Guide
Streaming Insider Part 6: Pro tips for Deezer
by Florian Bodoky
How to break down music into individual instruments
Splitting music into individual tracks is easy and reliable with AI. Ultimate Vocal Remover is a free and useful tool, and it does more than just isolate vocals.
Artificial intelligence isn’t wholly bad for musicians – there are some useful applications. One of them is the automatic separation of a track into vocals and individual instruments – a pretty practical feature if you want to analyse and replay a piece. Simply put, you can isolate your instrument from the rest, making it much easier to hear which notes are being played. Conversely, you can also remove your instrument from the mix to create the perfect practice track. This is particularly popular for singing, think karaoke.
Free of charge, no restrictions
Tools like these are called stem splitters. There are countless websites and apps for this, but almost all of them either charge a fee (argh, another subscription!) or have massive restrictions.
Luckily, one tool is completely free and does the job: Ultimate Vocal Remover, available for Linux, Mac and Windows. If you want to show your gratitude, buy the creator a coffee.
One note for Mac users: you probably won’t be able to open the app at first due to Apple’s strict security measures. You can fix this problem by scrolling down in the system settings under Privacy & Security and clicking on Open anyway for Ultimate Vocal Remover. After that, the app will open without any problems.
How Ultimate Vocal Remover works
To start off, use Select Input to specify the track you want to split. You can also select several files. Right under that, you can set the output folder. Following this, select the process method. VR Architecture and MDX-Net divide a piece into vocals and instruments, creating two audio files. Demucs is what you need if you want to isolate an instrument. This method separates a piece into bass, drums, vocals and any remaining instruments, generating four audio files.
You’ll also find different AI models for extraction exist across all these processes. These can be selected or downloaded using the drop-down menu directly below. However, simply using the default setting always works too.
Ensemble Mode isn’t a separate process, but allows several methods to be combined. Finally, you can use Audio Tools to change speed and pitch, among other things.
Remixing your new tracks
Individual tracks often sound shoddy on their own, and you’ll also quickly lose your bearings. Understandably then, they require some remixing. For example, to hear which bass notes are being played, I want to have both the bass track as well as the other parts together – only much quieter than the bass, allowing me to hear it perfectly. As a result, I recommend merging the individual tracks in audio software such as Audacity.
Drag the files into the program window where they’ll appear as individual tracks. You can adjust the volume using the plus/minus to the left side of the track. As soon as you’re happy with the mix, select Export file > Audio. Of course, you can also save the Audacity project to edit the mix later. Other apps such as GarageBand work similarly.
David Lee
Senior Editor
David.Lee@digitecgalaxus.chMy interest in IT and writing landed me in tech journalism early on (2000). I want to know how we can use technology without being used. Outside of the office, I’m a keen musician who makes up for lacking talent with excessive enthusiasm.
Audio
Follow topics and stay updated on your areas of interest
Guide
Practical solutions for everyday problems with technology, household hacks and much more.
Show allThese articles might also interest you
12 comments
later