Vocal Chains Explained: Compression, EQ, and Effects for Singers

A raw vocal recording almost never sounds the way you want it to in a finished mix. The microphone captures every detail of the performance: the breath noise, the plosive bursts, the sibilant harshness, the dynamic range swings between a quiet verse and a powerful chorus. Processing that raw signal into a polished, present, and controlled vocal is what a vocal chain does.

The order of processing in a vocal chain matters. The same processors applied in a different sequence produce meaningfully different results. This guide explains each stage in the order it typically appears, what it does to the audio, and how to set it up practically.

This assumes your recordings are already as clean as possible at the source. If you are dealing with background noise in your recordings, address that first with our noise removal guide before applying the chain below.

What You Will Learn

• The standard vocal processing chain in order
• What each processor does and what settings to start with
• The difference between corrective and creative processing
• How effects (reverb and delay) fit into the chain
• Common mistakes that make vocals sound amateur

The Standard Vocal Chain: An Overview

A fully processed vocal typically passes through these stages in this order:

1. High-pass filter (corrective EQ)
2. De-esser
3. Compression (first pass, gain riding)
4. Corrective EQ
5. Compression (second pass, character/color)
6. Saturation
7. Creative EQ
8. Reverb and delay (on return tracks)

Not every vocal requires every stage. A simple folk recording might need only a high-pass filter, light compression, and a touch of reverb. A polished pop vocal might use all eight stages plus pitch correction. Start simple and add complexity only where the sound needs it.

Stage 1: High-Pass Filter

The first thing to apply to almost any vocal recording is a high-pass filter, also called a low-cut filter. This removes low-frequency content below the fundamental range of the voice.

Most human voices do not produce meaningful musical content below 80 to 100Hz. What does exist below that range is room rumble, microphone stand vibration, HVAC noise, and handling noise. Removing it cleans up the low end and makes room for the bass and kick drum in the mix.

Practical setting: High-pass filter at 80Hz with a gentle slope (12dB per octave or 18dB per octave). Listen as you raise the cutoff frequency: stop when you notice the vocal losing body or warmth. For a bright, forward vocalist, you might be comfortable cutting at 100Hz. For a deeper baritone voice, 80Hz may be the right place to stop.

Stage 2: De-esser

A de-esser is a frequency-selective compressor that targets the sibilant frequency range, typically between 5kHz and 10kHz, and reduces it when sibilance exceeds a set threshold. This prevents "esses" and "shes" from sounding harsh and distracting in the final mix.

Sibilance becomes more problematic with condenser microphones (which are more sensitive in the high frequencies), in rooms with a lot of high-frequency reflections, and with voices that naturally produce pronounced sibilance.

Practical setting: Set the de-esser's frequency to the specific region where sibilance appears harshest. Use a spectrum analyzer on your voice to find the peak during an "S" sound. Reduce the threshold until the sibilance is controlled but the overall high-frequency presence of the vocal is not noticeably reduced. Over-de-essing creates a lispy, "sshh"-removed quality that sounds unnatural.

Good de-essers include the FabFilter Pro-DS, Waves DeEsser, and built-in de-essers in most DAWs.

Stage 3: First Compression (Gain Riding)

The first compressor in the chain handles the large dynamic swings in the vocal: the jump from a whispered verse line to a belted chorus, or the difference between a close-mic quiet phrase and a pulled-back loud one.

At this stage, you want a compressor with a relatively fast attack, moderate release, and enough ratio and threshold to gain the loud peaks by 6 to 10dB. The goal is evening out the extreme dynamics before the recording hits any further processing.

Practical settings: Ratio 4:1, attack 5 to 15ms, release 50 to 100ms, threshold set so the compressor is engaging on the louder phrases. Aim for 6 to 10dB of gain reduction on loud peaks.

The VCA-style compressors (API 2500, SSL G-Bus, their plugin equivalents) work well here because they respond quickly and transparently. The Waves SSL Channel, FabFilter Pro-C2, and Klanghelm DC8C are strong plugin options.

Stage 4: Corrective EQ

After compression has evened out the dynamics, you can hear the true tonal character of the vocal more clearly. Corrective EQ addresses frequency problems: a nasal buildup in the 300 to 500Hz range, a boomy low-mid resonance, harsh energy in the 2 to 4kHz range, or a dull lack of clarity above 10kHz.

This is subtractive EQ work: finding and reducing problematic frequencies rather than boosting desirable ones.

Method: Use a parametric EQ with a narrow bandwidth. Boost a narrow band by 6 to 8dB and sweep it slowly through the frequency range while the vocal plays. When you find a frequency that sounds objectionable when boosted, you have found a problem area. Switch the boost to a cut and reduce that frequency by 3 to 6dB with a moderate bandwidth.

Common corrective cuts on vocals:

• 250 to 400Hz: muddy buildup, reduced by 2 to 4dB
• 800Hz to 1kHz: boxy, telephone quality, reduced by 2 to 3dB
• 2 to 4kHz: harsh presence, reduced by 2 to 4dB

The FabFilter Pro-Q 3 is the standard tool for this work due to its dynamic EQ capabilities and spectrum analyzer display.

