Google Using AI to Create Never-Before-Heard Instruments

Google is developing a new program to synthesize sound using machine learning techniques. Developed by Google company Magenta, a part of their artificial intelligence division Google Brain, the new project is dubbed NSynth (“Neural Synthesizer”). The program is fed samples of real-life instruments (organs, flutes, basses, etc), which are then analyzed using machine learning to synthesize a new sound. While the result sounds like a garbled, digital rendering of the original sample, the NSynth gets interesting when the vectors produced by analyzing musical samples are combined. Check out some clips below via Wired. According to Wired, the Magenta team is looking to expand the limits of how AI can be used to create new tools for musicians. Check out more NSynth examples here.

Comments are closed.