Audio Processing for SSB: Compression, EQ, and Getting Loud Without Splattering

You've got 100 watts. The station you're trying to work has 100 watts. So why does he sound like a kilowatt and you sound like you're whispering into a tin can from across the room?

The answer is audio processing. Done right, it's the closest thing to free power you'll ever get in ham radio. Done wrong, you become that guy everyone complains about on 14.300.

This guide covers how to set up compression, equalization, and transmit audio shaping for SSB that actually works. We're talking real-world settings, not theory-only explanations that leave you more confused than when you started.

Why Audio Processing Matters on SSB

SSB (Single Sideband) transmits your voice as a varying-power signal. Speak softly, and your output drops to maybe 10-15 watts on a 100W rig. Hit a loud syllable, and you spike to full power. The average power over a typical sentence? Somewhere around 20-30 watts if you're not processing anything.

That's where speech processing comes in. The goal is simple: raise the average power of your transmitted signal without increasing the peak power. More average power means more perceived loudness at the receiving end.

A 6 dB increase in average power is roughly equivalent to quadrupling your transmitter output. That's going from 100 watts to 400 watts, just by shaping your audio. Not magic. Physics.

Compression: The Foundation

Most modern transceivers have a built-in speech processor or compressor. The Icom IC-7300 calls it "COMP," Yaesu rigs label it "PROC" or "Speech Processor," and Kenwood uses "PROC" as well. They all do roughly the same thing.

How It Works

A compressor reduces the dynamic range of your voice. It turns down the loud parts and brings up the quiet parts, resulting in a more consistent signal level feeding your transmitter's final amplifier.

Think of it like an automatic hand constantly adjusting your mic gain. Loud syllable comes in? It pulls back. Quiet consonant? It pushes forward.

Setting Compression Level

Here's where most people screw up. They crank the compression to maximum because more is better, right?

Nope.

I ran this exact experiment on my IC-7610 last winter. Set compression to maximum (20 dB on that rig), transmitted on a dummy load, and asked three locals to compare. Every single one said it sounded worse than no compression at all. Distorted, harsh, fatiguing to listen to.

The sweet spot for most rigs:

• Light compression: 4-6 dB of gain reduction. Noticeable improvement, zero distortion.
• Moderate compression: 8-10 dB. Good punch for contest or DX pileup work. Some coloration.
• Heavy compression: 12+ dB. Contest-only. Sounds aggressive. Not for ragchewing.

Start at 6 dB and work up. Ask for honest reports. Not "How's my audio?" but "Does my audio sound natural or processed?"

Attack and Release Times

If your rig (or external processor) lets you adjust attack and release:

• Attack: 2-5 ms. Fast enough to catch transients before they clip.
• Release: 100-250 ms. Slow enough to avoid "pumping" artifacts between words.

Most built-in processors don't expose these settings, and honestly? Their defaults are fine for 90% of operators.

Equalization: Shaping What Gets Transmitted

SSB on HF has a defined audio bandwidth, typically 300 Hz to 2700 Hz for most transceivers. Some open up to 2900 Hz or even 3100 Hz in wider filter settings. Everything outside that range gets chopped off regardless of your EQ.

So we're working within a narrow window. The question is: how do you distribute energy within that window for maximum intelligibility?

The Speech Intelligibility Curve

Human speech intelligibility (the ability to understand words) lives primarily between 1000 Hz and 3000 Hz. The consonants that differentiate words — the "s," "t," "f," "k" sounds — sit in the 1500-2500 Hz range.

The low frequencies (300-800 Hz) carry warmth and body but very little actual information. On a noisy HF band with QRM and QRN, that low-end energy is just competing with the noise floor.

Practical EQ Settings

For DX and contest work, I've settled on this after years of tweaking (on an IC-7610 with the built-in equalizer):

• 300-500 Hz: -3 to -6 dB (cut the mud)
• 500-800 Hz: flat or -2 dB
• 800-1200 Hz: flat
• 1200-2000 Hz: +2 to +4 dB (presence boost)
• 2000-2700 Hz: +1 to +3 dB (clarity/bite)

For ragchewing where you want a fuller, more pleasant sound:

• Keep everything flatter. Maybe a gentle +2 dB bump at 1500 Hz.
• Don't cut the lows as aggressively.

