Juan Pablo Bello
Bello's current research is concerned with analysing digital music signals, such as those stored in CDs or personal computers in wav, aiff or mp3 format, for distinctive patterns indicative of musical structure.
"Music," says Juan Pablo Bello, "is my long-term love relationship." He's played the bass, guitar and keyboards in many bands, and composed music too. "When I was young it was my dream to be a recording engineer in a studio. When signal processing appeared in my life, I saw it as a way to bring together my passion for music and my interest in engineering."
Signal processing is the development of tools and techniques for the analysis, manipulation and synthesis of digitized signals. "In my research," says Bello, "I am concerned with digital music signals, such as those stored in CDs or personal computers in wav, aiff or mp3 format - to name a few examples."
Bello analyzes these audio signals for distinctive patterns indicative of musical structure. "Music is made out of an interconnected series of events that, in the western tradition, relate to substructures commonly known as keys and chords. I'm currently developing an analysis framework that uses signal processing and machine learning to characterize chords from polyphonic, multi-instrumental music, and approximate the many complex relationships that are present in chord progressions."
This framework has the potential to offer music listeners a more interactive experience with their music. "The idea is to take the music collection on your computer or iPod and analyze it for the similarities between the different songs you have. If you have a lot of songs that are very fast paced, those could be grouped together. If you have a lot of songs with slow dynamics, those will be clustered together in another part of your collection. You could organize your collection according to intrinsic similarities in the musical signals."
Bello is exploring ways of making his research available to NYU students and faculty interested in undertaking their own studies in musicology. "They could visualize musical signals in such a way that would allow them to identify whether the same motif was playing in more than one version of a song. They could sort through the differences between various interpretations of the same piece of music. They could switch between two different versions of a song to see how pianist X played the song compared to pianist Y." He also believes that these technologies can be used to engage high school students in science, mathematics and engineering education through music.
Bello says his students effortlessly understand the music/technology overlap. "They are part of the first generation who understand both. They are very concerned with aesthetics and creative issues in music, but they are also very aware of recent scientific and technological developments and how they can shape the way music is experienced. They are curious and enthusiastic about this line of work, and fully qualified to take it to the next level."
He is equally effusive about his hopes for the Music Technology program. "It's extremely exciting for me to be here right now. Our new PhD program is quickly attracting an outstanding group of researchers who are going to push the boundaries of music technology. We're also getting a new state-of-the-art music facility where we'll be able to gather and share ideas. It's a fantastic time and the possibilities are endless."