Spectral Analysis and Detection of Extreme Vocal Effects by Oriol Nieto, Pandora

Abstract:

Extreme Vocal Effects are aggressive singing techniques that have increased in popularity in the past two decades mostly due to the evolution of heavy metal music. In this talk we introduce and classify such effects, discuss some of their spectral qualities, and show how they might be able to be automatically detected using deep convolutional architectures.

Bio:

Senior Data Scientist at Pandora who previously pursued a Ph.D in Music Data Science at the Music and Audio Research Lab (MARL) in NYU. My current research focuses on topics such as music information retrieval, large scale recommendation systems, and machine learning with especial emphasis on deep architectures. My Ph.D thesis is about trying to better teach computers at "understanding" the structure of music. You can find it here. I develop open source Python packages (such as msaf), play guitar, violin, and sing (and scream) in my spare time.