Runs in your browser · no install
Audio Test — Left, Right and Stereo Speaker Test
Audio test
Start at a low volume. Tap a test, listen, then compare with the hint.
Signals are generated in your browser. Protect your hearing: keep headphone volume moderate, especially for the sweep's high end.
An audio test is a measurement, not a repair. Its whole value is that it narrows the possibilities before you spend money: a channel that is silent in headphones but fine on the speaker is a headphone fault, and a channel that is silent on both is a phone fault. That single comparison eliminates half of everything people pay a repair counter to discover.
Work through the four checks in the order given. They are ordered so each one removes a class of cause — routing first because it is free, then identity, then balance, then bandwidth. Skipping ahead is how people conclude a speaker is blown when the audio was going to a car stereo two rooms away.
If the test says the hardware is fine but the sound is still dull, the problem is in front of the driver rather than behind it: go to why is my phone speaker muffled. If it is dull because something is in the grille, how to clean phone speakers is the method and grille cleaning tools is the kit. If liquid is involved, start at water eject sound instead of testing anything.
Result → Meaning → Next step
This table is the point of the page. Find the row that matches what you heard.
| What the test produced | What it means | Do this next |
|---|---|---|
| Both channels correct, sweep clean | Audio path is healthy | Nothing — if it still sounds wrong, check EQ and volume-limit settings |
| Sides reversed | Wiring, adapter or a channel-swap setting | Try another cable or port; check accessibility audio settings |
| One side quieter, both audible | Blocked opening on the quiet side | Clean the grille, then re-test |
| One side completely silent | Dead driver, broken conductor or routing | Elimination ladder |
| Silent side is the top of the phone | Earpiece channel — usually lint, often not testable from a browser | Earpiece speaker cleaning |
| Sweep has a silent gap | Driver band damage | Service assessment — cleaning will not restore a missing band |
| Buzz in one narrow region only | Debris or a lint mat resonating | Cleaner sound, then re-check |
| Buzz across the whole sweep | Torn diaphragm or a loose component | Diagnosis and repair routing |
| Gargling or bubbling on any tone | Liquid in the cavity | Water eject sound, then deep dry |
| Everything dull but balanced | Both openings loaded equally, or an EQ setting | Muffled speaker diagnosis |
| Headphones wrong in the same way | System-level audio fault, not the speaker | Restart, check audio settings, then general fix guide |
| Nothing at all from anything | Routing, mute state or amplifier | Speaker not working |
Earbud and headphone channel test
A dead earbud and a quiet earbud are different problems, and the test that separates them takes twenty seconds.
- Test each side alone. Play the left-only tone with both buds in, then the right-only tone. A bud that is silent on its own channel is either not connected or has failed.
- Swap ears. Put the suspect bud in the other ear and repeat. If the fault follows the bud, it is the bud; if it stays with the side, it is the source or the channel setting.
- Re-seat and clean. A quiet bud is very often a wax-blocked nozzle mesh rather than a dead driver. Wipe the mesh dry with a soft brush — never push anything through it — and re-test.
- Re-pair as the last software step. Forget the pairing, put both buds in the case, close it for ten seconds, then pair again. Single-bud dropouts after an update are usually a pairing state rather than hardware.
- If liquid was involved, stop testing and follow the AirPods water eject routine first — repeated tone playback through a wet driver is the one thing that can turn a recoverable bud into a dead one.
What this test cannot tell you
Stating the limits is what separates a diagnostic from a gimmick.
- It cannot drive a phone earpiece. Media audio does not route to the earpiece on iOS or Android, so any browser test that claims to check your call speaker is testing the bottom speaker instead. Test the earpiece with a real phone call.
- It cannot measure loudness in decibels. Your ears, your room and your volume setting all vary. The suite compares one channel against the other, which is a reliable relative measurement, rather than reporting an absolute level it cannot know.
- It cannot distinguish a lint mat from a hairline diaphragm tear when both buzz in the same region. That distinction comes from what happened before the fault: pocket use over months points to lint, a drop or a loud event points to damage.
- It cannot see inside. If the openings look grey and packed, cleaning is worth doing regardless of what the tones say.
- It is not a repair. A test changes nothing about the hardware. Cleaning happens on the cleaning page, clearing on clear my speaker, and device-level health on the speaker test.
Audio test or speaker test — which page do you want?
| Your question | Right page |
|---|---|
| Is my left/right correct, balanced and in the right ears? | This page |
| Is this phone speaker itself healthy before I buy the phone? | Speaker test |
| My speaker makes no sound at all | Speaker not working |
| It works but sounds muffled | Muffled diagnosis |
| I want to shift dust with a tone | Speaker cleaner sound |
| There is water in it | Water eject sound |
Sources & Citations
- W3C, Web Audio API — StereoPannerNode. w3.org/TR/webaudio — specification for the channel panning used by the tests on this page.
