Speaker |
Sander Dieleman |
|---|---|
Affiliation |
DeepMind |
Date |
Friday, 01 March 2019 |
Time |
13:00-14:00 |
Location |
Zoom |
Link |
Roberts 421 |
Event series |
Jump Trading/ELLIS CSML Seminar Series |
Abstract |
Realistic music generation is a challenging task. When machine learning is used to build generative models of music, typically high-level representations such as scores, piano rolls or MIDI sequences are used that abstract away the idiosyncrasies of a particular performance. But these nuances are very important for our perception of musicality and realism, so we embark on modelling music in the raw audio domain. I will discuss some of the advantages and disadvantages of this approach, and the challenges it entails. |
Biography |