Music Frequencies Explained: See Bass, Mids and Highs Live
Music frequencies run from about 20 Hz to 20 kHz. Producers split that range into seven named bands, from sub-bass to highs, and every instrument occupies a predictable slice of it. You can see all of it live in BeamTracer Visualizers: the Timeline paints a track red, green and blue by band, and the Audio Analysis Engine shows the full spectrum, free, in Chrome or Edge.
Hi, I'm Trina with BeamTracer.com. Let's make frequencies something you can see, not just something you nod along to.
If you DJ, you already read frequencies by eye every night. The colored waveform in Traktor, Serato, Rekordbox or Virtual DJ is a frequency chart: red for low energy, green for mids, blue for highs, and blends in between. A magenta block is kick plus hi-hats, so it is probably the chorus. A thin green stretch is a vocal over nothing, so it is probably the bridge. You learned song structure from color without ever opening a spectrum analyzer.
That is the whole trick of this post. Every frequency concept below comes with a way to confirm it on screen, so the numbers stop being trivia and start being something you recognize on sight.
The Timeline visualizer in EQ COLOR mode. Red is bass, green is mids, blue is highs, and the blends tell you what is playing before you hear it. Open it at /viz?v=timeline or read the Timeline guide.
What is a music frequency?
A frequency is how many times per second a sound wave repeats, measured in hertz (Hz). A kick drum's body pushes air back and forth about 60 times a second; a hi-hat's shimmer does it 8,000 to 16,000 times a second. Low frequencies are heard as bass, high frequencies as brightness, and a full track is thousands of them at once.
One correction to a myth you will read elsewhere: high notes are not "faster" sound. Sound in air travels at about 343 meters per second no matter the pitch. What changes is how often the wave cycles, not how fast it moves. A 20 kHz wave and a 20 Hz wave reach your ear at the same instant; one has simply cycled a thousand times more on the way.
See it: open the Audio Analysis Engine from any visualizer (the A key), watch the Spectrum card, and play a sine sweep or step through a synth's octaves. A single peak walks left to right. That peak is one frequency; the whole display is every frequency the track contains right now.
Why does pitch feel logarithmic?
Doubling a frequency raises the pitch by exactly one octave, so the same musical step covers twice as many hertz each time you go up. A4 is 440 Hz, A5 is 880 Hz, A3 is 220 Hz. Each step feels the same size to your ear even though the first jump is 220 Hz and the second is 440 Hz.
That is why serious spectrum displays use a logarithmic axis. On a linear axis the octave from 40 to 80 Hz gets the same width as the sliver from 19,960 to 20,000 Hz, and everything musical is squashed into the left edge. On a log axis every octave gets equal room.
For the curious, the math for any note:
frequency = 440 × 2^((midiNote − 69) / 12)Middle C (MIDI 60) works out to 261.63 Hz. The lowest key on a piano, A0, is 27.5 Hz; the highest, C8, is 4,186 Hz. Notice that the entire piano keyboard fits below 4.2 kHz. Everything above that in a mix is harmonics, noise and air, not fundamentals.
What are the seven frequency bands?
The seven bands are sub-bass (20–60 Hz), bass (60–250 Hz), low-mids (250–500 Hz), mids (500 Hz–2 kHz), high-mids (2–4 kHz), presence (4–6 kHz) and highs (6–20 kHz). The names are conventions, not physics, and different references draw the lines a few dozen hertz apart. This table is the split most mixing guides agree on.
| Band | Range | What lives here | How it sounds when it is wrong |
|---|---|---|---|
| Sub-bass | 20–60 Hz | 808 tails, sub-synths, the felt-not-heard part of a kick | Too much: rumble that eats headroom. Too little: thin on a club system |
| Bass | 60–250 Hz | Kick body, bass guitar and synth fundamentals, low toms | Too much: boomy. Too little: no weight |
| Low-mids | 250–500 Hz | Body of guitars, keys, snare shell, male vocal warmth | Too much: mud |
| Mids | 500 Hz–2 kHz | Most fundamentals of vocals and melodic instruments; the "heart" of a mix | Too much: boxy, honky |
| High-mids | 2–4 kHz | Attack of picks and beaters, vocal consonants, intelligibility | Too much: harsh, fatiguing |
| Presence | 4–6 kHz | Definition and forwardness; the difference between "in the room" and "behind a curtain" | Too much: brittle |
| Highs / air | 6–20 kHz | Cymbals, breath, reverb tails, sparkle | Too much: sibilance, hiss. Too little: dull |
Your ear is most sensitive between roughly 2 and 5 kHz, which is exactly where the high-mids and presence bands sit. That is not a coincidence; the ear canal resonates there, and it is where speech consonants live. It is also why a small boost in that region sounds like a big one.
