Audio swap

Audio swap

The audio swap is the single most important transform in the export pipeline. Read this once before you ship a clip you care about.

Why we mute the original

The platforms run audio fingerprint engines. TikTok runs ACRCloud. YouTube runs Content ID. Meta runs its own. They all do the same thing. They sample your audio, build a fingerprint, and check it against a database of every track they have ever indexed.

If the original soundtrack is mixed in even quietly the fingerprint match still fires. Mixing does not defeat detection. Only full replacement does.

That is why every export with the audio swap turned on mutes the original track at the file level. The new track stands alone. The original audio does not leak through.

The licensed pool

Open the Audio tab in the editor. The free tier shows you 25 hand curated tracks from Mixkit Stock Music. We picked them for vibe, length, and a clean rights trail. You can drop any of them into any export at no cost.

The Pro tier adds AI generated tracks. Pro users see a second tab with mood prompts that generate fresh music to fit your clip.

What the license does not cover

Read this slowly. The pool gives you cleared rights to use the audio inside a video posted to social platforms. That is the full scope.

What it does not cover:

1. Content ID registration. You cannot claim any of these tracks as your own audio on YouTube. The license forbids it.
2. Standalone audio releases. You cannot upload a track from the pool to Spotify, Soundcloud, or anywhere else as music.
3. Resale. You cannot bundle the audio into a product.

If you need a track you can claim and monetize on YouTube you need a custom license from the original rights holder. The pool is for posting clips. Not for music distribution.

Bringing your own audio

You can also drop in your own MP3 or WAV. Use the Upload button in the Audio tab. If you own the rights or have a separate license you are good. If you do not, you are on your own with the platforms. We do not check what you upload.

How the export wires it in

The render pipeline replaces the audio at the ffmpeg level. The output file carries the swapped track only. The original audio is gone, not faded. Verified in the variant export test suite.