CoyoPedal – Full-Size Neural Amp Modeler Captures on an ESP32-S3 (github.com)

CoyoPedal has introduced a standalone guitar amp and effects pedal built on the Waveshare ESP32-S3-Touch-AMOLED-2.06, capable of running full-size Neural Amp Modeler A2 captures in real-time. This innovative device transforms the ESP32-S3 into a high-performance digital audio solution, allowing musicians to utilize complex audio processing without needing a desktop converter. The pedal features a responsive touchscreen UI developed in TSX compiled to native C++, eliminating the typical dependency on a JavaScript engine, contributing to its efficiency. This advancement is significant for the AI/ML community as it demonstrates the power of edge computing in audio applications, showcasing the capability of low-cost hardware to handle sophisticated neural network models for sound processing. The pedal operates at 48 kHz with a 23-layer, eight-channel WaveNet architecture, utilizing custom Xtensa kernels across both cores. Additionally, it supports user-loaded captures via microSD, offers a full effects chain, and includes features like remote diagnostics and maintenance through Wi-Fi. This mix of technical sophistication and user accessibility positions CoyoPedal as a noteworthy development in the realm of digital audio and AI integration.