AI and machine learning seem like buzzwords at this point, with their mention spattered across press releases and new product launches. The ubiquity may have numbed us, but some cool and weird things are happening at least. Algoriddim leveraged AI for an update to its djay Pro software a couple years back, improving its Automix capability. It seems the company has been developing some more interesting uses since then with today’s launch of Neural Mix, as part of the new djay Pro AI. These new tools leverage Apple’s Core ML framework and the A12 Bionic chip (or higher) to provide on-the-fly audio separation -- similar to what you get with stem files -- but extracted from a single streaming waveform.
The Automix AI feature is still on board, but the Neural Mix tools and related UI refresh are what’s interesting with this update to djay for iOS. Some songs have been on the market for a while as stem files, giving buyers multi-track audio so they could mix using any of four separate song elements. What Neural Mix does, is actually extract elements from a track on-the-fly. You have sliders to adjust the levels of three segments from the audio track: drums, vocals and harmonic elements.