You can also drag them sideways to change the vowels they're singing. The way you use it is that the interface shows the four blobs, and you can drag them up or down to change the pitch they're singing in. You don't hear the voices of the original singers, you hear what the ML model learned opera sounds like. They recorded many hours of singing, and this was used by the machine learning model to learn what opera singing sounds like. This is the end result of a machine learning experiment by artist David Li, and a team at Google that developed a machine learning model trained on the voices of four opera singers: tenor, Christian Joel, bass Frederick Tong, mezzo-soprano Joanna Gamble and soprano Olivia Doutney.
Once you get beyond the initial horror, it's an interesting (ok, and fun) idea. Google says you can be sure this will be the first blob opera, though of course, depending how cruel a set of techie friends you have, the inboxes of your contacts may already be overflowing with blob operas. You can then record your creation - and here comes the good part - share it with family & friends, especially the person who already seems to have everything. Google suggests that you can guide the pitch and vowel sound of four festive blobs who stand ready to transform your musical ideas (no matter how good or bad) into beautiful harmonies. The Blob Opera looks as bad as the idea sounds (or as wonderful, if you're on the "isn't winter wonderful" side of the bah humbug spectrum). Google has put its immense resources to a new use - developing the Blob Opera, a new machine learning experiment that lets you create your own festive song inspired by Opera on Google Arts & Culture. If you're already traumatized by festive season overload, look away now.
Additional singing from Ingunn Gyda Hrafnkelsdottir and John Holland-Avery.īlob Opera is an experiment using the latest web audio technology and may not perform optimally on older devices.Create Festive Songs With Google's Blob Opera Tenor Christian Joel, bass Frederick Tong, mezzo-soprano Joanna Gamble and soprano Olivia Doutney. Another machine learning model lets the blobs respond to and harmonise your input in real time.Ĭlick the Christmas Tree for a holiday surprise based on the top searched Christmas Carols of the year. Or forwards and backwards for different vowel sounds. In the experiment you don’t hear their voices, but the machine learning model’s understanding of what opera singing sounds like, based on what it learnt from them.ĭrag the blobs up and down to change pitch. Tenor, Christian Joel, bass Frederick Tong, mezzo‑soprano Joanna Gamble and soprano Olivia Doutney recorded 16 hours of singing. We developed a machine learning model trained on the voices of four opera singers in order to create an engaging experiment for everyone, regardless of musical skills.
This experiment pays tribute to and explores the original musical instrument: the voice. “Blob Opera is a machine learning experiment by David Li in collaboration with Google Arts and Culture.
0 kommentar(er)
