independence

SpeakerEject is not affiliated with Apple, Google, Samsung or other device manufacturers. Product names belong to their respective owners and are used only to describe compatibility and official guidance. The site currently shows no ads and has no affiliate links or sponsored content. The footer on the home page shows badges for directories where the site is listed or has applied to be listed; they are listings, not endorsements, and the privacy page explains them, along with how analytics and any future advertising work.

Corrections and updates

Guidance is reviewed against linked manufacturer and platform documentation. If a source changes or a statement needs correction, email hello@speakereject.com with the page and supporting source. A confirmed correction is made on the page, and its reviewed date changes only after the source has been checked again.