High-pass and low-pass filters are common tools used to remove unwanted sounds from audio recordings. They have a long history in use for audio postproduction due to their simplicity and ease of use. In this post, we’ll discuss what high-pass and low-pass filters are, as well as how and when to use them.
High and Low-Pass Filters
Technically speaking, a high-pass filter (sometimes abbreviated “HPF”) passes signals with a frequency higher than the designated cutoff frequency and attenuates signals with frequencies lower than the cutoff frequency. This means that low frequencies are strongly reduced while high frequencies are allowed to pass through unaffected. The opposite is true for low-pass filters (sometimes abbreviated “LPF”), which passes signals with a frequency lower than the designated cutoff frequency and attenuates signals with frequencies higher than the cutoff frequency. Simply put, high frequencies are strongly reduced while low frequencies are allowed to pass through unaffected.
High and low-pass filters are found in nearly all mixing consoles and digital audio workstations, with some offering a fixed cutoff frequency and others offering a “sweepable”, or user-selected cutoff frequency. The cutoff frequency is the point in which the filter begins its attenuation, indicating which frequencies get attenuated and which frequencies pass through unaffected.
In DAWs, these filters typically come with a variety of slopes, or rates of attenuation below the cutoff frequency, typically quantified in decibels per octave (note: an octave is a relative doubling or halving of the frequency). A 6dB per octave high-pass filter set to 200Hz would result in 6dB of relative attenuation at 100Hz and 12dB at 50Hz. This would continue into very low frequencies, attenuating the signal to an indiscernible level. Unlike some effects, high and low-pass filters tend to work predictably and consistently across mixing consoles, DAWs, and plugins given that the cutoff frequency and slope are the same.
High-pass filters are incredibly useful, especially for vocal audio recording, and work wonders for other content as well. High-pass filters remove low-frequency noise that’s difficult to reproduce on headphones and small speakers. Even when it’s inaudible, low-frequency noise occupies space in the recording, reducing the ability to record and playback audio at a louder level. More importantly, this noise is difficult to monitor on headphones and speakers, but might be audible on full range speaker systems. This means that you may be recording and transmitting information that your audience can hear, but that you can’t. Listen to local news broadcasters on a home theater system and the subwoofer will pop incessantly with “plosives” that result from addressing the microphone too closely or wind blowing into the microphone. This type of noise can be terribly off-putting to an audience playing audio through their home entertainment system or listening in a car, and can be virtually eliminated by using a high-pass filter.
For music recording, high-pass filters are incredibly practical, especially when recording multiple instruments together. Bass-heavy instruments and sounds work better without interference from other bass-heavy sounds. For example, a bass guitar and acoustic guitar typically sound better together with the low-frequency content removed from the acoustic guitar mic(s). Consider using high-pass filters on all instruments except for bass guitar, contrabass, floor toms, kick drums, bass synthesizers, organ, or anything else with strong, low-frequency content.
For all recorded vocals, use a high-pass filter set to 90Hz, and select the steepest slope available, up to 64 decibels per octave (dB/Oct). This helps to substantially reduce wind noise, vocal pops and “plosives”, yet doesn’t interfere with the tone of the human voice. Even if you can’t hear it, low-frequency content may exist in the recording and become apparent to a listener with a larger playback system. The only exception to this rule applies to voices with extreme low-frequency energy, like a voiceover artist with a deep voice or bass or baritone singer, in which case you may the filter to 60Hz.
High-pass filter set to 100 Hz and slope set to 64 decibels per octave (dB/Oct).
Unlike high-pass filters, low-pass filters offer little benefit for much other than music mixing and creative applications. Vocal recordings and most instruments contain defining characteristics at high frequencies, and sound better with these frequencies either boosted or left alone. Consonants and breaths contain a large degree of high-frequency information, and spoken words, singing, and dialogue lose clarity when high-frequencies are removed. High-frequencies also provide a very subtle, pristine quality that makes a recording sound more polished and professional. Removing high-frequencies often results in an inferior sound for vocals, so we recommend avoiding the use of low-pass filters unless you’re skilled with EQ or plan to use it for a creative effect. For example, telephone or walkie-talkie “low-fi” effects are usually created using aggressive low-pass filtering in combination with mild distortion.
For music recording, low-pass filters rarely offer little benefit except for detailed mixing. Even bass-heavy instruments like electric basses, kick drums, and upright basses contain large amounts of high-frequency information. Low-pass filters can effectively reduce high-frequency noise, but it often comes at the expense of important defining characteristics of the instrument, such as the clicking sound of a bass string. We recommend avoiding using low-pass filters unless you’re skilled with EQ or plan to use it for advanced mixing or for creative effect.
Both high and low-pass filters are common equalization tools found in mixing consoles and DAWs, but high-pass filters are far more practical for music and voice recording applications. High-pass filters remove low-frequency noise that may go unheard by the person recording but that may be distracting to anyone listening in their car stereo or large speaker system. As a rule of thumb, use a high-pass filter set to 90Hz, and select the steepest slope available, up to 64 decibels per octave (dB/Oct) for vocal recordings. This helps to substantially reduce wind noise, vocal pops and plosives, yet doesn’t interfere with the tone of the human voice
We hope this helps improve your recordings using these simple, yet highly effective tools. If you’ve recently tried EQing with high and low-pass filters, share your thoughts in the comments below.