Research Leader, Music Team, Sony Computer Science Labs
Stefan Lattner brings a rich confluence of tech and music to the table, currently heading the research of the music team at Sony CSL Paris. His expertise in generative AI for music production is paving the way for innovative solutions in the music industry. Over the years, Stefan has worn several hats -- from researching music information retrieval to understanding computational music perception.
Earlier in his career, Stefan was the lead developer at "Re-Compose," contributing to developing "Liquid Notes," a tool designed for re-harmonizing polyphonic music. It's this blend of academic insight and practical experience that shapes his perspective.
As a musician himself, Stefan has a deep understanding of the challenges and joys of music creation. He is deeply passionate about leveraging technology to augment human-computer interaction in music creation, pushing the boundaries of live staging and further exploring computational aesthetics. His knowledge base extends to the intricacies of information theory and how it ties back to musical perception.
On the technical side, Stefan has worked on various projects involving generative models (both symbolic and audio), audio restoration, audio representation learning, and musical structure learning. At Mondo.NYC, Stefan looks forward to sharing his experiences and insights from both the music and tech worlds.
The Well-Trained Model: Ethically Sourced AI for Artists
10/11/23, 9:00 PM
How does an artist know what dataset the AI tool they are using was trained on, and whether their own music was used? A discussion of how AI is trained, and the future of attribution and compensation for artists.