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.

Enjoy.Blob Opera is a machine learning experiment by David Li in collaboration with Google Arts and Culture. Here is a Blob Opera original that I crafted on the fly. If you can't get your blobs performing the way you'd like, you can toggle a switch and listen to them singing Christmas songs like Jingle Bells and Silent Night. What comes out of the blobs' mouths, though, is a machine learning interpretation of what opera sounds like. The blobs' voices are modeled on actual opera singers. 'This experiment pays tribute to and explores the original musical instrument: the voice,' Google said in a blogpost Tuesday. Using machine learning, the blobs automatically harmonize with each other, and if you like what they're signing, you can record your musical creation. With Blob Opera, you have a set of four blobs - a bass, mezzo-soprano, soprano and tenor - whose pitch and vowel sounds you can change by moving them up or down, forward or back. A quartet of opera signing blobs is at your command, courtesy of an interactive arts and culture offering from Google and artist David Li called Blob Opera.

Arts and culture blob opera