See it: the Audio Analysis Engine's Band energies card compares the bands from Sub through Ultra as live meters. Play a track you know and watch which meters jump for the kick, the vocal, the hats. The Frequency bands table in the guide lists the hertz range behind every meter.
Where does each instrument sit in the frequency spectrum?
Every instrument has a fundamental range, the pitched part, and a character range of harmonics well above it. A kick's fundamental is 40–100 Hz but its click is at 2–5 kHz; a hi-hat has almost nothing below 1 kHz at all. That is why a kick and a hat never fight, and why a kick and a bass synth always do.
| Instrument | Fundamental range | Where its character lives |
|---|---|---|
| Kick drum | 40–100 Hz body | Click of the beater at 2–5 kHz |
| Bass guitar / sub | 41 Hz (low E) to ~350 Hz | Finger noise and growl at 700 Hz–2 kHz |
| Snare | 100–250 Hz shell | Crack at 2–5 kHz, wires up to 10 kHz |
| Male vocal (speech) | ~90–155 Hz | Presence at 3–5 kHz, sibilance at 5–8 kHz |
| Female vocal (speech) | ~165–255 Hz | Same presence and sibilance zones |
| Sung voice (bass to soprano) | ~80 Hz to ~1,050 Hz | Formants at 1–4 kHz decide the vowel |
| Guitar (standard tuning) | 82 Hz (low E) to ~330 Hz open strings | Pick attack at 2–4 kHz, fizz up to 10 kHz |
| Piano | 27.5 Hz to 4,186 Hz | Hammer noise above 5 kHz |
| Hi-hats and cymbals | Mostly above 1 kHz | Shimmer at 8–16 kHz |
See it: solo a kick in your DAW, route it to BeamTracer as tab audio, and watch the Spectrum card. You get a fat lump at the bottom and a faint spike near 3 kHz. That spike is the beater. Now solo a hi-hat: nothing below 1 kHz.
Route your DAW without a cable
In any visualizer, open Settings, press CHANGE next to the audio source and pick TAB / SYSTEM AUDIO. Anything playing on the computer, including a soloed channel in your DAW, feeds the analyzer. The audio input guide covers the other sources: microphone, an audio file, internet radio and virtual devices.
How do DJ decks color a waveform by frequency?
EQ-colored waveforms give each band a light color and add them like light: red for lows, green for mids, blue for highs. Where two bands overlap the colors mix, so bass plus mids is yellow, mids plus highs is cyan, bass plus highs is violet, and all three together read as pink or white. BeamTracer's Timeline visualizer does the same thing with an EQ COLOR mode, split at 250 Hz and 4 kHz.
Read it like a DJ does:
- Red or orange stretch: bass and kick with little else. A breakdown, an intro, or the last bars of an outro.
- Green stretch: mids alone. A vocal over a pad, a bridge, a melodic intro.
- Blue burst: highs on their own. A hi-hat build-up, a riser, a snare roll without a kick.
- Pink, magenta or white block: everything at once. The chorus, the drop, the busy part you mix into.
- Big waveform versus small: loud, dense sections stand tall; bridges and intros sit low. Size tells you structure even before color does.
- Open Visualizers, expand CORE ANALYTICS in the left list and choose TIMELINE
- In the settings panel on the right, pick EQ RGB from the STYLE grid, or set COLOR MODE to EQ COLOR yourself
- Adjust BASS (20-250Hz), MID (250-4kHz) and HIGH (4k-20kHz) if you want a different palette; the deck convention is red, green, blue
- Play the demo track, an audio file or an internet-radio station; a file is colored once it loads, a live source is colored as it rolls
- Turn on BAR MARKERS and set SCROLL SPEED to 8 seconds or more to read a whole phrase at a glance
The EQ RGB preset on the demo track: yellow where bass and mids overlap, green where the melody carries the phrase, cyan and blue on the hats.
The other two color modes on the same panel tell you different things. WAVE COLOR is the amplitude heat map audio editors use: the level most of the audio stays under glows hot at the center of each column, and the rare peaks fade cool toward the edges, so density and dynamics jump out. SOLID is the plain envelope. All three are described control by control in the Timeline guide.
What is a harmonic, and why do two instruments on the same note sound different?
A harmonic is energy at an integer multiple of a note's fundamental frequency, and the mix of harmonics is what we hear as timbre. Play A2 (110 Hz) on a bass guitar and on a piano: both have a fundamental at 110 Hz, but each also produces energy at 220, 330, 440 and 550 Hz in different proportions. A pure sine wave has no harmonics, which is why it sounds like a test tone. A distorted guitar is mostly harmonics, which is why it fills a mix.
