Video service YouTube is demonstrating new generative artificial intelligence features that can create music based on text cues or humming a tune.
Here's What We Know
One of them, Dream Track, generates 30-second tracks in the style of popular artists such as Charlie Puth, Sia and T-Pain. This feature is already being tested by some YouTube creators.
Another tool turns singing or beatboxing into musical compositions using virtual instruments. As YouTube notes, artist permission has been obtained for these features to work.
According to YouTube representatives, the purpose of the experiments is a desire to provide creators with new ways to create content in the format of short videos. Previously, the service has already launched tools for generating visual content based on artificial intelligence.
These functions utilise Google's DeepMind Lyria music generation model. According to the developers, all generated tracks will contain a SynthID watermark that is inaudible to the naked ear and can be saved when the track is changed.
AFFILIATE DISCLOSURE Some posts may contain affiliate links. Gagadget.com is a participant in the Amazon Services LLC Associates
Program, an affiliate advertising program designed to provide a means for sites to earn advertising fees by advertising and linking