Voice & language

The right track.At the right volume.

Licensed music chosen for your video's mood, beat-matched to your cuts, and auto-ducked under every word you say.

8,000+
licensed tracks
−12 dB
auto-duck under speech
100%
cleared for monetized content
audio_agent
source_track.wavprocessed
Levels
−14 LUFS
Bed
auto-ducked
Noise floor
−58 dB

Picked by mood, not by menu

Scrolling a stock library is its own job. The agent reads your video — energy, pacing, subject — and proposes five tracks that fit, each previewed in place under your actual footage. Filters by genre, mood, and energy are there when you'd rather drive.

  • Five fitted suggestions per video, previewed under your footage
  • Genre, mood, energy, and instrument filters
  • 8,000+ tracks, all cleared for monetized use on every platform
  • Memory learns the sound you keep choosing
audio_agent
Five fitted suggestions per video, previewed under your footage
Genre, mood, energy, and instrument filters
8,000+ tracks, all cleared for monetized use on every platform

Mixed like a sound engineer

Music that fights the voice is worse than no music. The bed is sidechain-ducked −12 dB under speech, eases back up in pauses, lands its cut points on the beat grid, and resolves cleanly at your outro instead of stopping mid-bar.

  • Sidechain ducking: −12 dB under speech, adjustable
  • Loop and trim to length with cut points on the beat
  • Intro and outro resolved musically, never clipped mid-bar
  • Full mix editable on the audio track afterward
audio_agent
Sidechain ducking: −12 dB under speech, adjustable
Loop and trim to length with cut points on the beat
Intro and outro resolved musically, never clipped mid-bar

How Background Music works

  1. 1

    Read the room

    The analysis agent classifies your video's tone and pacing, then the audio agent shortlists five tracks that fit — not five hundred to scroll.

  2. 2

    Fit the track

    The music is trimmed and looped to your runtime, with its strongest section under your strongest moment and cut points on the beat.

  3. 3

    Mix it under you

    Sidechain ducking drops the bed −12 dB whenever you speak and breathes back up in the gaps. Speech stays first, always.

Frequently asked

Ready to try Background Music?

No credit card. 10 free AI minutes every month. The audio agent is ready when you are.