Skip to main content
PremiereCopilot logo
The native AI plugin for Adobe Premiere Pro
Adobe official partner

Mixed likea sound engineer.

Auto Mix is an AI sound engineer for your Premiere Pro timeline. It levels every source, cleans up the voice and ducks the music under each line, with real, editable audio effects. You review the mix before it is applied.

Voice at -16 LUFS, music under the dialogue, only the processing each clip needs. Name your platform, review the plan, apply in 1 click. Native to Premiere.

Download freeSee pricing
Auto Mix panel listing the mix of each clip, with voice levels on A1 and music ducking on A2 of a Premiere Pro timeline

How it works

Three steps. A finished mix.

  1. Brief

    Say how it should sound.

    Name your platform and the voice and music balance, like "podcast mix for YouTube, music well under the voice". Optional.

    Typing a podcast mix brief in the Auto Mix panel

  2. Review & apply

    Levels, effects and ducking in one pass.

    Each clip is read as speech, music or SFX and gets only what it needs. Review the plan, then apply it to your timeline.

    Applying the mix: voice clips leveled on A1, music ducked under the voice on A2

  3. Fine-tune

    Real effects. Real keyframes.

    Everything lands as standard Premiere audio effects and volume keyframes. Drag a keyframe to reshape a duck, or undo it all.

    Dragging a ducking keyframe on the music track, then playing the passage

What ships

A clean mix in minutes, not hours.

0

Delivery targets: podcast -16 LUFS, YouTube -14 LUFS, broadcast -23 LUFS.

1 min

Roughly one minute of processing per hour of timeline.

Review first

See every clip's gain, effects and ducking before anything is applied.

Auto ducking

The music dips before each line, holds under it and comes back after.

CH · Premiere Pro

Standard Premiere audio effects and keyframes. Undo it all with Cmd/Ctrl+Z.

Stack the rest

Explore other tools.

See all 24
From the Discord

Real users, real reactions.

PremiereCopilot user testimonial from the Discord community
PremiereCopilot user testimonial from the Discord community
PremiereCopilot user testimonial from the Discord community
PremiereCopilot user testimonial from the Discord community
PremiereCopilot user testimonial from the Discord community
PremiereCopilot user testimonial from the Discord community
PremiereCopilot user testimonial from the Discord community
PremiereCopilot user testimonial from the Discord community
PremiereCopilot user testimonial from the Discord community
PremiereCopilot user testimonial from the Discord community
PremiereCopilot user testimonial from the Discord community
PremiereCopilot user testimonial from the Discord community
PremiereCopilot user testimonial from the Discord community
PremiereCopilot user testimonial from the Discord community
PremiereCopilot user testimonial from the Discord community
PremiereCopilot user testimonial from the Discord community
PremiereCopilot user testimonial from the Discord community
PremiereCopilot user testimonial from the Discord community
PremiereCopilot user testimonial from the Discord community
PremiereCopilot user testimonial from the Discord community
PremiereCopilot user testimonial from the Discord community
PremiereCopilot user testimonial from the Discord community
PremiereCopilot user testimonial from the Discord community
PremiereCopilot user testimonial from the Discord community
PremiereCopilot user testimonial from the Discord community
PremiereCopilot user testimonial from the Discord community
PremiereCopilot user testimonial from the Discord community
PremiereCopilot user testimonial from the Discord community
Join the Discord

FAQ

What you're probably wondering.

It listens to every audio clip of your sequence, tells speech, music and SFX apart, and mixes them like a sound engineer: a level for each source, a cleaned-up voice, and the music ridden down under every line.

Only what each clip needs: level, high-pass, compression, de-noise, de-esser, limiter. A music bed gets a level and nothing else. Wherever music or SFX overlap speech, volume keyframes duck them under the voice.

The voice is the anchor: -16 LUFS for podcast (default), -14 LUFS for YouTube and social, -23 LUFS for broadcast. Name your platform in the brief and the whole mix re-anchors around it.

Yes. Analysis and application are separate. You see every clip with its type, gain, effects and ducking, then click Apply Mix.

Yes. The mix lands as standard Premiere audio effects and volume keyframes. Retune anything in Effect Controls, drag a keyframe to reshape a duck, or undo it all with Cmd/Ctrl+Z.

Premiere Pro only, on macOS (Intel and Apple Silicon) and Windows.