Limited-Time Offer: Up to 75% Off Coursera Plus!
7000+ certificate courses from Google, Microsoft, IBM, and many more.
This course teaches learners about a generative model for music that can create entire songs with high quality and consistency. Students will understand how the model can be conditioned on genre, artist, and lyrics to generate music with specific styles. The course covers the technical aspects of using a multiscale VQ-VAE and autoregressive Transformers to compress and model raw audio for music generation. The teaching method involves explaining the concepts and techniques used in the Jukebox model. This course is intended for individuals interested in music generation, artificial intelligence, and machine learning.