Podcast Compression Best Practices: Control Dynamics for Consistent Audio
TL;DR: Compression reduces the gap between quiet and loud moments, creating consistent volume that listeners don't need to adjust. For podcasts, use 3:1 to 4:1 ratio, threshold around -18dB, attack 10-30ms, release 100-200ms. Apply moderate compression to each speaker, then gentle bus compression to glue everything together. Over-compression sounds lifeless; under-compression causes volume swings.
Table of Contents
- What Compression Does
- Compression Parameters Explained
- Podcast Compression Settings
- Compression Workflow for Podcasts
- Common Compression Mistakes
- Advanced Compression Techniques
- FAQ
What Compression Does
Compression is automatic volume control. It turns down loud moments, allowing you to raise overall volume without clipping.
Here's the thing: without compression, you're stuck choosing between volume that's comfortable for quiet sections (but blasting during loud moments) or comfortable for loud sections (but whisper-quiet during softer moments). Compression lets you have both.
The Dynamic Range Problem
Dynamic range = the gap between quietest and loudest moments.
Speech naturally has wide dynamic range:
- Quiet words spoken softly
- Loud emphasis and exclamations
- Volume changes as speakers lean toward or away from microphones
- Different speakers at different natural volumes
Listeners adjust their volume based on average levels. Wide dynamic range means quiet parts disappear into background noise while loud parts become uncomfortable.
What Compression Achieves
Reduced dynamic range: Quiet and loud moments closer together Increased average loudness: Can raise overall level without clipping Consistent listening experience: No volume adjustments needed Glue and cohesion: Multiple speakers sound like they belong together
The Trade-Off
Compression reduces natural expression. Heavy compression makes speech sound flat and lifeless. The goal is finding the balance: enough compression for consistency, not so much that voices lose character.
Compression Parameters Explained
Understanding each control helps you dial in appropriate settings.
Threshold
What it does: Sets the level above which compression begins.
How it works: Audio below threshold passes unchanged. Audio above threshold gets compressed.
Podcast range: -18dB to -12dB typically
Lower threshold: More compression (more audio exceeds threshold) Higher threshold: Less compression (only loudest moments affected)
Ratio
What it does: Determines how much loud signals are reduced.
How it works: A 4:1 ratio means for every 4dB above threshold, only 1dB comes out.
| Ratio | Effect | Use Case |
|---|---|---|
| 2:1 | Gentle | Light smoothing |
| 3:1 | Moderate | Standard podcast |
| 4:1 | Medium | Dynamic speakers |
| 8:1+ | Heavy | Limiting effect |
Podcast range: 3:1 to 4:1 for most applications
Attack
What it does: How quickly compression begins after signal exceeds threshold.
Fast attack (1-10ms): Catches transients quickly, can sound aggressive Slow attack (30-100ms): Lets transients through, more natural sound
Podcast range: 10-30ms balances control with naturalness
Release
What it does: How quickly compression stops after signal drops below threshold.
Fast release (20-50ms): Quick recovery, can cause pumping Slow release (200-500ms): Smooth recovery, can reduce average level
Podcast range: 100-200ms provides smooth recovery without audible artifacts
Knee
What it does: How gradually compression transitions from uncompressed to compressed.
Hard knee: Abrupt transition at threshold Soft knee: Gradual transition around threshold
Podcast recommendation: Soft knee sounds more natural for speech
Makeup Gain
What it does: Restores volume lost to compression.
Compression turns down loud moments, reducing overall level. Makeup gain compensates.
Setting makeup gain: Aim for similar perceived loudness with compression on and off.
Podcast Compression Settings
Starting settings for common podcast scenarios.
Single Speaker (Solo Podcast)
Goal: Consistent volume without obvious processing
| Parameter | Setting |
|---|---|
| Threshold | -18dB |
| Ratio | 3:1 |
| Attack | 15ms |
| Release | 150ms |
| Knee | Soft |
Gain reduction target: 3-6dB on average, occasional peaks hitting 8-10dB
Two-Person Conversation
Goal: Both speakers sound equally present
Apply to each speaker individually:
| Parameter | Setting |
|---|---|
| Threshold | -16dB |
| Ratio | 3:1 to 4:1 |
| Attack | 20ms |
| Release | 150ms |
Then add light bus compression:
| Parameter | Setting |
|---|---|
| Threshold | -12dB |
| Ratio | 2:1 |
| Attack | 30ms |
| Release | 200ms |
Dynamic Speaker (Expressive Delivery)
Goal: Control wide volume swings while preserving expression
| Parameter | Setting |
|---|---|
| Threshold | -20dB |
| Ratio | 4:1 |
| Attack | 10ms |
| Release | 100ms |
More aggressive settings handle greater dynamic variation.
Quiet Speaker (Low Recording Level)
Goal: Bring up level without amplifying noise
| Parameter | Setting |
|---|---|
| Threshold | -24dB |
| Ratio | 3:1 |
| Attack | 20ms |
| Release | 200ms |
| Makeup gain | Higher than usual |
Important: Address noise before compressing quiet recordings—compression amplifies noise proportionally.
Compression Workflow for Podcasts
Where compression fits in your editing process.
