Audio7 min readLessonBy · Production Lead at Studio432

EQ and Compression Basics for Video Editors

EQ shapes the tone of a sound by cutting or boosting specific frequencies; compression tames the peaks and evens out the levels — together they are the two most useful audio tools a video editor can learn.

If your dialogue sounds muddy, boxy, or uneven, EQ and compression are almost certainly the fix — but they are also the two tools most editors either skip entirely or use too aggressively. Understanding what each one actually does, rather than just moving sliders until something sounds different, is the difference between audio that feels polished and audio that sounds over-processed. Neither tool requires an engineering background. You need a mental model for each one, a few practical starting points, and a clear sense of the order in which to apply them.

What EQ Does — and What It Is Not

EQ, short for equalization, lets you adjust the volume of specific frequency ranges within a single audio track. Think of it as a tone control that operates on slices of the sound spectrum rather than the whole thing at once. The low end of that spectrum contains bass, rumble, and body. The midrange is where most of a human voice lives — its core tone, warmth, and character. The high end holds clarity, air, and the sharpness of consonants like S and T.

The most important thing to understand about EQ is that it does not add quality to a recording. It reveals what is already there or removes what is getting in the way. If dialogue was recorded in a bad room with a cheap microphone, EQ can make it slightly more listenable, but it cannot make it sound like a professional studio recording. EQ is best used to remove problems — and then, secondarily, to make subtle enhancements. Start by cutting, not boosting.

The High-Pass Filter: Your Most-Used EQ Move

A high-pass filter — sometimes labeled HPF or low-cut on your EQ — rolls off everything below a set frequency, letting the higher frequencies pass through untouched. For dialogue and voice recording, this is the single most practical EQ adjustment you can make, and it costs you almost nothing in audio quality while removing a surprising amount of noise.

Below roughly 80 to 100 Hz, most microphones pick up things you do not want: air conditioning rumble, traffic vibration traveling through floors and furniture, handling noise from someone touching the mic stand, and low-frequency hum from electrical interference. None of these sounds contain useful voice information. Rolling them off with a high-pass filter clears up the low end of your audio, making the voice sound cleaner and more present without actually changing the voice itself.

A typical starting point is to set your high-pass filter somewhere between 80 Hz and 120 Hz for dialogue and voice-over. Some microphones and rooms need it higher; some well-treated environments barely need it at all. Sweep it up slowly from 80 Hz and listen for where the voice starts to sound thinner or loses its natural warmth — that is where to stop.

  • Apply a high-pass filter to almost all dialogue tracks as a default first step
  • Start around 80 to 100 Hz and adjust by ear
  • Stop raising the filter point when the voice begins to lose body or warmth
  • This removes rumble, hum, and handling noise without touching the voice itself

Cutting Mud and Adding Presence

After the high-pass filter, the two most common EQ problems in dialogue are muddiness and a lack of presence. Muddiness is that thick, congested quality where the voice sounds like it is buried in blankets — hard to understand, indistinct, tiring to listen to. It usually lives in the range between about 200 Hz and 400 Hz. A gentle, narrow reduction in that area — pulling it down by 2 to 4 dB — often clears up the voice significantly.

Presence is the quality that makes a voice feel close, clear, and easy to understand. It lives in the range between roughly 2,000 Hz and 5,000 Hz. A small boost in that area — again, 2 to 3 dB is usually enough — can add intelligibility and bring a voice forward in the mix. Be careful here: boosting too much in the upper midrange can tip a voice from clear into harsh and fatiguing very quickly.

A useful approach is to turn on your EQ, boost a frequency range more than you intend to keep, and sweep it slowly across the spectrum. When you hit the problem frequency, you will hear the issue amplified. Then pull that frequency down rather than leaving the boost in place. This technique — boost to find, cut to fix — is a reliable way to use EQ purposefully rather than guessing.

  • Muddiness: try a gentle reduction between 200 Hz and 400 Hz
  • Harshness or boxiness: try a gentle reduction between 800 Hz and 2,000 Hz
  • Presence and clarity: try a small boost between 2,000 Hz and 5,000 Hz
  • Boost to find a problem frequency, then cut — do not leave the boost in

What Compression Does — in Plain Terms

Compression automatically reduces the volume of audio that exceeds a set level threshold. When a speaker's voice gets louder than that threshold, the compressor turns it down by a controlled amount. When the voice drops back below the threshold, the compressor lets it pass through at full volume. The result is a track with less difference between the loudest and quietest moments — a smaller dynamic range — which is easier to mix and easier for an audience to follow.

Think of a compressor as a very fast, very consistent volume knob being turned by someone who never blinks. When you apply light compression to a dialogue track where the speaker shifts between a quiet conversational tone and a sudden emphatic word, the compressor catches that loud word and pulls it back, so you do not have to ride the fader manually throughout the entire edit.

The key parameters on a compressor are threshold, ratio, attack, and release. Threshold sets the level at which the compressor kicks in. Ratio controls how aggressively it reduces audio above that threshold — a 2:1 ratio is gentle, a 4:1 ratio is moderate, anything above 8:1 starts to become limiting rather than compressing. Attack controls how quickly the compressor responds when audio crosses the threshold, and release controls how quickly it lets go afterward. For dialogue, slow-to-medium attack times preserve the natural punch of consonants, while the release should be fast enough to recover cleanly between words.

Gentle Starting Points for Dialogue Compression

Heavy compression is one of the most recognizable signs of amateur audio processing. When the ratio is too high or the threshold too low, voices start to pump and breathe — you can hear the compressor working, which is never the goal. The target is compression that does its job invisibly, evening out the performance without calling attention to itself.

