The ESP32 microcontroller has proven remarkably capable for real-time audio digital signal processing, and the YummyDSP library makes this power accessible to hobbyists and professionals alike. This open-source framework provides a collection of audio processing building blocks optimized for ESP32, enabling you to create guitar pedals, studio effects, synthesizers, and audio tools with professional results at a fraction of commercial product costs.
What is YummyDSP
YummyDSP is a lightweight, efficient DSP library specifically designed for audio effects processing on the ESP32 platform. Unlike general-purpose DSP libraries that try to do everything, YummyDSP focuses on real-time audio with low latency and clean implementation of essential audio processing blocks. The library handles the complexities of I2S audio interface configuration, buffer management, and efficient arithmetic while providing an intuitive API for building effect chains.
The library includes common building blocks: filters (low-pass, high-pass, band-pass, notch), delays, reverbs, compressors, limiters, distortion, and modulation effects like chorus, flanger, and phaser. Each module is optimized for the ESP32’s architecture, taking advantage of hardware floating-point support and minimizing computational overhead. This optimization enables combining multiple effects in series while maintaining low latency crucial for real-time performance.
Hardware Configuration
A typical YummyDSP effects processor uses the ESP32-S3 or ESP32-WROVER module for maximum performance and RAM. Audio input and output connect through I2S-compatible codec chips like the PCM5102 DAC for output and PCM1808 ADC for input, though many projects use combination audio codec chips like the ES8388 or WM8978 that handle both input and output with a single device.
The audio interface operates at professional sample rates, typically 44.1kHz or 48kHz, with 16-bit or 24-bit resolution. I2S uses three signals: bit clock (BCLK), word select (LRCLK), and serial data (DIN/DOUT). The ESP32’s I2S peripheral handles all timing automatically once configured, leaving your code free to focus on processing audio rather than managing low-level protocol details.
User interface might include potentiometers connected to the ESP32’s ADC inputs for real-time parameter control, rotary encoders for menu navigation, and an OLED display for visual feedback. Some advanced designs add MIDI input for parameter automation or preset switching. The beauty of the ESP32 platform is the abundance of GPIOs allowing comprehensive control interfaces without external multiplexers or I/O expanders.
Creating Effects Chains
YummyDSP’s architecture revolves around processing blocks connected in chains. A simple overdrive pedal might include: input gain → high-pass filter → soft clipper → tone control → output gain. Each block processes the audio buffer in sequence, transforming the signal step by step. The library handles buffer passing automatically, allowing you to focus on the audio processing algorithm rather than memory management.
More complex effects combine serial and parallel processing. A reverb pedal processes the dry signal through early reflections, sends that to a late reverb algorithm, then mixes both with the dry signal for the output. A multi-effects processor might split the signal, sending one path through distortion and another through delay, combining them with adjustable mix ratios.
The real power emerges when dynamically reconfiguring effect chains. Your processor can become a different pedal by loading different effect chain configurations. Store multiple presets in flash memory, switch between them via MIDI program change messages or foot switches, and you’ve essentially built multiple pedals in one enclosure.
Performance Considerations
Audio latency is critical for guitar effects—humans perceive delays above 10-12 milliseconds as distinct echoes rather than immediate response. YummyDSP achieves latency typically under 5 milliseconds on ESP32-S3, well within acceptable limits for live performance. This low latency results from efficient code, DMA-driven I2S transfers, and careful buffer sizing.
Computational budget determines how many effects you can chain. Simple effects like filters and gain consume minimal CPU time, while complex reverbs and pitch shifters demand more resources. The ESP32’s dual cores help—dedicate one core to audio processing while the other handles user interface, WiFi communication, or control functions. This separation ensures audio processing never gets interrupted by non-critical tasks.
Floating-point arithmetic on ESP32 executes quickly thanks to hardware FPU support, but integer math remains faster for less critical operations. YummyDSP uses floating-point for audio path processing (where precision matters) and integers for control logic, striking an optimal balance between audio quality and computational efficiency.
Practical Applications
Guitar pedal designs represent the most common YummyDSP application. Build anything from simple boosters to complex multi-effects units. The programmability means you can update algorithms remotely via WiFi, add new effects without hardware changes, or create parameter mappings impossible with analog circuits. Imagine a chorus where LFO rate syncs to tap tempo, or distortion that automatically adjusts gain based on input level.
Studio effects processors benefit from ESP32’s connectivity. Stream audio wirelessly via I2S over WiFi, control parameters from a web interface, integrate with DAW software through MIDI or OSC protocols. The same hardware that runs your effect also serves web pages for remote configuration—something analog pedals certainly can’t do.
Synthesizer voice cards use YummyDSP for oscillators, filters, and envelopes. Build polyphonic synthesizers where each ESP32 generates one voice, receiving MIDI data over a shared bus. The modular approach scales from monophonic instruments to full polysynths without redesigning software architecture.
Advanced Features
The ESP32’s WiFi enables unprecedented connectivity. Download new effects algorithms from online repositories, back up presets to cloud storage, or control effects from smartphone apps. Some projects implement wireless audio streaming, eliminating cables between instruments and amplifiers. While WiFi audio introduces latency unsuitable for real-time monitoring, it works excellently for recording applications or processing already-recorded material.
Bluetooth MIDI simplifies control without physical connections. Modern MIDI controllers support BLE MIDI, allowing wireless parameter changes, preset selection, and synchronization. Your effect pedal becomes truly wireless—battery powered, Bluetooth connected, and cable-free except for audio.
Machine learning integration opens fascinating possibilities. Train models to replicate specific amplifiers or effects units, then run inference in real-time on ESP32. While complex models exceed ESP32’s capabilities, simpler networks emulating analog circuits run successfully, bringing AI-powered tone matching to affordable DIY pedals.
Getting Started
The YummyDSP library and examples are available on GitHub with comprehensive documentation. Start with simple examples like a clean boost or tone control, understand the code structure, then progress to complex effects. The community actively shares projects, providing inspiration and proven designs to learn from.
Hardware-wise, development boards like the ESP32 Audio Kit provide everything needed—ESP32 module, audio codec, buttons, and audio jacks on a single PCB perfect for prototyping. Once satisfied with your design, create custom PCBs tailored to your specific effect pedal requirements.
Conclusion
YummyDSP and the ESP32 platform democratize professional-quality audio DSP, bringing effects processing that rivaled commercial products just years ago into the hands of hobbyists. Whether building guitar pedals, studio processors, or experimental sound generators, the combination of powerful hardware, efficient software libraries, and active community support makes creating custom audio effects more accessible than ever. The only limit is your imagination—and maybe computational budget.
Explore YummyDSP on GitHub and join the ESP32 audio community to see what’s possible and get started building your own DSP-powered audio projects.