Stage 5: Second Compression (Character and Color)

The second compressor adds density, warmth, and cohesion to the vocal after corrective EQ has cleaned up the tone. Where the first compressor was transparent and functional, this one can have character.

Tube-style and optical compressors work well here: the LA-2A, Fairchild 670, and their emulations. These compress in a musically natural way that "glues" the vocal without making it sound processed.

Practical settings: Lower ratio (2:1 to 3:1), slower attack (30 to 50ms), program-dependent or slow release. You are adding 2 to 4dB of gentle compression that makes the vocal feel consistent and present.

Stage 6: Saturation

Saturation adds harmonic distortion to the vocal signal. Subtle saturation adds warmth, presence, and that slightly "pushed" quality that makes vocals sound like they were recorded through high-quality analog hardware. It also adds harmonics that help the vocal sit audibly in a dense mix without raising the actual level.

Many producers use a dedicated tape saturation plugin (iZotope Tape in Ozone, Waves J37, Softube Tape) or a console emulation plugin (SSL, API, Neve emulations). A few percent of drive goes a long way on a vocal. More than a few percent starts to sound like intentional distortion.

Practical setting: Drive the saturation plugin just to the point where you can hear the character change in A/B comparison. If you cannot hear the difference, add slightly more. If you can hear it clearly, pull back until it is subtle.

Stage 7: Creative EQ

After all corrective processing is complete, apply additive (boosting) EQ for character and presence. This is where you shape the vocal's personality in the mix rather than fix its problems.

Common creative EQ boosts on vocals:

• Gentle wide boost at 8 to 12kHz: air, shimmer, and detail
• Moderate boost at 2 to 5kHz (presence region): cut-through in a dense mix
• Wide low-mid boost at 200 to 300Hz: warmth and body

Use a high-quality EQ for creative boosts. Analog-modeled EQs (Neve 1073, Pultec EQP-1A, SSL 4000 E) and their emulations produce pleasing boosts that "breathe" rather than sounding surgical.

Stage 8: Reverb and Delay (Return Tracks)

Reverb and delay should be applied as send effects on return tracks rather than inserted directly on the vocal channel. This lets you control the blend independently of the dry signal and apply different rooms to different vocal elements.

Short plate reverb: A short plate reverb (0.8 to 1.5 seconds decay) adds space and warmth without pushing the vocal back in the mix. This is the most common vocal reverb in pop and R&B production.

Room reverb: A small room reverb (0.3 to 0.6 seconds) adds a sense of acoustic realism, as if the vocal was recorded in a real space rather than a padded booth.

Slapback delay: A single repeat at 60 to 120ms with no feedback creates a vintage, rockabilly-influenced quality and adds width when panned opposite to the dry vocal.

Long modulated delay: A longer delay (200 to 400ms) with one or two repeats and light feedback fills space in ambient and electronic productions.

Keep reverb and delay blended subtly in the context of the full mix. Soloceing the reverb return to set it usually leads to over-wet results that wash out the vocal in the mix.

Pitch Correction: Where It Fits

Pitch correction (Auto-Tune, Melodyne) is not strictly part of the signal processing chain because it edits the audio content rather than processing it. It is typically applied as the first step, before any of the above processors, so that the corrected pitch is processed rather than the pitch error.

For a full comparison of pitch correction tools, see our Auto-Tune vs Melodyne guide.

Common Mistakes in Vocal Processing

Over-compressing. Too much compression reduces dynamics to the point where the performance sounds lifeless and "squashed." Dynamic range is part of the expressiveness of a vocal performance. Compress enough to control the dynamics, not to eliminate them.

Boosting before subtracting. Always make corrective cuts first, then creative boosts. Adding more energy to a frequency range that already has a problem makes the problem worse.

Setting reverb in solo. The appropriate reverb level for a vocal is determined in the context of the full mix. What sounds lush and spacious in solo will often sound washed out and distant once the mix is playing.

Applying identical settings to every vocal. Every voice and every microphone combination behaves differently. Use the chain above as a starting point, not as a recipe. Trust your ears over your presets.

Frequently Asked Questions

Q: Do I need all eight stages for a simple demo?

A: No. A high-pass filter, one compressor, and a small amount of reverb will handle most demo vocals. Add stages progressively as the need arises.

Q: Should I use a gate or an expander before the de-esser?

A: A noise gate can be useful before the de-esser to reduce steady-state noise floor during pauses. Set it with a slow attack so it opens before the first transient of each phrase.

Q: How do I know if my vocal chain is too heavy?

A: Compare your processed vocal to a reference recording in the same genre. If your vocal sounds more processed, more squashed, or less natural than the reference, reduce processing at each stage until the result sounds competitive rather than over-worked.

Q: Can I use the same vocal chain on background vocals?

A: Background vocals typically receive less processing than lead vocals: less presence boost, more reverb, and sometimes more compression for a more "glued" group sound. Start with a lighter version of your lead vocal chain.

What to Do Next

With a finished vocal sound, the next step is placing it correctly in the mix alongside the rest of your production. Our mixing vs mastering guide explains how the vocal fits into the broader mix process. For producers who want to record more than just vocals, our guide to recording acoustic guitar at home covers miking technique, room positioning, and processing in the same practical depth as this guide.