Add music to video with AI
Replace a video's soundtrack, keep its ambience or let the original audio pull the music down as someone speaks.
To add music to video with AI, start by deciding which sounds in the footage matter. A potter's wheel in a demonstration might be worth keeping, while wind noise on a silent product shot might not. With audio-mix-maker, the choice between those two edits is explicit. Its default replaces the original soundtrack.
A third mode lets the original sound control how far the added track drops. This can help a conversation sit above the music. A loud tap on the workbench can trigger the same response, so listen to both a spoken passage and a noisy one before choosing the mix.
What it does
- Replace an unwanted camera soundtrack with a permitted piece of music.
- Keep workshop ambience audible underneath a background track.
- Try automatic ducking against the original sound.
- Adjust how the added track enters and fades out.
- Save separately named versions for a comparison from the same point.
Install and use
- Provide a local video and audio file, the maker's shared Python runners and FFmpeg. Mixing runs locally and needs no provider account.
- Ask, for instance, to keep the room sound, lower the music around the instructor and write a new file without touching the original.
- Listen to the speech, a quieter passage and the ending. If ducking follows background noise too closely, compare a lower constant overlay.
FAQ
Source
Made by Mikhail Savchenko. Original skill: audio-mix-maker, published on GitHub under the MIT licence.
View on GitHub