Research · Audio ML
Guitar Pedal AI
An audio-to-control system that listens to processed guitar and works toward reconstructing an editable effect chain instead of generating an opaque replacement waveform.
Project overview
This final-year Berklee project investigates the inverse problem behind guitar tone: given a wet guitar recording, can a model infer effect structure and controls that a musician can inspect, edit, and reuse? The research evolved through several architectures as I studied where musical content, dry-signal variation, and effect identity become entangled.
In the current direction, DDSP provides the underlying structure for reconstructing delay-line-based effects. The broader aim is to combine machine learning with explicit DSP modules so that the output remains useful inside a producer’s workflow. The final case study will include dataset construction, model iterations, evaluation, and honest failure analysis.
Project demonstration
Explore the Guitar Pedal Interface
Watch the project showcase above. The interactive model interface is currently available in the local version of this project.