- W3C, Web Audio API — OscillatorNode. w3.org/TR/webaudio#OscillatorNode — specification for the locally generated test tones and sweep.
- International Telecommunication Union, Recommendation ITU-R BS.775: Multichannel stereophonic sound system. itu.int/rec/R-REC-BS.775 — standard definition of channel identity and loudspeaker placement.
- Apple Inc., How to clean your Apple products, Apple Support. support.apple.com/en-us/102330 — manufacturer limits on what may touch a speaker or earbud mesh.
- World Health Organization, Make Listening Safe. who.int/activities/making-listening-safe — guidance on sustained playback level, relevant to testing with headphones on.
Step by step
Cut every other audio route first
Turn Bluetooth off completely rather than just disconnecting, and unplug any USB-C or 3.5 mm headset. A paired car system, TV, earbud in a case or laptop dongle can hold the audio route while the device shows full volume. This one step resolves a large share of “no sound” reports before any tone is played.Fix the listening conditions so results are comparable
Take the case off, set media volume to a fixed level you can repeat — about two-thirds is ideal — and hold the device at a consistent distance over a hard surface. A soft bed or sofa absorbs high frequencies and will make a healthy speaker fail a balance check.Test one: channel identity, left then right
Play the left-only tone and confirm it arrives from your left, then the right-only tone on the right. If the sides are reversed, the fault is in the cable, adapter, or a software channel-swap setting — the drivers themselves are fine. If one side produces nothing at all, note which, and continue rather than stopping.Test two: balance between the two channels
Play the same tone on each side at the same volume setting and compare loudness, not tone quality. A clearly quieter side is a partial fault. On a phone the second channel is normally the earpiece above the screen, which is small and easily blocked by pocket lint, so a quieter top channel is far more often dirt than damage.Test three: the frequency sweep, low to high
Listen for a continuous rise with no silent gaps and no buzzing. A gap in the middle of the sweep suggests a damaged driver band; buzzing that appears only in one narrow region usually means a loose particle or a lint mat resonating rather than a torn diaphragm. Note the point where you stop hearing anything — on most phone speakers the low end simply rolls off below roughly 100 Hz, and that is normal.Test four: compare against headphones
Plug in or pair a headset and repeat the identity and balance checks. This is the decisive comparison in the whole suite: if the headphones are correct and the speaker is not, the fault is the speaker or its grille; if both are wrong in the same way, the fault is in the phone’s audio processing or settings, and no amount of cleaning will change it.Repeat the failing test once, deliberately
Rotate the device so the suspect opening faces you, repeat only the test that failed, and confirm the result. Single-trial results in audio are unreliable because room position and hand placement both change what reaches your ear. A fault that survives a repeat with the position changed is a real fault.Convert the result into one next action
Use the verdict table below. Quiet goes to cleaning, silent goes to diagnostics, gargling or crackling after a splash goes to the water pages, and swapped channels goes to cables and settings. The test is finished when you can name one page to open next, not when the tones stop.
Device reference
What each test is measuring, and what a normal result looks like:
| Test | Signal used | Normal result | What a failure narrows it to |
|---|---|---|---|
| Channel identity | Single-channel tone, hard panned | Sound arrives from the named side | Cable, adapter, or channel-swap setting |
| Channel balance | Same tone, each side in turn | Equal perceived loudness | Blocked opening (quiet) or dead driver (silent) |
| Frequency sweep | Continuous rising sine | Smooth rise, no gaps, no buzz | Driver band damage or resonating debris |
| Low-end limit | Sustained low tone | Rolls off gradually, no rattle | Torn diaphragm or loose particle |
| Headphone comparison | Repeat of tests one and two | Headphones correct, speaker matches | Speaker hardware vs system-wide audio fault |
| Routing check | Any tone, Bluetooth off | Sound from the device in your hand | Audio held by a paired or plugged device |
Typical usable band by device class — use this so you do not fail a device for a limit it never had:
| Device class | Practical low limit | Practical high limit | True stereo? |
|---|---|---|---|
| Flagship phone (recent iPhone, Galaxy S, Pixel Pro) | Around 70–90 Hz | Near 20 kHz | Yes — earpiece plus bottom driver |
| Budget or compact phone | Around 120–150 Hz | Around 18 kHz | Often mono |
| Laptop | Around 50–100 Hz | Near 20 kHz | Yes, sometimes multi-driver |
| In-ear earbuds | Around 20–40 Hz | Near 20 kHz | Yes, one driver per ear |
| Portable Bluetooth speaker | Around 55–80 Hz | Near 20 kHz | Model dependent, many are mono |
| Desktop monitor pair | Around 45–60 Hz | Near 20 kHz | Yes, one cabinet per channel |
Note that the high limit you personally hear falls with age — many adults stop hearing above roughly 15–16 kHz, and that is your hearing rather than the speaker. Judge the sweep on continuity and buzz, not on how high you can follow it.