See it: play a single sustained note and watch the Spectrum card. The harmonics appear as a row of peaks marching to the right, and the spacing tightens as you go up, which is the same effect as octaves feeling equal.
Why does bass disappear when you turn the volume down?
Bass disappears at low volume because human hearing is not flat: at quiet listening levels a 100 Hz tone needs roughly 25 dB more energy than a 1 kHz tone to sound equally loud, and at loud levels the gap shrinks to around 10 dB. These are the equal-loudness contours, first measured by Fletcher and Munson in 1933 and standardized today as ISO 226, last revised in 2023.
The practical consequence: a mix that feels balanced at club level sounds bass-light on a laptop, and a mix that sounds right on a laptop is bass-heavy in the club. A spectrum analyzer does not have this problem. It shows the actual energy, not your perception of it.
This is also why BeamTracer's loudness meter runs the signal through a K-weighting filter before measuring LUFS. K-weighting rolls off the lows and adds a shelf boost above about 2 kHz, so the number tracks what people hear rather than raw electrical level. If you have never used one, How to Read a Loudness Meter covers the five readings.
How does an audio visualizer see frequencies?
An audio visualizer takes a short window of samples, runs a Fast Fourier Transform (FFT), and gets back one magnitude per frequency "bin." Every BeamTracer visualizer reads that same shared analysis, and the two numbers that matter are the window length and the bin width the window produces.
- FFT size is the window length in samples. BeamTracer's shared analyzer defaults to 2048, selectable from 512 to 8192 in the Audio Analysis Engine's Analyser settings card.
- Bin width is the sample rate divided by the FFT size. At 48 kHz with an FFT size of 2048 that is 23.4 Hz per bin. At 8192 it is 5.9 Hz.
So the FFT is a trade. A bigger window resolves frequencies more finely but covers a longer slice of time, so it reacts more slowly. A smaller window snaps to transients but cannot tell 60 Hz from 80 Hz. For a beat-driven visual, 1024 or 2048 is usually the sweet spot. For reading a bass note's exact pitch, go to 8192.
There is one more catch. FFT bins are spaced linearly in hertz. With 23 Hz bins, the whole sub-bass band gets about two bins while the top octave gets over 400. That is why a raw FFT bar display looks like all the action is in the first few bars: it is. A good visualizer groups bins into log-spaced or named bands before it draws anything, which is what BeamTracer's band analyzers do.
See it: change FFT size while a track plays and watch the Spectrum card. At 512 the display is fast and chunky; at 8192 it is detailed and slightly behind the beat.
How do bass, mids and highs drive the visuals?
Most BeamTracer visualizers read three numbers every frame: bass, mid and high energy, split at 250 Hz and 4 kHz. Some also read spectral centroid (the "center of mass" of the spectrum, a proxy for brightness) and spectral flux (how much the spectrum changed since the last frame, the raw material for beat detection). The mapping conventions are almost universal because they match how the body hears.
| Audio feature | What it usually drives | Why it works |
|---|---|---|
| Bass energy (20–250 Hz) | Scale, pulse, camera shake, zoom | Low frequencies are felt as motion |
| Mid energy (250 Hz–4 kHz) | Color, hue shift, shape morph | Melody and vocals carry the emotional content |
| High energy (4–20 kHz) | Particles, sparkle, fine detail, strobe | Highs read as texture and light |
| Spectral flux / onset | Scene cuts, flashes, triggers | A sudden change is a beat |
| Centroid | Warm-to-cool palette, blur amount | Brightness of sound maps to brightness of image |
See it: in the Audio Analysis Engine, click any meter on the Band energies card to start a source-to-control assignment, then pick the visualizer control it should drive. Put Sub on scale and Air on particle count, play a drop, and watch them separate. That is what a VJ does by hand with a MIDI controller, except the frequencies are doing the fader work. The Spectrum Analyzer visualizer is the plainest place to watch the bands move before you assign them.
Same analysis, every visualizer
The Spectrum card, the Band energies meters and every visualizer's audio response read one shared analysis of the same signal. What you see on the meters is exactly what the artwork is reacting to, so a change to FFT size or the processing window changes both at once.
What does a frequency look like on an oscilloscope?
A spectrum shows which frequencies are present; an oscilloscope shows the waveform itself, and it teaches something the spectrum hides: shape. Feed a sine into OSC-1 and you get a smooth curve. Feed a square wave and you get a square: same fundamental, but the sharp corners are the odd harmonics stacked up. A kick drum is a single fat cycle that decays; a hi-hat is a fuzz of tiny cycles with no repeating pattern at all.
