Audio Filter

In-Depth Explanation

Audio filter is a signal processor that attenuates or boosts specific frequency ranges of an audio signal. Common types include low-pass (removes highs), high-pass (removes lows), band-pass (keeps a mid range), and notch (cuts a narrow range). Filters shape tonal character by removing or boosting selected frequencies. They are fundamental to both mixing and synthesizer sound design.

How Audio Filter Works

A filter is a frequency-dependent processor. It passes some frequencies unchanged while reducing the level of others. The point where the filter begins to take effect is called the cutoff frequency. The rate at which the filter reduces frequencies beyond the cutoff is called the slope, measured in dB/oct. A 12 dB/oct slope reduces gradually. A 24 dB/oct slope reduces steeply. A 48 dB/oct slope is nearly a brick wall.

Low-Pass Filter

The most common filter type. It passes frequencies below the cutoff and attenuates frequencies above it. A low-pass filter on a synth at 800 Hz removes all harmonics above 800 Hz, leaving a warm, dark sound. This is the primary filter on most analog synthesizers, including the Moog Minimoog and the Roland Juno.

High-Pass Filter

The opposite of low-pass. It passes frequencies above the cutoff and attenuates frequencies below it. High-pass filters are used on every mixing session to remove rumble and low-frequency mud from tracks that do not need bass. A high-pass filter at 100 Hz on a vocal removes air conditioning noise and mic handling noise without affecting the voice.

Band-Pass Filter

Passes a band of frequencies around a center frequency while attenuating everything above and below. Band-pass filters are used for telephone effects (centering around 1 to 3 kHz) and isolating frequency ranges in sound design. A band-pass filter on a drum loop at 200 Hz creates a lo-fi, AM-radio texture.

Notch Filter

Cuts a narrow range of frequencies while passing everything else. Notch filters remove problem frequencies without affecting the surrounding sound. A notch filter at 3 kHz with a narrow bandwidth removes a harsh resonance from a vocal without dulling brightness. Notch filters also eliminate feedback in live sound and remove hum at 50 or 60 Hz.

Resonance (Q)

Most filters have a resonance (or Q) control. Resonance boosts frequencies at the cutoff point before attenuating everything beyond it. Low resonance (0.1 to 0.7) produces a gentle filter curve. High resonance (5 to 20) produces a sharp, whistling peak at the cutoff. Extreme resonance on a synth low-pass filter creates the classic acid house squelch of the Roland TB-303.

Real-World Example

An independent producer records a vocal in an untreated bedroom. The recording has low-frequency rumble from a nearby air conditioner at 60 Hz. There is also a harsh resonance at 2.5 kHz that makes the vocal unpleasant on earbuds.

The producer inserts an EQ plugin (the free TDR Nova) on the vocal. They add a high-pass filter at 80 Hz with a 24 dB/oct slope. The 60 Hz rumble drops by approximately 20 dB, becoming inaudible. The vocal's lowest frequencies are unaffected.

They add a notch filter at 2.5 kHz with a Q of 6 and -4 dB of cut. The harsh resonance disappears. The vocal sounds smoother without losing clarity. The producer spent $0 and fixed both problems in 3 minutes.

On a synthesizer track, the producer uses a low-pass filter at 1.2 kHz with resonance at 60%. They modulate the cutoff with an ADSR envelope (fast attack, 400 ms decay). Each note starts bright, then darkens as the envelope decays. This creates the classic plucky synth sound used in house and techno.

Why It Matters for Independent Artists

Filters are the most used processors in music production. Every mix uses high-pass filters on nearly every track. Every synth patch uses a filter to shape harmonic content. Understanding filter types and slopes is non-negotiable.

Three rules:

1. High-pass everything that does not need bass. Apply a high-pass filter at 80 to 120 Hz on vocals, guitars, and synths that are not providing bass content. This removes rumble and creates headroom for the kick and bass. On a 30-track mix, this technique frees up 3 to 6 dB of headroom on the master bus. Cost: $0 with any free EQ plugin.

2. Choose the right slope. Steep slopes (24 to 48 dB/oct) remove unwanted frequencies decisively but can sound unnatural. Gentle slopes (6 to 12 dB/oct) sound musical but let more unwanted frequencies through. Use steep slopes for cleanup. Use gentle slopes for tonal shaping.

3. Use resonance intentionally. On synthesizers, resonance adds character and movement. On mixing EQ, high resonance (narrow Q) is useful for surgical cuts but sounds unnatural for broad tonal changes. For wide tonal shaping, use low Q values (0.5 to 1.5). For surgical fixes, use high Q values (4 to 10).

The Fletcher-Munson curve explains why filter decisions matter at different volumes. A mix that sounds bright at 85 dB may sound harsh at 35 dB because the ear is most sensitive to 1 to 5 kHz. Use filters to control the 2 to 5 kHz range so your mix translates across levels. Read our music production fundamentals guide for more on filter techniques. Our mixing vs. mastering guide covers how filters affect the final master.

Related Terms

• EQ (Equalization) - The mixing tool that uses multiple filter types to shape frequency balance
• Synthesizer - The instrument where filters are a primary sound design tool
• ADSR (Attack, Decay, Sustain, Release) - The envelope that often modulates a synthesizer's filter cutoff
• Fletcher-Munson Curve - The psychoacoustic principle that explains why filter decisions matter at different volumes