Andrea Martelloni

Audio ML Engineer

Blackstar Amplification

About Me

Andrea is a DSP Engineer and a PhD in Artificial Intelligence and Music at Queen Mary University of London. Currently at Blackstar Amplification, he worked on porting Neural Amp Modeler onto the BEAM range of digital guitar processing devices. Previously, he co-created the meMLP library to allow training ML models directly on musical instruments, and developed the HITar (MIDI Innovation Awards 2023 winner). His ADC 2023 talk “Deep learning for DSP engineers” was co-presented alongside Franco Caspe.

Sessions

  • Audio ML on Bare Metal

    GEMM, Memory Layout, and Real-Time Dataflow on Embedded ARM
    00:00 - 00:00 UTC | Monday 9th November 2026 |
    Advanced

    Audio hardware built around classical DSPs and embedded CPUs (guitar amps, pedals, synths) wasn't designed with machine learning in mind. Yet, Neural Amp Modeler (NAM) can be made to run on the ubiquitous ARM Cortex-M7 at 2x real-time, and models with hundreds of parameters can be trained in tens of milliseconds on the RP2350, a £1 chip. This talk is all about taking a trained model and getting it to run in real time on hardware that has no NPU and no GPU. The core of any ML model is the General Matrix Multiply (GEMM), but on these chips the […]