Harmonic Overtone Detector

Detect and visualize up to 16 harmonics of any sound in real time. See amplitude (dB) per harmonic, odd/even classification, THD percentage, instrument timbre matching, and harmonic-to-fundamental ratios. Click any harmonic bar to hear it isolated. 100% browser-based, no audio uploaded.

Harmonic Overtone Detector Tool

Understanding Your Results

Harmonics and Overtones

A harmonic is an integer multiple of a fundamental frequency. If a guitar string vibrates at 220 Hz (A3), its harmonics appear at 440 Hz (H2), 660 Hz (H3), 880 Hz (H4), and so on up to H16 at 3520 Hz. The first harmonic (H1) is the fundamental itself. Overtones are all harmonics above the fundamental — the first overtone is the second harmonic (H2). The unique blend of harmonic amplitudes is what makes a clarinet sound different from a trumpet playing the same note.

Odd vs. Even Harmonics

Instruments with predominantly odd harmonics (H1, H3, H5, H7…) have a hollow, woody tone — the clarinet is the classic example, because its cylindrical bore naturally suppresses even harmonics. Instruments with a full harmonic series (both odd and even) sound richer and brighter — trumpet, violin, and open-bore flutes. A perfectly symmetrical waveform (square wave) contains only odd harmonics. The odd/even balance indicator shows you this ratio in real time.

Total Harmonic Distortion (THD)

THD quantifies how much energy exists in harmonics compared to the fundamental. It is calculated as the square root of the sum of squared harmonic amplitudes divided by the fundamental amplitude: THD = sqrt(H2² + H3² + … + H16²) / H1. A pure sine wave has 0% THD. A square wave has ~48% THD. Typical instrument THD ranges: flute 5–15%, clarinet 10–25%, trumpet 20–40%, violin 30–60%. THD below 1% indicates an extremely clean sine tone; above 50% indicates a very harmonically rich or distorted signal.

Instrument Timbre Matching

The timbre matcher compares your harmonic profile (relative amplitudes of H1–H16) against stored templates for common instruments. It uses normalized correlation — the instrument with the closest harmonic envelope is shown with a confidence score. 90%+ means a very strong match. 50–89% is a reasonable match. Below 50% means the harmonic profile does not closely resemble any template — the sound may be synthetic, distorted, or a non-template instrument.

Harmonic-to-Fundamental Ratio

The ratio table shows each harmonic's amplitude relative to the fundamental (H1 = 0 dB reference). A ratio of −20 dB means that harmonic is 20 dB quieter than the fundamental. In most natural sounds, higher harmonics decay — H2 might be −6 dB, H8 might be −30 dB. Harmonics that are unexpectedly strong (closer to 0 dB) at high numbers indicate distortion, resonance, or formant reinforcement.

Click-to-Play Harmonic Isolation

Clicking any harmonic bar plays a pure sine tone at that harmonic's exact frequency, so you can hear what each overtone sounds like individually. This is useful for ear training, identifying which harmonics create perceived tone color, and verifying that the detected frequencies are correct by listening to them.

Technical Background

How Harmonic Detection Works

The Harmonic Overtone Detector uses a two-stage approach to identify and measure harmonics in real-time audio. First, the fundamental frequency (F0) is detected using time-domain autocorrelation — the signal is correlated with delayed copies of itself, and the lag with the highest correlation peak corresponds to the period of the fundamental. Autocorrelation is preferred over pure FFT peak-picking for F0 detection because it correctly identifies the true pitch even when the fundamental is weak or missing entirely (the "missing fundamental" phenomenon common in telephone audio and small speakers).

Once F0 is established, the tool performs an FFT (Fast Fourier Transform) with a window size of 8192 samples at the browser's native sample rate (typically 44,100 or 48,000 Hz). The FFT converts the time-domain audio into a frequency-domain spectrum. The tool then examines the FFT bins at each integer multiple of F0 — specifically at 2×F0 (H2), 3×F0 (H3), 4×F0 (H4), and so on up to 16×F0 (H16). For each harmonic, it uses parabolic interpolation across the three bins nearest the expected harmonic frequency to get a precise amplitude measurement in decibels, compensating for spectral leakage caused by the FFT windowing.

Total Harmonic Distortion (THD) is calculated from these measurements using the standard IEEE definition: THD = sqrt(sum of squared harmonic voltages from H2 through H16) divided by the H1 (fundamental) voltage. Amplitudes are converted from dB back to linear scale for this calculation, then the result is expressed as both a percentage and in dB (20 × log10 of the ratio). The odd/even harmonic classification follows the mathematical definition: H1, H3, H5, H7… are odd (n is odd), while H2, H4, H6, H8… are even. The ratio of total odd-harmonic energy to total even-harmonic energy characterizes the waveform symmetry — a ratio strongly favoring odd harmonics indicates a waveform with half-wave symmetry, such as a square or triangle wave.