Expressive Deep Learning with Magenta DDSP w/ Jesse Engel - #452
Analysis
This article summarizes a podcast episode of Practical AI featuring Jesse Engel, a Staff Research Scientist at Google's Magenta Project. The discussion centers on creativity AI, specifically how Magenta utilizes machine learning and deep learning to foster creative expression. A key focus is the Differentiable Digital Signal Processing (DDSP) library, which combines traditional DSP elements with the flexibility of deep learning. The episode also touches upon other Magenta projects, including NLP and language modeling, and Engel's vision for the future of creative AI research.
Key Takeaways
Reference
““lets you combine the interpretable structure of classical DSP elements (such as filters, oscillators, reverberation, etc.) with the expressivity of deep learning.””