Coursera Spring Sale
40% Off Coursera Plus Annual!
Grab it
Explore the development of a transformer-based VST plugin that captures and transfers the expressive groove characteristics of Latin American rhythms to electronic music production in this conference talk from ADCx India 2025. Learn how researchers Anmol Mishra and Satyajeet Prabhu employed machine learning techniques to analyze microtiming and dynamics from onset timing and strength annotations of Uruguayan Candombe drum performances. Discover the technical implementation of their transformer model that learns the subtle variations in timing, dynamics, and other rhythmic aspects that give Latin American music its distinctive expressive feel. Understand how this research bridges the gap between algorithmic rhythm creation and authentic live performance qualities by allowing users to apply learned Candombe micro-rhythms to quantized MIDI drum performances. Gain insights into the challenges of replicating the groove-based elements that are fundamental to Latin American musical expression through electronic music production tools. Examine the broader implications of this work for preserving and digitally transferring the authentic rhythmic characteristics of various Latin American genres into modern music production workflows.
