4.5K ·GitHub
What is this?
AudioNoise is a small, playful collection of digital audio effect simulations (phaser-like all-pass filters, delay-based echoes, etc.) written in C by Linus Torvalds as a side project for learning digital signal processing basics, originally tied to his DIY guitar-pedal hardware.
AudioNoise is a small, playful collection of digital audio effect simulations (phaser-like all-pass filters, delay-based echoes, etc.) written in C by Linus Torvalds as a side project for learning digital signal processing basics, originally tied to his DIY guitar-pedal hardware.Is this the actual guitar-pedal hardware code, or just a software simulator?It is a pure software simulation of the effects; the physical RP2354/TAC5112 guitar-pedal hardware project is a separate, currently archived effort, and this repo just models the same digital processing on its own.
It is a pure software simulation of the effects; the physical RP2354/TAC5112 guitar-pedal hardware project is a separate, currently archived effort, and this repo just models the same digital processing on its own.How do I build and run the effects?The repo ships with a Makefile, so you build with 'make'; beyond that, the README doesn't walk through specific usage steps.
The repo ships with a Makefile, so you build with 'make'; beyond that, the README doesn't walk through specific usage steps.Do I need anything beyond a C compiler to try this out?A C toolchain and make are enough to build the effect code itself; there's also a separate Python-based visualizer tool for inspecting the audio samples.
A C toolchain and make are enough to build the effect code itself; there's also a separate Python-based visualizer tool for inspecting the audio samples.Are these anything like modern amp-sim or cab-emulation effects?No — these are intentionally simple toy effects using only IIR filters and basic delay loops with no latency, not fancy FFT-based processing or AI-style amp/cab modeling, and the author explicitly says not to take them seriously.
No — these are intentionally simple toy effects using only IIR filters and basic delay loops with no latency, not fancy FFT-based processing or AI-style amp/cab modeling, and the author explicitly says not to take them seriously.Why build digital simulations instead of finishing the hardware pedal?The author wanted to keep exploring digital effects while the physical guitar-pedal hardware design is paused, so simulating everything in software lets the learning continue without needing the actual board.
The author wanted to keep exploring digital effects while the physical guitar-pedal hardware design is paused, so simulating everything in software lets the learning continue without needing the actual board.What does the Python visualizer tool do?It's a separate helper for visualizing the audio samples/effects; the author admits it was largely built via vibe-coding with Google Antigravity rather than hand-written Python expertise.
It's a separate helper for visualizing the audio samples/effects; the author admits it was largely built via vibe-coding with Google Antigravity rather than hand-written Python expertise.Does the audio processing add any noticeable latency?No — every effect is designed as single-sample-in, single-sample-out with no added latency, mirroring the sub-millisecond ADC-to-DAC latency the author achieved on the real TAC5112 hardware.
No — every effect is designed as single-sample-in, single-sample-out with no added latency, mirroring the sub-millisecond ADC-to-DAC latency the author achieved on the real TAC5112 hardware.What are similar projects?sloop-fm1, coyopedal, ultimatevocalremovergui, AutoEq, Athena-Public, onthespot, vhs-decode, audio-plugin-coder
sloop-fm1, coyopedal, ultimatevocalremovergui, AutoEq, Athena-Public, onthespot, vhs-decode, audio-plugin-coderWhich topics is it in?filtering, vibe-coding, antigravity, audio-processing, simulation, gplv2, music, hardware, audio, c, learning, python
filtering, vibe-coding, antigravity, audio-processing, simulation, gplv2, music, hardware, audio, c, learning, python