Frequency Sweep Test: Play Sine Sweeps from 20 Hz to 20 kHz
Click Start Sweep on this frequency sweep test to sweep 20Hz to 20kHz and catch any dropouts or rattles. Run the check your hearing range directly in your browser — it uses your camera or microphone live, and nothing you capture is ever uploaded.
Press play and the frequency sweep test above glides one pure tone through every pitch your audio gear can make, so you hear where the sound swells, thins or buzzes. Unlike fixed sweep test tones or separate test tones, these sweeps cover 20 Hz to 20 kHz (or any band in Hz and kHz) in seconds, and weak spots in your playback chain stand out the moment the level jumps.
How the Frequency Sweep Test Works
This frequency sweep generator produces one continuous signal whose frequency rises or falls between two points you set. The sweeper runs in Chrome, Safari and Firefox, so the frequency sweep tool needs no download. With one tone at a time, the sweep generator gives each frequency the full signal level. Run the left and right stereo test directly in your browser — it uses your camera or microphone live, and nothing you capture is ever uploaded.
Setting the Start Frequency and Stop Frequency
Enter a start frequency, a stop frequency and a duration in seconds, then pick a waveform: sine for clean checks, or square, sawtooth and triangle, which add harmonics. Start volume and end volume boxes let the signal fade in or out.
Log Sweep or Linear Sweep Test Sound
A log sweep spends equal time on every octave, the way your ears hear pitch. A linear sweep adds the same hertz each second, so it races through the low end; its energy spreads like white noise, while a logarithmic one sounds closer to pink. The logarithmic frequency at time \(t\) is:
$$f(t) = f_{1} \times \left(\frac{f_{2}}{f_{1}}\right)^{t/T}$$
Take 25 Hz to 12,800 Hz over 18 seconds: \(12{,}800 \div 25 = 512 = 2^{9}\), so 9 octaves of 2 s each, passing 566 Hz at 9 s. The linear version climbs 709.7 Hz per second and is already at 6,412 Hz halfway through.
| Octave | Logarithmic (s) | Linear (s) |
|---|---|---|
| 25–50 Hz | 2.0 | 0.035 |
| 100–200 Hz | 2.0 | 0.141 |
| 400–800 Hz | 2.0 | 0.564 |
| 1.6–3.2 kHz | 2.0 | 2.254 |
| 6.4–12.8 kHz | 2.0 | 9.018 |
Below 200 Hz the log run plays for 6 s, the linear one for 0.25 s, so use the log version by ear.
What a Sine Sweep Reveals About Frequency Response and Room Modes
As the sine sweep rises, loudness should change smoothly. Any jump reveals the frequency response of your audio system at your listening location: Run the measure audio round trip delay directly in your browser — it uses your camera or microphone live, and nothing you capture is ever uploaded.
- Peaks: a boost from room modes or other resonant frequencies of the listening room.
- Dips: holes where reflections cancel (anti-resonances).
- Rattling: furniture or trim buzzing at its resonances.
- Fade-out: the speaker's limit at either end.
No real room gives a flat response, so look for sharp, narrow changes, not gentle slopes.
Bass and Subwoofer Test Sound
The low frequency band is where rooms misbehave most. Sweep 20–120 Hz slowly to hear where your subwoofer becomes audible and where bass booms. If the hand-off to your loudspeakers dips, flip the sub's phase switch and rerun. Full-bandwidth speakers benefit too.
Hearing the Test Tone at a Safe Level
Run each pass at a comfortable volume: the test tone puts all its energy at one pitch as it climbs, which can harm your hearing or a tweeter. As the signal rises through the upper mids it seems louder than it is (the Fletcher-Munson, or equal loudness, effect), so judge only sudden jumps along the way. A raspy edge that appears partway up means harmonic distortion from an overdriven amp, which music tends to hide.
Tracking Down a Bass Boom With the Frequency Sweep Tool
A new subwoofer sits in the corner of a 4.37 m by 3.62 m den, and one note in every film booms while the next sounds hollow. The goal is to decide whether to move the sub or just change its crossover, and the answer comes in under a minute: a boom at about 39 Hz, the room's own length mode, not a fault in the sub.
Before pressing play, I predict where the trouble should sit. The first axial mode of a room is the speed of sound over twice its length: \(343 \div (2 \times 4.37) = 39.24\) Hz. The width gives \(343 \div 7.24 = 47.38\) Hz, and the second length mode doubles to 78.49 Hz.
