Enhanced Drums Extraction Quality with Phoenix

LALAL.AI expands the possibilities of Phoenix, the revolutionary stem separation technology, with drums support. Extract or remove percussion from any audio/video source with utmost precision.

Enhanced Drums Extraction Quality with Phoenix

Just two months ago LALAL.AI released Phoenix, the in-house developed technology for voice and vocal stem extraction from audio and video sources. We’ve done a lot of research and training to make the new neural network not only better than our previous one but also surpass all other AI-based vocal splitters available on the market.

Numerous quality tests and overwhelmingly positive feedback from users are a testament to how much more powerful and sophisticated Phoenix is in comparison to Cassiopeia, the previous generation stem separation technology. It inspired us to expand the possibilities of Phoenix by training it to extract other stems with the same unprecedented precision as it does with vocals.

Today we present the improved isolation and removal of drums, the most popular out of the six musical instrument stems LALAL.AI supports.

What’s New

Drum stem separation is not a new feature per se – it’s been available within LALAL.AI for a hot minute. However, now that the drums isolation is handled by Phoenix instead of Cassiopeia, the process and the results are better by a long shot. Here is what has changed:

1) Processing speed. The time it takes to detect the percussion sounds and cutting them out of a track depends on a lot of factors, including file duration and size. LALAL.AI has always been swift and hasn’t left users hanging for more than a few minutes to get the splitting results. Now it’s even faster – with Phoenix, the drums stem is isolated twice as fast as before.

2) Splitting accuracy. Since Phoenix is capable of observing more data, recognizing all instruments that make up the composition, and considering various characteristics of drums better than Cassiopeia, it can extract the drums stem with maximum precision.

3) Sound clarity. Thanks to the improved architecture and innovative processing techniques, Phoenix delivers superior splitting results with very few to none sound artifacts and noise. The sound of drums and other percussion instruments is clear and crisp, in many cases comparable to studio quality.

How to Extract the Drums Stem with LALAL.AI Phoenix

The drums isolation process hasn’t changed on the user’s end, it’s exactly the same with Phoenix as it was with Cassiopeia. The only difference you will notice is that the drums stem now has the icon ⓟ next to it (just like the Vocal and Instrumental stem), signifying that the splitting is performed by the Phoenix neural network. Follow the steps below.

  1. Open LALAL.AI in your browser.
  2. Click the Vocal and Instrumental button to open the full Stem Separation Types list.
  3. Select Drums in the drop-down menu.
  4. Click the Select File button to upload an audio or video track.
  5. Wait a few seconds while the service generates stem previews.
  6. Play the stem previews  - one containing the drums track, the other with the rest of the music.
  7. Sign up and click Process the Entire File if you like what you hear. If you don’t, upload another version of the track or try another file.
  8. Click the download icon (⤓) next to the stems to download them to your device.

That’s it! The stems will be saved in the same format and quality as the track you’ve uploaded. The download links will also be sent to the email you’ve signed up with.

What’s Next

LALAL.AI’s goal is to provide the easiest, fastest and most precise audio source separation. Drums is the first but definitely not the last musical instrument stem that can be extracted in high quality. We are going to continue improving our revolutionary technology and support the extraction of more instruments in the future. Stay in the know of all our updates, news and new features – follow LALAL.AI on Twitter, TikTok, Instagram, Facebook, YouTube, and Reddit.

Cookies

For magic to happen, we use cookies. Read our Privacy Policy to learn more.