Audio Stack

AI Voiceover & Audio

Fractal's audio stack covers voiceover generation, background music, ducking, narration-driven pacing, and waveform cleanup. It is the layer that turns a visually assembled edit into something that actually sounds finished.

See Pricing Read Docs See Create With AI →

Best for creators who need voice, music, and pacing to work together instead of treating audio as an afterthought.

What It Solves

Why Audio Matters In Fractal

Finished videos need finished audio

Strong visuals still feel rough if the voiceover is weak, the music is too loud, or the pacing around narration is off.

Supports different workflows

Use fast TTS, saved narrators, manual voiceovers, or cleaned-up recordings depending on how much control you need.

Integrated, not bolted on

Audio connects directly to Advanced Editor, Video Creator, Script Shufflr, and Create With AI instead of living in a separate disconnected app.

Quick Workflow

How The Audio Layer Works

Add or generate the voice

Choose an existing voiceover, generate TTS from text, or use a saved narrator when you want more consistent brand voice.

Add background music

Use the audio controls to choose BGM and set the relative balance between music and voice.

Use ducking and preview

Lower BGM automatically during voice segments and preview the mix before exporting so narration stays intelligible.

Clean up the recording if needed

Open Voice Editor to trim silences, adjust gain, or export a cleaner audio file back into the wider workflow.

Audio Controls

What You Can Adjust

BGM and voice levels

Set the relative volume of background music and narration instead of settling for one fixed mix.

Ducking

Automatically lower background music when voiceover is present so speech remains clear without constant manual balancing.

Preview Mix

Audition the blend before export rather than waiting for a finished render to discover the music is overwhelming the voice.

Keep or remove source audio

Use remove-audio behavior for silent B-roll workflows or preserve source audio when the original sound should remain part of the final edit.

Voice Workflows

TTS, Narration, And Cleanup

Text-to-speech

Generate narration directly from text and use it in the audio card or larger production workflows.

Saved narrators

Create With AI narration can use either a simple voice or a saved narrator so sound and writing style stay more consistent across videos.

Voice Editor cleanup

Record, open, preview, trim silence, adjust volume, and export a cleaner voice track when you need a more manual polish step.

Narration And Pacing

Why Voice Often Controls The Video

Voice drives timing

The Create With AI narration docs treat spoken audio as a pacing backbone, so more words usually means a longer project and fewer words means a shorter one.

Match total length to the VO

If you already have a finished voiceover, set total length to it so the visual plan is built around the actual spoken runtime.

Skip generated narration when needed

Use your own voiceover and skip generated narration when preserving your original script and delivery matters more than speed.

Plan Context

Where Audio Fits In video shufflr

Core Pro capability

The docs tie audio mixing, voiceover, and Voice Editor capabilities to the Pro stack, making this a meaningful part of the paid workflow.

Platinum expands the narration story

Create With AI narration and some adjacent managed AI access live higher in the stack, especially when you want narrators and AI-driven builds.

Works across multiple editors

This is not limited to one surface. Audio behavior shows up in shuffle workflows, Advanced Editor, Video Creator, and Create With AI.

Related Features

Need The Full Audio Workflow?

Read the docs if you want the deeper voice and narration breakdown, or use pricing if audio control is one of the key reasons you want Fractal.

Read The Docs See Pricing