Into the frequency sweep tool go a 28 Hz start, a 179.2 Hz stop and a 16-second duration, sine, logarithmic. That range is 2.68 octaves, so each octave gets almost 6 seconds, slow enough to hear a single mode. From the listening seat, the swell arrives about 2.9 s in, which the log formula puts at 39.2 Hz; a softer bump follows at 4.5 s (47 Hz), then a dip near 8.9 s.
| Event | Predicted | Heard at |
|---|---|---|
| Length mode | 39.24 Hz | 2.9 s |
| Width mode | 47.38 Hz | 4.5 s |
| Second length mode | 78.49 Hz | 8.9 s |
For a sensitivity check I rerun the same band over 6 seconds. The window within 3 Hz of the mode (36.25–42.25 Hz) shrinks from 1.32 s to 0.50 s and the boom turns into a blink that is easy to miss, which is why a quick sweep can make a bad room sound fine.
The caveat that matters: the second length mode lands only 1.5 Hz below the common 80 Hz crossover, so the dip at 8.9 s may be the hand-off between sub and mains, not the room. Flipping the sub's phase and rerunning separates the two; if the dip moves, it was the crossover. The 39 Hz peak won't move with any setting, so the next step is sliding the sub 60 cm out of the corner and sweeping again from the same seat. Moving your chair changes what you hear as much as moving the sub, so keep the seat fixed between runs.
Sweep Measurements With a Spectrum Analyzer
For an objective audio measurement, record the signal with a microphone. On a real time analyzer or spectrum analyzer, turn on peak hold so the whole audio spectrum builds up in one pass; an FFT analyzer with linear bins pairs with the linear version. Measurement software derives the impulse response from a swept sine, then the amplitude response, phase response and distortion. A time window removes reflections, and on a speaker driver the same test can plot impedance. Phones with no audio input play a stored signal after a short trigger tone so the analyzer can sync.
Glide Sweep, Chirp and Stepped Sweep
A glide sweep, or chirp, moves continuously, so a short one loses no resolution. A stepped sweep jumps between fixed points and waits for settling at each. Hold the frequency and step the amplitude instead for an amplitude sweep, the standard linearity test.
Sine Sweeps Compared With Pink Noise and Other Test Signals
Sine sweeps put full power into one frequency at a time, so they stay clear of background sound and pin down a single resonance. Pink noise plays everything at once with equal energy per octave, which suits quick A/B checks. Other test signals fill gaps: a short sweep tone for subwoofers, a full sine sweep for whole-range checks, or a sweep signal 6 dB below full scale for safe gain staging.
Saving Sweep Tones for Your Audio Equipment
To test audio equipment away from a browser, such as a car stereo, save the sweep tone generator output as a lossless WAV rather than MP3, since MP3 encoding can trim the extremes of the audible range.
Frequently Asked Questions
- What does a frequency sweep test check?
- It plays one tone that moves steadily through the frequency range, so you can hear where your speakers, headphones or room boost or cut certain notes. Sudden peaks, dips, rattles or a fade-out at either end show where the playback chain is uneven.
- Should I use a linear or a logarithmic sweep?
- Use a logarithmic (exponential) sweep when listening by ear, because it gives every octave the same time, the way your ears hear pitch. A linear sweep rushes through the bass in a fraction of a second and spends most of its time in the treble, which suits a linear FFT analyzer.
- Which waveform should I choose?
- Choose sine for response checks, since it contains a single frequency at a time. Square, sawtooth and triangle waves add harmonics above the fundamental, which is useful for hearing distortion or buzzes but makes the response harder to judge.
- How can I find room modes and resonances with a sweep?
- Sweep the bass range slowly from your usual listening position and note where the level suddenly swells or drops. Swells are usually room modes or resonances, dips are cancellations, and buzzing at one note points to something in the room rattling.
- Why do some frequencies sound louder even if my system is flat?
- Your ears are more sensitive in the upper midrange, an effect described by the Fletcher-Munson equal loudness curves. Ignore gentle, broad changes in loudness and look only for sharp, narrow peaks and dips.
- Is a frequency sweep safe for my ears and speakers?
- Start at a low volume and only raise it to a comfortable level. A sweep concentrates all its energy at one frequency at a time, so a loud high-frequency tone can damage your hearing or a tweeter.
- How is a sine sweep different from pink noise?
- A sine sweep plays one frequency at a time with the full signal level, so it stands out above background noise and pinpoints individual resonances. Pink noise plays all frequencies at once with equal energy per octave, which is better for quick A/B comparisons and real-time analyzer calibration.