Processing Order
Compression typically comes after EQ in the signal chain. Maintaining this order consistently is key to an effective editing workflow:
- Noise reduction (if needed)
- EQ (shape tone before dynamics processing)
- Compression (control dynamics)
- De-essing (if needed)
- Limiting (catch remaining peaks)
Per-Speaker vs. Bus Compression
Per-speaker compression:
- Applied to individual tracks
- Addresses each speaker's unique dynamic characteristics
- Usually the primary compression stage
Bus compression:
- Applied to combined tracks
- Glues multiple speakers together
- Usually lighter than per-speaker compression
Recommended approach: Moderate per-speaker compression + light bus compression
Gain Staging Through Compression
Maintain proper levels throughout:
Before compression: Peaks around -12dB to -6dB After compression: Similar peak levels (use makeup gain) After bus compression: Peaks around -6dB to -3dB After limiting: Peaks at -1dB to -1.5dB
A/B Testing Your Compression
Always compare compressed audio to the original.
Testing method:
- Match loudness (bypass compression, adjust to same perceived level)
- Toggle compression on/off
- Listen for improvement in consistency
- Listen for artifacts or unnatural sound
If the original sounds more natural without major volume problems, reduce compression.
Common Compression Mistakes
Avoid these patterns that undermine your results.
Over-Compression
The mistake: Too much gain reduction, creating lifeless, flat sound.
Symptoms:
- Voice sounds distant or "squashed"
- Natural emphasis disappears
- Words run together
- Pumping or breathing artifacts
- Gain reduction meter constantly pinned
Fix: Raise threshold, reduce ratio, or reduce makeup gain.
Under-Compression
The mistake: Settings too gentle to address dynamic problems.
Symptoms:
- Volume still swings noticeably
- Quiet moments hard to hear
- Loud moments uncomfortably loud
- Gain reduction barely visible
Fix: Lower threshold or increase ratio.
Wrong Attack Time
Too fast: Transients get squashed, words sound mushy or lose punch. Too slow: Loud moments punch through before compression catches them.
Fix: Adjust attack while watching gain reduction meter. You should see it respond to words without clamping down on every syllable.
Wrong Release Time
Too fast: Pumping sound—volume rises and falls audibly between words. Too slow: Compression doesn't recover between phrases, reducing overall level.
Fix: Listen for natural pauses. Volume should recover smoothly without audible breathing.
Compressing Without Addressing Source Problems
The mistake: Using compression to "fix" fundamental recording issues.
Compression can't fix:
- Clipped audio (distortion)
- Severe background noise (gets amplified)
- Bad microphone technique (inconsistent distance)
- Room reverb problems
Fix: Address source problems before compression. Use compression to enhance good recordings, not rescue bad ones.
Advanced Compression Techniques
Beyond basic compression for specific situations.
Parallel Compression
Blend compressed and uncompressed signals for aggressive compression that retains dynamics.
How it works:
- Duplicate the track
- Compress the duplicate heavily (6:1 ratio, low threshold)
- Mix the compressed track in under the original
- Adjust blend to taste
Result: Transients and dynamics from original + body and consistency from compressed
Multiband Compression
Different compression settings for different frequency ranges.
When to use:
- Plosives or low-frequency problems
- Sibilance control (alternative to de-essing)
- Different dynamic issues in different frequency ranges
Podcast application:
- Low band: Control plosives and rumble
- Mid band: Standard voice compression
- High band: Tame harsh transients or sibilance
Sidechain High-Pass
Many compressors can filter the detection signal (sidechain).
Problem: Low frequencies (plosives, mic handling) trigger compression, affecting the whole voice.
Solution: High-pass filter on the sidechain so the compressor ignores low-frequency energy.
Typical setting: 80-150Hz high-pass on sidechain
Serial Compression
Two or more compressors in series, each doing light work.
Benefits:
- More transparent than one compressor working hard
- Different compressor characters can complement each other
- Easier to control overall effect
Typical setup:
- First compressor: Fast attack, catches peaks (2:1 ratio)
- Second compressor: Slower attack, shapes overall dynamics (3:1 ratio)
FAQ
How much gain reduction should I see on podcast vocals?
For typical podcast speech, aim for 3-6dB of gain reduction on average, with occasional peaks hitting 8-10dB on louder moments. If you're seeing 10-15dB constant reduction, your threshold is too low or ratio too high. If you're barely seeing 1-2dB, compression may not be doing much useful work.
Should I compress before or after EQ?
Generally compress after EQ. EQ changes can significantly alter dynamics—a presence boost might create new peaks that need compression. However, some engineers prefer EQ after compression to shape the compressed tone. Try both approaches and use what sounds better for your specific material.
What's the difference between compression and limiting?
Compression uses ratios typically 2:1 to 10:1 with gradual gain reduction. Limiting uses very high ratios (10:1 to infinity:1) with fast attack to create a hard ceiling that audio can't exceed. Compression shapes dynamics; limiting prevents peaks from exceeding a maximum level. Podcasts typically use both—compression for shaping, limiting for peak control.
Can compression make my podcast louder?
Yes, but indirectly. Compression reduces loud peaks, allowing you to increase overall volume (makeup gain) without clipping. The result is higher perceived loudness because average level increases even though peak level stays the same. Compression alone doesn't increase loudness—the combination of compression plus makeup gain does.
Why does my compressed audio sound worse than the original?
Common causes: over-compression (too much gain reduction), wrong attack/release times creating artifacts, or trying to fix problems compression can't address. Reduce your compression intensity, verify attack and release settings, and ensure source audio doesn't have fundamental issues. Sometimes less compression or no compression sounds better.
Ready to Control Your Podcast Dynamics?
Compression transforms inconsistent recordings into polished, professional audio that listeners can enjoy without touching their volume controls. Start with moderate settings and adjust based on what you hear—your ears are the best judge.
Consistently leveled episodes deserve permanent preservation. Transcription transforms your dynamically controlled audio into searchable archives where every properly balanced word becomes findable.
Try PodRewind free and make every well-compressed episode part of a searchable archive.