Do you think you have an unpleasant voice? Well, some people can sing while others can’t. It’s just the way of life. But there is nothing wrong in wishing for a tool that can transform your awful hums to magical melodies. All thanks to Google, its new machine-learning algorithm experiment, Tone Transfer, makes that dream come true — at least to an extent.

Tone Transfer uses machine learning models to transform sounds into musical instruments used in a variety of styles, from Baroque to jazz. You can re-render the chirp of a bird into a flute or turn a cello piece into a saxophone solo.

Also Read | Scientists develop AI tool to identify COVID-19 cases using CT scans

“AI isn’t just creating new kinds of art; it’s creating new kinds of artists. We want to extend, not replace, the creative process,” says Doug Eck, research lead of the project.

All that one has to do to transform voice is to visit the website and record 15-second hum or tune. Google’s machine learning algorithm will convert that tune into a digital signal, and then you can convert it into a tune with Flute, Saxophone, Violin, or Trumpet.

The experiment was created by open-source research project Magenta and Google Research. The team includes a group of musicians, researchers, engineers, and designers who collaborate together to make music even more playful and experimental than ever.