A practical starting point for dialogue: set the threshold so that the compressor is engaging during normal speech but not constantly pinned. A ratio of 2:1 to 3:1 is a safe range for voice. Set the attack at a medium speed — fast enough to catch peaks, slow enough to let the natural transient of each word through. Set the release so it recovers cleanly before the next word arrives. Then adjust the output gain to compensate for the overall level reduction the compressor introduces.

After applying compression, listen back to a passage where the speaker's level varies significantly. The loud moments should feel tamed without sounding squashed. The quiet moments should feel closer in level to the loud ones without sounding artificially boosted. If you can hear the compressor working, back off the ratio or raise the threshold.

  • Ratio: start at 2:1 to 3:1 for dialogue — go higher only if the performance is very uneven
  • Threshold: set so the compressor engages during speech, not just on peaks
  • Attack: medium speed to preserve natural consonant transients
  • Release: fast enough to recover cleanly between words
  • Output gain: turn it back up to compensate for the level reduction

Order of Operations: EQ Before or After Compression?

The order in which you place EQ and compression in your signal chain matters, and the right answer depends on what you are trying to do — but for dialogue in video editing, the most practical default is EQ first, then compression.

Applying EQ before the compressor means you are cleaning up the signal before the compressor analyzes it. If you have a significant low-end rumble, that rumble contains energy that the compressor will react to, potentially causing it to pump and behave unpredictably. Running your high-pass filter first removes that energy so the compressor sees a cleaner signal and responds to the actual voice rather than the noise beneath it.

That said, some editors place a second, lighter EQ after the compressor to make final tonal adjustments once the dynamics have been evened out. This is not mandatory on every project, but it is useful when the compressor has changed the character of the tone in a way you want to gently correct. Think of the first EQ as preparation and cleanup, and any post-compression EQ as finishing and polish.

  • Default order: high-pass filter, then EQ cuts, then compression
  • EQ before compression keeps the compressor reacting to the voice, not background noise
  • A second, light EQ after compression is optional — use it for final tonal polish
  • Do not overthink the order; consistent, gentle processing is more important than perfect chain architecture

When to Stop and How to Check Your Work

The most common mistake with both EQ and compression is doing too much. A voice that has been heavily EQ'd and compressed loses its natural quality — it starts to sound like it has been processed, which undermines trust and listener comfort. The goal in every case is to make the audio sound like a very clean, consistent version of the original recording, not like something that has been visibly manufactured.

A reliable check is to bypass your processing temporarily and listen to the unprocessed audio, then switch it back on. The processed version should feel noticeably cleaner and more even, but the character of the voice should be recognizable as the same sound. If the processed version sounds dramatically different in tone or unnatural, you have likely pushed too far somewhere.

Also check your work on different playback devices before delivering. What sounds balanced on studio headphones can sound harsh on laptop speakers. Running through your dialogue with the EQ and compression applied, at normal listening volume, on the playback device your audience is most likely to use, gives you a much more honest picture of whether the processing is working or working against you.

FAQ
Do I need to EQ and compress every audio track in my edit?

Not necessarily. Dialogue almost always benefits from at least a high-pass filter and light compression. Music tracks from production libraries are usually already professionally processed and rarely need additional EQ or compression from you. Sound effects vary — most are ready to use as delivered. Focus your processing on the dialogue first, then only touch other elements if something specific is not working in the mix.

My dialogue sounds thin and hollow after applying the high-pass filter. What went wrong?

The filter point is probably set too high. If you roll off everything below 200 Hz or higher on a voice, you start removing the natural body and warmth of the speaker's tone. Drop the filter point back down toward 80 to 100 Hz, or lower if the voice sounds thin even there. Some voices with naturally warm, deep tones need the filter set lower than average to avoid sounding hollow.

What is the difference between compression and limiting?

Both compress audio, but limiting is a more extreme version. A compressor with a high ratio — often cited as anything above 10:1 — acts as a limiter, which means it allows almost nothing above the threshold to pass at full volume. Limiters are typically used at the very end of a signal chain as a safety ceiling to prevent clipping rather than as a way to shape the dynamics of a performance. For dialogue processing, stick with lower, gentler ratios. Use a limiter later in your chain only to catch any stray peaks before export.

Can I fix bad audio with EQ and compression?

EQ and compression can make poorly recorded audio more usable, but they cannot fully fix a fundamentally bad recording. If the dialogue was recorded in a highly reverberant room, EQ cannot remove that reverb. If the signal was recorded at a very low level with a lot of noise floor, compressing it will also raise the noise. The best audio comes from a good recording — processing is for refinement, not rescue. Where you do have unavoidably difficult audio, noise reduction tools applied before EQ and compression will generally give you a better starting point.

I applied compression and now the whole track sounds quieter. Is that normal?

Yes, and it is expected. Compression works by reducing the level of audio above the threshold, which lowers the overall perceived loudness of the track. Most compressors include an output gain or makeup gain control specifically to compensate for this reduction. After dialing in your compression settings, bring the output gain back up until the processed track sits at roughly the same loudness as the unprocessed version when you bypass and compare. This lets you judge the compression on its dynamic effect rather than the level difference.

F
Written & reviewed by
Faran@432
Production Lead & Consultant · Studio432

Faran is the production lead at Studio432 — the studio arm of Club432, the Karachi collective behind 100+ filmed live sessions and concert films. He plans and consults on shoots worldwide, and owns the gear, settings and craft standards behind everything published here.

Don’t want to learn it — want it done?

Studio432 edits concert footage, music videos, gaming content, vlogs and short-form for creators worldwide. Send the footage, get a quote by email.

Start a project →