DevicedialSpeaker test
Speaker test
Playing now
silent
The tone is generated in this tab โ there is no audio file to download. Pick a channel and say out loud where the sound comes from.
- Toneโ
- Output levelโ20 dBFS
- Channelsโ
- Sample rateโ
One gear check a month
MonthlyOne thing a month worth measuring before it fails properly โ a mic that reset itself, a stick that drifts, a switch sending two clicks. Two sentences and the test.
Try also
How do I test left and right speakers?
The speaker test sends a tone to one channel at a time and names the channel it is sending to, so a silent side becomes proof rather than a guess. Play the left tone: the sound should come only from the speaker on your left as you face the screen. Then the right, then both. Anything else is a wiring, balance or output-device problem, and the test has already told you which side to look at.
The tones run from 100 Hz to 12 kHz because speakers do not fail evenly across the range. Small laptop and monitor speakers put out almost nothing below about 150 Hz, so a missing bass tone there is physics rather than a fault; a missing 1 kHz tone is a real fault, because that is the middle of the range speech lives in. The page controls only what it sends โ at the other end, your ears are the meter.
Work upward through the frequencies, one channel at a time. Which tones you cannot hear says more than whether sound comes out at all, because each one fails for its own reason.
| Tone | What it checks | If it is missing |
|---|---|---|
| 100 Hz | Bass output, and cabinet or desk rattle | Normal on laptop and monitor speakers; a fault on a system with a subwoofer |
| 250 Hz | Lower voice range and warmth | Look at bass redirection and any “small speakers” setting |
| 440 Hz | Reference pitch; the easiest tone to judge level on | Missing on one side means that side is receiving nothing |
| 1 kHz | The middle of speech and the standard reference tone | A dead 1 kHz channel is a genuine fault, not a speaker limitation |
| 4 kHz | Consonants and clarity, and where harshness shows up | Muffled here usually means an obstruction or a failed tweeter |
| 8 kHz | Cymbals, sibilance, the sense of air | Often quiet on small drivers; compare the two sides |
| 12 kHz | The top end | Frequently inaudible for innocent reasons โ check both channels before blaming hardware |
If the tones are correct through speakers but wrong through a headset, the split is at the output device: the headphone test checks channels and polarity on the plugged-in pair. When the complaint is that people cannot hear you rather than the other way round, that belongs to the mic test.
Frequently asked questions
Only one speaker plays. Is it dead?
Test that side from another source before replacing anything, because the three usual causes are all software. A balance slider pushed to one side in the system sound settings, a mono audio accessibility option that folds both channels into one, and a 3.5 mm plug that is not pushed fully home all produce the same symptom. If the tone follows the cable to a phone and plays correctly there, the speaker itself is fine.
Sound works in this test but not in a game or a call
Applications keep their own output device, and Windows keeps a separate volume and output per application in the volume mixer. Check that the app is not muted there and that it points at the same device the browser is using. Games in particular remember the device that existed when they were first launched, so a headset unplugged weeks ago can still be selected and playing to nothing.
Which side counts as left?
Left means your left while you face the screen, which is the convention every recording is mixed to. If the test announces left and the sound arrives from the right, two channels are crossed somewhere: reversed speaker cables, an extension lead wired the other way, or a virtual surround feature reassigning channels. Swapping the cables at one end only, then retesting, isolates it in about ten seconds.
I cannot hear the 12 kHz tone at all
That is usually normal rather than a fault. Hearing above roughly 15 kHz fades with age, and small drivers roll off at the top of the range as well, so a missing 12 kHz tone on laptop speakers has two innocent explanations before any hardware one. What matters is symmetry: hearing it clearly on one channel and not the other at the same volume is a real difference and worth chasing.
Should every tone be the same loudness?
No. The tones leave the browser at the same digital level, but no speaker is flat and human hearing is least sensitive at the bottom of the range, so 100 Hz will sound quieter than 1 kHz on almost any system. Compare each tone against the same tone on the other channel rather than against its neighbours โ that comparison is the one that means something.
Does this work over Bluetooth?
Yes, with two caveats. Bluetooth adds latency, typically over 100 ms, so the tone starts noticeably after the button โ that delay belongs to the link, not the speaker. And if the device also carries a microphone, the system may switch to the hands-free profile, which is mono and band-limited; everything above roughly 8 kHz then disappears for reasons that have nothing to do with the drivers.
How the test works
Sources:
What a speaker test can and cannot prove
It proves routing, not quality. If the left tone arrives on the left, the whole chain from the browser to the driver is intact on that side. Frequency response, distortion and stereo imaging need a measurement microphone and a quiet room, and no web page has either.
Silence on one channel is the useful result. It cuts the problem in half immediately: everything shared between the two channels is exonerated, and the fault has to live in the part that is not shared.
Volume is not part of the test. Set the system volume to a normal listening level and leave it there. Changing it between tones makes every comparison meaningless, which is the fastest way to turn a working test into a confusing one.