How to improve dialogue in Final Cut Pro

Good dialogue is the difference between a video people finish and one they click away from. Music and color get the attention, but if the voice is thin, harsh, or uneven, viewers feel it before they can name it. The good news is that Final Cut Pro gives you everything you need to take a rough recording and make it sit clearly in the mix — as long as you work in the right order.

This guide walks through a repeatable dialogue workflow: fix the level first, shape tone with EQ, control dynamics with compression, tame harshness, then check it in context and hit a delivery target.

Improve dialogue in Final Cut Pro by working in the right order

Most dialogue problems get worse when you reach for plugins too early. A clip that’s too quiet will only amplify its own noise once you compress it, and EQ decisions made in solo rarely survive contact with the music. A dependable order looks like this:

1. Level and gain staging
2. Subtractive EQ (cut problems)
3. Compression (even out dynamics)
4. De-essing and harshness control
5. Additive EQ (presence and air)
6. Mix context and delivery loudness

Each step assumes the one before it is done. Skipping ahead is the most common reason a fix sounds worse than the raw clip.

Start with level, not plugins

Open the Audio Inspector and look at the clip’s gain before touching any effect. If speech is consistently sitting below roughly −18 LUFS, add gain in small steps until it lands in a workable range. You want the loudest words peaking comfortably without clipping, and the quietest still audible.

If you’re working across many interview clips, assign them to a dialogue role so you can process the whole role at once instead of clip by clip. It keeps your treatment consistent and makes later changes a single edit rather than fifty.

Cut problems before you add anything

Subtractive EQ does most of the heavy lifting. Add an EQ effect to the dialogue clip or role and address issues in this order:

• High-pass below about 80 Hz to remove rumble, handling noise, and air-conditioning hum that carries no speech information.
• Cut the mud with a gentle 3–6 dB dip around 200–350 Hz when the voice sounds boxy or buried. Use a medium Q and A/B against bypass so you don’t thin the voice into a telephone sound.
• Notch harsh resonances if a specific frequency rings — sweep a narrow boost until the ugliness jumps out, then cut there instead.

Cut first, boost later. It’s easier to hear what a voice actually needs once the problems are out of the way.

Even out the dynamics with compression

Interview and voiceover levels drift — someone leans in, trails off, or emphasizes a word. Compression smooths that so every phrase lands at a consistent level. A ratio between 2:1 and 3:1 with a medium attack lets consonants punch through while the release breathes with the sentence. Keep makeup gain modest; overall loudness is a mix decision you’ll make later, not something to chase per clip.

If you hear the room noise pumping up between phrases, you’re compressing too hard or your threshold is too low. Back off until the effect is felt but not heard.

Tame the harsh "s" sounds

Once you compress and add presence, sibilance often gets sharper. A de-esser targeting the 5–8 kHz range calms hard "s" and "t" sounds without dulling the whole voice. Aim for a light touch — a couple of dB of reduction on the loudest sibilants is usually enough. Over-de-essing gives you a lisping, lifeless voice, which is harder to fix than the harshness you started with.

Add presence and air last

With problems removed and dynamics under control, small additive moves bring the voice forward:

• A gentle lift around 3–5 kHz adds intelligibility and presence.
• A soft high shelf above 10 kHz adds "air" and openness.

Keep these boosts subtle — 1 to 3 dB. If you find yourself adding a lot here, revisit your subtractive EQ, because you’re probably compensating for a problem you didn’t fully cut.

A faster path with Dialogue Enhancer

That whole chain — high-pass, mud cut, presence, compression, a touch of saturation for warmth — is a lot to balance by hand on every project. If you’d rather get most of the way there in one move, our plugin Dialogue Enhancer is built natively for Final Cut Pro and drives EQ, compression, and saturation from a single Enhance knob. You pick a starting tone — Neutral, Warm, Bright, or Broadcast — then dial the amount to taste.

It won’t rescue a badly recorded take or replace heavy noise reduction, but for thin lavs, dull rooms, and uneven interviews it’s a quick way to make dialogue clearer and fuller without stacking five plugins yourself. It releases 31 July — subscribe on the home page or visit the product page if you want a heads-up on launch day.

Check the mix, don’t trust solo

Solo fixes lie. After each change, toggle the music and effects back in — a voice that sounds perfect alone can still disappear under a bed. If dialogue hides, duck the music 2 to 4 dB under speech rather than pushing the voice into distortion. Automation or a ducking effect on the music role handles this cleanly.

Hit a delivery target

Finish by measuring loudness on the summed output, not per clip. For YouTube and most streaming, aim near −14 LUFS integrated with true peak below −1 dBTP. For broadcast, check your deliverable spec — many European broadcasters expect −23 LUFS. Put a loudness meter on the master or dialogue sum and confirm the whole program lands where it should before you export.

Get these steps in order and dialogue stops being the thing you fight in the mix — it becomes the thing that carries it.