Show HN: What if your synthesizer was powered by APL (or a dumb K clone)?
This project unveils k-synth, an experimental web-based synthesizer that harnesses a minimalist, K-inspired array language to intuitively sketch audio waveforms. It stands out by merging niche array programming with audio synthesis, offering a novel approach to sound design and sample generation. The developer also utilized AI for significant portions of the project's bootstrap, appealing to HN's interest in both unconventional tech and AI-assisted development.
The Lowdown
k-synth is an experimental web-based synthesizer built by 'octetta' to explore whether a minimalist, K-inspired array language could offer a faster and more intuitive method for sketching audio waveforms compared to traditional coding. This project, which is currently in a "will-it-blend" state, aims to provide a compact way to generate samples for larger audio projects rather than replacing a full Digital Audio Workstation (DAW).
Key aspects of k-synth include:
- Language Design: It features a simplified, right-associative array language, inspired by K, using concise symbols (e.g., 's' for sine, 'p' for pi) to manipulate and generate sound.
- Web Toolkit: The entire experience is delivered as a web-based toolkit, leveraging WebAssembly (WASM) and the Web Audio API for live-coding and instant auditory feedback directly in the browser.
- AI-Assisted Development: The creator employed AI agents to bootstrap the parser and web boilerplate, significantly accelerating the development process and allowing for language design vetting in weeks instead of months.
- Interactive Demo: Users can quickly try out the system by selecting a patch like 'dm-bell.ks', running it, and then playing the generated waveform through a performance grid that allows for melodic playback.
The project actively seeks feedback from the array language and DSP communities regarding its operator choices and the right-to-left evaluation logic, highlighting its open-ended and experimental nature.