I'll get hate for this, but the "broadcast quality" audio crowd on 75 meters is optimizing for the wrong thing. If someone needs noise-canceling headphones to tell your audio from background hiss, you're winning. If they think your audio is "beautiful" but can't copy you through moderate QRM, you've lost the plot.

Mic Choice Matters More Than EQ

A Heil PR-781 or Heil GM-5 into a rig with zero EQ will sound better than a stock hand mic with perfect EQ applied. The mic's native frequency response is your starting point.

Dynamic mics with a presence peak around 2 kHz (like the Heil elements) are purpose-built for communications intelligibility. That's not marketing fluff — it's measurable frequency response data.

If you're using the stock mic that came in the box, EQ becomes more important because you're compensating for a mic designed for general purpose, not communications.

The ALC Trap

Here's the thing most guides skip.

Your rig's ALC (Automatic Level Control) meter is telling you something. If you're seeing ALC deflection during normal speech after setting up compression and EQ, you're overdriving the input. The ALC is clamping your signal, and that clamp is happening after your carefully crafted audio processing.

ALC action on voice peaks means distortion. Period.

Set your mic gain so that ALC barely flickers on the loudest syllables. If you've added compression and your ALC is now slamming into the red, turn down the mic gain to compensate. The compression is doing the work now — you don't need the same mic gain level you used before.

I spent six months running too hot because nobody told me this. Got reports ranging from "sounds great" to "you're splattering" depending on who was listening and where they were in frequency relative to my signal. (Turns out the guy saying "sounds great" was 5 kHz away and the splatter victims were right next door.)

External vs. Internal Processing

Modern rigs like the IC-7300, IC-7610, Yaesu FTDX101D, and Kenwood TS-890S have onboard EQ and compression that's genuinely good. Ten years ago, you needed an external box. Now?

For most operators, the built-in tools are sufficient. Use them first.

External processors make sense if:

• You want more precise control over attack/release/ratio
• You're running a vintage rig without built-in DSP
• You're doing contest operating where every fraction of a dB matters
• You want to chain multiple stages (EQ → compressor → limiter)

Popular external options: Symetrix 528E (discontinued but found used), W2IHY EQplus, Behringer Ultragain Pro MIC2200 (budget, works fine), or a USB audio interface running software processing into a rig's USB/LINE input.

Monitor Your Own Audio

You can't tune what you can't hear. Use your rig's monitor function to listen to your transmitted audio through headphones. Most modern rigs support this.

On Icom: Menu → MONITOR → ON, adjust MON level. On Yaesu: MONI button or menu setting. On Kenwood: TX MON function.

This gives you the processed audio before it hits the final PA. It's not identical to what goes over the air (it's pre-filtering on some rigs), but it's close enough to catch obvious problems.

The Quick Setup Checklist

1. Start with compression OFF, EQ flat
2. Set mic gain so ALC just barely shows on loud voice peaks
3. Turn compression ON, set to 6 dB
4. Reduce mic gain until ALC behavior returns to step 2 levels
5. Apply EQ — cut below 500 Hz by 3 dB, boost 1.2-2.0 kHz by 3 dB
6. Get an on-air report from someone you trust (not your buddy who always says "sounds good")
7. Adjust in small increments based on feedback

Do not change three variables at once. One knob at a time.

Wrapping Up

Good audio processing is the cheapest upgrade in ham radio. A $30 dynamic mic element and 20 minutes of careful adjustment gives you the equivalent of doubling or tripling your power on SSB.

But the keyword is careful. Crank everything to max and you're just the loudest nuisance on the band. Take the time to dial it in, get honest reports, and resist the urge to keep pushing for "more."

Your fellow hams' ears will thank you.