Franco Caspe is a PhD researcher exploring the intersection of acoustic instruments and synthesizers. His research focuses on bridging the gap between these two worlds, using artificial intelligence (AI) for timbre transfer: analyzing performance features from instrument audio and generating synthetic sound renderings.

His work addresses two challenges in AI model design. First, he aims to create systems that enable meaningful musical interaction, allowing musicians to explore new sonic possibilities in ways that support intimate control, embodiment, and appropriation. Second, he addresses the technical challenge of latency, designing models that minimize it, providing a lower bound for responsiveness.

As an instrumentalist, Franco enjoys playing electric guitar and singing in band settings, with styles ranging from funk and R&B to bossa nova. As a maker, he has developed various personal projects related to audio synthesis and musical performance. He also enjoys analog photography, taking travel pictures with medium-format press cameras.

Franco holds an Erasmus Mundus Joint Master's Degree in Computer Vision and a degree in Electronic Engineering from UTN-FRBB. He has interned at Yamaha, contributing to their Vocaloid voice synthesis software, and has experience in fields such as antenna design, real-time systems for image processing, and communications. His expertise lies at the intersection of signal processing, machine learning, and audio, focusing on real-time applications and performance-driven synthesis.

— Electronic Engineering (UTN FRBB) - MsC Image Processing and Computer Vision (EMJMD IPCV)