LUMIÈRE
Start free

Lumière · Tools

The music knows when to step back.

You know the fader ride — down for every line, up for every gap, keyframe by keyframe. The engine reads the story, finds music that fits its mood, and does that pass for you. Smooth ramps, every spoken passage.

Start free See the full suite

How it works

Three steps. No timeline required.

  1. 01

    The engine reads the story

    From the transcript it builds a content profile: mood, energy curve, narrative arc. Not a genre dropdown — a music brief drawn from what was actually said.

  2. 02

    Music is found and mixed

    It searches a royalty-free catalog for tracks that match the brief, then mixes the pick under the voice: lowered through every spoken passage, back up in the gaps.

  3. 03

    You make the call

    The Audio panel is yours — search the stock catalog, upload your own track, set the level. The engine’s pick is a starting point, not a verdict.

What you get

Built in, not bolted on.

  • Story-matched search

    Mood, energy and arc come from the transcript itself. A quiet testimony does not get an upbeat bed.

  • The ducking pass, automated

    The music lowers under every spoken passage and returns in the gaps — smooth ramps in and out, not gates snapping open and shut.

  • Your track, your mix

    Stock search and upload live in the Audio panel. Override the pick, adjust the volume, keep what works.

  • AI-generated beds

    With an AI music provider connected, the engine can generate a bed to the brief instead of searching for one.

The catalog and AI generation each need their provider connected. And the ducking follows the transcript — speech that was never transcribed won’t lower the music. The mix stays yours.

One suite, every tool

The same engine, other doors.

Each tool here is the same production suite, entered where you need it. Start with one; the rest is already there.

Start free