How it works

How the Frequency Detector Measures Hz

The tool takes a short slice of sound and runs a Fast Fourier Transform, which splits that slice into a row of frequency bins and measures how much energy sits in each one. The loudest bin is the dominant frequency, but the bins are coarse: an 8192-point FFT at a 48 kHz sample rate spaces them roughly 5 to 6 Hz apart, so reading the bin alone would round every result to the nearest 5 Hz.

To do better, it looks at the loudest bin and its two neighbours and fits a parabola through the three levels. The peak of that parabola almost never lands exactly on a bin, and its position gives a sub-bin estimate of the true frequency. This parabolic interpolation is standard practice in pitch detection and pulls a clean, sustained tone to within about a hertz, far tighter than the raw bin spacing. From that frequency it computes the nearest equal-tempered note and how many cents sharp or flat the sound sits, using the same math as the note frequency chart.

Uses

What You Can Measure

• Tune an instrument to a reference. Play or sing a note into the mic and read the exact Hz plus the cents deviation from the nearest note, so you can hear and see how far it is out.
• Find a feedback frequency to notch. When a PA or monitor starts to ring, the detector locks onto the offending frequency so you know exactly which band to pull down on the EQ.
• Check a tone generator or test signal. Confirm that a generated test tone is really at the frequency it claims, whether you are testing a subwoofer or calibrating levels.
• Identify a hum or buzz. A steady low tone that reads 50 or 60 Hz is almost always mains hum from the power grid; once you know the frequency you can hunt down the ground loop or notch it out.

Honest limits

Dominant Frequency Is Not the Key of a Song

This matters enough to be blunt about: the dominant frequency of a full music track is usually its loudest bass component, not the key of the song. On a mixed track the kick drum and bassline carry far more energy than the melody or chords, so the single loudest frequency almost always sits down in the low end, somewhere around 40 to 120 Hz. That number tells you where the bass lives, not what key the track is in.

Key is a relationship across many notes, and detecting it needs harmonic analysis of the whole track, not a single peak. If you dropped in a full song to find its key, use the song key and BPM finder instead. To see the whole frequency picture of a track over time rather than one number, the spectrogram analyzer plots energy across every frequency and every moment. The frequency detector is the right tool when you genuinely want one number: a single note, an instrument, a tone, or a feedback or hum frequency.

Mic tips

Getting a Clean Mic Reading

In mic mode the tool asks the browser to turn off echo cancellation, noise suppression, and automatic gain control, because those enhancements reshape the signal and can shift or smear the very frequency you are trying to measure. That is handled for you, but a few things are still up to the room. Get close to the source so the tone you care about is clearly the loudest thing the mic hears, and keep other sound sources quiet: a fan, an air conditioner, or nearby speakers will compete for the peak.

Hold a steady, sustained tone for the tightest reading, since short or wobbling sounds give the interpolation less to lock onto. When the signal drops below the noise gate the readout holds the last stable value rather than flickering to zero, and the peak-hold marker records the highest frequency seen so you can catch a brief tone or a feedback spike. Mic access needs a secure connection, so the page must be served over HTTPS or run on localhost for the browser to grant permission.

Frequently Asked Questions