X-Y mode goes further. Put one frequency on the left channel and another on the right and the beam traces a Lissajous figure whose shape encodes their ratio. A 1:1 ratio draws a circle or ellipse depending on phase; 2:1 draws a figure-eight; 3:2 draws a bow tie. This is the foundation of oscilloscope music and of everything BT Laser Studio does, since a laser is an X-Y scope drawing with light. The Oscilloscope Music Guide goes deep on it.
See it: the Audio Analysis Engine has an X/Y card beside the Spectrum, left channel on X and right on Y. A mono-compatible mix draws a thin diagonal line; a wide mix fills the screen as a round cloud. Stereo width, which is invisible on a spectrum, is obvious here. For the full instrument, OSC-1's audio sources page explains why its bundled tracks are 96 kHz.
How do I learn frequencies by eye in five exercises?
Each exercise takes under two minutes with the Audio Analysis Engine or the Timeline open, and each one ties a number in this post to something you can see.
- The sweep. Play a 20 Hz to 20 kHz sine sweep. Watch the peak cross the Spectrum card and note where it leaves each band. Around 15–16 kHz most adults stop hearing it while the display keeps showing it. That gap is your hearing, not the track.
- Solo the kick. Loop a kick alone. Sub and Bass light up on the Band energies card; a small blip appears in Presence. Now add the bass line and watch the two fight for the 60–120 Hz region. That is masking, and it is why side-chain compression exists.
- Bright versus dark. Load two masters of similar loudness. The brighter one holds more energy in Treble and Air. If you cannot hear the difference on your monitors, you can still see it.
- Find the mud. Play a busy mix and watch Low Mid. If it sits higher than the mids most of the time, the mix is probably muddy. Cut 250–400 Hz on the guitars or pads and watch the meter drop.
- Read the structure. Switch to the Timeline in EQ COLOR. Choruses are pink and tall; breakdowns turn red or blue and shrink; bridges go green. It is the DJ colored-waveform trick, with the bands labeled in hertz.
How does a frequency view help when mastering for streaming?
Streaming platforms normalize playback to about −14 LUFS integrated (Spotify, YouTube, Tidal, Amazon) or −16 LUFS (Apple Music), so mastering louder gains nothing. What a frequency view adds to the loudness number is where the loudness is coming from.
A track can measure −14 LUFS and still sound weak on a phone if most of its energy is below 100 Hz, because the phone cannot reproduce it and K-weighting counts it lightly anyway. Watch the Band energies card and the LUFS meter together: if Sub and Bass dominate while Presence and Air sit low, you are spending your loudness budget where small speakers cannot hear it. Streaming Loudness Targets has the platform-by-platform table.
Loud is not the same as full
A limiter can push any mix to −14 LUFS. It cannot move energy from the sub band into the presence band. If the phone test fails, the fix is in the spectrum, not the limiter.
How does BeamTracer compare with a DJ deck or a DAW analyzer?
| BeamTracer | DJ software colored waveform | DAW spectrum analyzer plug-in | |
|---|---|---|---|
| Where it runs | A browser tab, Chrome or Edge, nothing installed | Inside the DJ app, with its library | Inside the DAW, on one channel or the master |
| What it shows | EQ-colored Timeline, full spectrum, band meters, X/Y and LUFS from one shared analysis | One EQ-colored overview waveform per deck | The spectrum of one signal at a time |
| Any source | Files, internet radio, microphone, tab or system audio, virtual devices | The library only | The DAW's channels only |
| Drives visuals | Every band is an assignable modulation source for 140 visualizers | No | No |
| Cost to try | Free, no account, in the tab you already have open | Bundled with the DJ software | Bundled or paid plug-in |
Frequently asked questions
Is 20 Hz to 20 kHz the range I can hear? It is the textbook range for a young adult. Most people lose the top few kHz by their thirties and the ceiling keeps dropping. Loud environments accelerate it, which is one reason to wear earplugs at shows.
Which band is "treble"? Consumer EQs usually mean everything above about 4 kHz. In BeamTracer's assignable sources Treble is specifically 4–8 kHz, with Air (8–16 kHz) and Ultra (above 16 kHz) separated out.
What is the difference between a spectrum analyzer and a visualizer? A spectrum analyzer draws the FFT directly so you can read numbers off it. A visualizer feeds the same FFT into artwork. BeamTracer does both from one shared analysis, so the Spectrum card and the visualizer are always looking at identical data.
Do I need to know any of this to use the visualizer? No. Every visualizer ships with sensible band mappings. Knowing it lets you change them on purpose instead of by accident.
Open the Timeline with a track you know, switch it to EQ COLOR, and put a color next to every band in this post. It sticks faster than any chart.
See your own track in red, green and blue. Free, in the browser, no account.
OPEN THE TIMELINE →