AI Music Composition in VR

This session will provide a hands-on experience into machine-generated music for video games. After a short presentation about how Artificial Intelligence (AI) can enable people with no musical skills to create highly immersive soundtracks, attendees will try a Virtual Reality (VR) experience with music generated in realtime by an AI music system called Melodrive. While in the virtual world, attendees will be able to change the emotional state of the music in realtime.

Episode 39 of the re:publica 18 - Arts & Culture podcast, hosted by Valerio Velardo, Ryan Groves, Christian Tronhjem, titled "AI Music Composition in VR" was published on May 2, 2018 and runs 107 minutes.

May 2, 2018 ·107m · re:publica 18 - Arts & Culture

Summary

Valerio Velardo, Ryan Groves, Christian Tronhjem Is it possible for machines to compose interesting music? Can we leverage intelligent machines to help people with no musical skills to create music? What’s the best use of machines in music creation? These are some of the questions we’ll explore in this session, which focuses on the use of Artificial Intelligence (AI) for generating music for video games. At the beginning of the session, we’ll give a brief presentation about the impact of AI on music creation. We’ll provide a short overview of what has been done in this field to date, and analyse what are the weaknesses and strengths of using AI for generating music. The core of the presentation will be about how AI can enhance people’s musical creation abilities, regardless of their musical skills. Towards the end of the presentation, we’ll introduce Melodrive - an AI music system that is able to create music in realtime for interactive content like video games and Virtual Reality (VR) experiences from scratch. After the presentation, attendees can try Melodrive in a VR demo. Once in the VR experience, people can wander around in a virtual stylish house. The soundtrack of the experience is generated in realtime by Melodrive. People will be able to spawn spheres, assign them emotions, and place them wherever they prefer in the virtual environment. By doing so, they’ll modify the music generation process. If a person gets closer to a ‘happy’ sphere, the music generated will start sounding happier. When attendees get back to the real world, they’ll be asked to fill a short form to provide feedback on the demo. This workshop is hosted by the Berlin Music Commission

Episode Description