New Script: Demucs Stem Splitter for Vegas Pro (AI Audio Separation)

kjetil wrote on 8/17/2025, 11:57 AM

Hi everyone,

I’ve been working on a script for Vegas Pro that integrates Demucs to separate audio tracks into stems (vocals, drums, bass, guitar, piano, etc.) and then automatically imports them as separate audio tracks inside Vegas.

I’m now sharing my current build in case anyone wants to test it, play around with it.

What’s included

demucs_cli.exe (required AI engine, ~5.6 GB unpacked – zipped download provided)
Vegas Pro .dll script file
Configuration .ini file
Demo screenshot (JPG)
Demo video (MP4)
install.txt with simple installation instructions

How to use

Download and unzip the package (link below).
Follow the steps in install.txt to copy the files into your Vegas Pro script folder.
Run the script on an audio event – Demucs will process it and bring the stems back into your Vegas project as new tracks.

Download

👉 Google Drive link to package: https://drive.google.com/drive/folders/1Iw3w6y_2MVOLo5zN6YYIYvdL6W2xq_GX?usp=sharing

(Please note: the AI engine is large – around 5.6 GB when unpacked.)

Demo

Here’s a short preview video showing how it works:

Here is a screenshot of the menu showing options

Important

This is an early build and provided as-is.
Tested in Vegas Pro 21/22 (should also work in other recent versions).
Performance depends on your system (GPU recommended).
No warranties – use at your own risk.

Thanks,
Kjetil

Back to post

Comments

JimmyH wrote on 11/5/2025, 2:00 AM

Also, just want to add that there is a free program which provides a rough idea of how well Demucs might work with your audio files in Vegas before you render and take up space on your computer. Not saying it is better or the same or worse, just interesting.

It is called Virtual DJ and it is kinda built for live entertainment needs. But to my point, it allows for immediate stem-splitting which in some cases saves you trouble when debating whether to put an audio file into the Demucs tool and gobble up more space on your computer. Interface comes in three different styles depending on your preferences/skill, timeline operates differently than Vegas', default scheme is dark, VST controls will feel familiar, full workstation in one window so put on your reading glasses.

JimmyH wrote on 11/5/2025, 2:05 AM

I learned of this software on a really back-alley website from a post which was made in 2005, so it looks like a pretty stable tool which is updated with new capabilities from time to time and for those of us who are non-pro, non-competitors, it can be a lot of fun.

mamomo wrote on 11/5/2025, 11:11 AM

Buen trabajo!

john_dennis wrote on 11/5/2025, 2:21 PM

Comment on a particular use-case from another thread.

Demucs for VEGAS Pro

zzzzzz9125 wrote on 11/5/2025, 7:15 PM

Comment on a particular use-case from another thread.

Demucs for VEGAS Pro

Since @Former user's thread was deleted, I'll post my ideas here.

@john_dennis's comment:

I used the scripted version of Demucs to see if I could reduce the background noise of a speaker in a room full of chatty elementary school students and parents. I would classify the effort as a failure.

I wasn't surprised. This is a thought-experiment for me.

@john_dennis Based on an open-source and free AI audio separation project: MSST WebUI, I recorded a test for you.

MSST WebUI is a locally deployed open-source and free AI audio separation tool that can use multiple types of models such as BS RoFormer and Demucs. Install it here for free: https://huggingface.co/Sucial/MSST-WebUI/tree/main/1.7.0

In the video, I used the model_mel_band_roformer_karaoke_aufr33_viperx_sdr_10.1956.ckpt model from BS RoFormer series, which can separate the main vocal track within the audio.

(If your original audio doesn't contain other noisy vocals like this, you can try using other models like model_bs_roformer_ep_317_sdr_12.9755.ckpt to separate full vocals, which effect is better.)

I just want to clarify that Demucs is not the only option. There are better alternatives than it.

In fact, the AI audio separation tool for VEGAS Pro that I plan to develop is just a shortcut to invoke the API of MSST WebUI. MSST WebUI has a complete API for script users to refer to. It won't take much effort to develop this, I think.

In short, I'll release an open source free AI audio separation tool for VEGAS Pro based on MSST WebUI in the future. When it's complete, I'll open a new thread to promote it.

john_dennis wrote on 11/6/2025, 8:31 AM

@zzzzzz9125 The results you achieved are much better. Thank you. This example was a straw man, I'm not particularly bothered by a room full of chatty elementary school students and parents.

Samsarulz wrote on 11/17/2025, 8:43 PM

Tested on Vegas 20 and worked too. CPU seems to take a bit more time to process tho. Have Vegas 20 (403) and 23 (278) in my PC right now and have no issues so far. Thanks