Adobe’s new experiment turns tone deaf humming into real music

What if your tone deaf humming could actually sound… pretty good? That’s the premise of a new experiment from Adobe that aims to turn your voice into music.

It’s called Project Kazoo, and it could one day make soundtrack creation as simple as humming a quick tune.

The software, which is still in a prototype phase, uses a technology called voice conversion to translate dissonant sounds into music. Hum a couple bars into a microphone, and the software can turn the sound into that of the well-tuned instrument of your choice. It can even convert music from instrument into the sound of another — play a few notes on a recorder and the software will spit out the same notes on a saxophone.

Watch the video, below, to see (and hear) it in action.

Behind the scenes, Project Kazoo uses machine learning to translate the sounds into the corresponding instrument. 

“It listens to a voice and takes the essence of your voice, how the pitch variates and how it changes over time, and then it converts that kind of essence into music using machine learning,” says Adobe research scientist Zeyu Jin.

It’s a neat concept, and one that Jin says could be particularly useful as a plug-in for some of the company’s audio and video editing services, like Audition and Premiere. Instead of having to find or create original sound effects or music for projects, Jin says, Kazoo would instead enable creators to make their own, regardless of relative musical ability.

The project is still just an experiment for now, Adobe previewed it as part of the “Sneaks” portion of its Max conference, so it may not ever see the light of day. But, if it did, it could give even the most tone deaf new confidence in their abilities.