opal is a C11 implementation of the Yamaha YMF262 (OPL3) and its OPL2 predecessor — the chips behind AdLib and Sound Blaster cards. The core covers the full register set: two- and four-operator FM, all eight waveforms, tremolo/vibrato LFOs, percussion/rhythm mode with noise generator, timers, and stereo routing including the hardware pipeline delay between left and right channels. Sample-for-sample stereo output matches Nuked-OPL3 at the chip's native 49716 Hz, with internal resampling to any target rate.
The public API is eight functions. The chip instance holds no internal pointers, making it trivially copyable as a complete save state. Registers can be written immediately or queued with buffered timing for closer hardware fidelity. A bundled example player streams DRO v1, HSC, IMF, and WLF files to audio via miniaudio, and a WebAssembly build runs the same core in-browser.
Builds as a static library via CMake on Windows, Linux, macOS, Android, and iOS. The emulation core is public domain (Shayde of Reality/OpenMPT lineage); the surrounding C API and tooling are MIT.
16bits-audio-mcp is a Zig-based MCP server that exposes game audio generation as tools callable by Claude. It produces 16-bit PCM WAV files using FM synthesis, reverb, ADSR envelopes, and multi-track mixing, all from the Zig standard library with zero external dependencies.
It covers three asset types: BGMs in 15 styles (adventure, dungeon, boss, cyber, horror, etc.) with 4-track arrangements; 20 sound effect types (jump, coin, explosion, laser, etc.) with pitch and volume control; and 12 jingle types for common game events like stage clear and game over. Scales, tempo, key, and a deterministic seed parameter give reproducible iteration.
A post-processing tool (wav_fx) chains effects such as reverb, delay, bitcrusher, chorus, and distortion, and a mixer tool combines multiple WAV files. Additional tools handle direct note/FM synthesis, WAV info, and playback. The workflow is prompt-driven: describe what you want to Claude, and WAV files are written to disk.
gion (擬音) is a chiptune/sfxr-style audio workbench and Go library for generating retro 8-bit sound effects and music. Rather than shipping WAV files, you author small deterministic parameter recipes in a .gion document (a Filo s-expression format); the same seed always produces the same samples on every platform. It can render at runtime (under a millisecond for effects, a few dozen milliseconds for music loops) or bake WAVs at build time via the gion-render CLI.
The interactive workbench lets you sculpt effects across seven families (pickup, laser, explosion, powerup, hit, jump, blip) with controls for waveform, envelope, frequency slide, vibrato, arpeggio, low-pass, and bit crush. Music generates perfectly-looping chiptune tracks across six moods (upbeat, heroic, dark, chill, battle, boss) with per-instrument mixer and mute switches. A 3D spectral waterfall updates live as you drag sliders; a plain 2D waveform view is also available.
The Go library API lets you load a .gion document and render effect parameters to []int16 samples. gion.Mutate derives slightly varied siblings of an effect from a seed, preventing repeated identical sound playback with zero asset overhead. Built on Ebitengine; a browser-based workbench is also available. Not a port of sfxr — written from scratch in Go.
retro-sfx-gen is a zero-dependency Python 3 script that procedurally synthesizes retro/chiptune-style sound effects from scratch using oscillators, pitch glides, arpeggios, ring modulation, filtered noise, and envelope shaping. Output is 44.1 kHz / 16-bit mono WAV. No recordings, no pip installs — just the standard library.
Covers 16 effect categories (coin, jump, laser, explosion, powerup, hit, ui_click, alarm, pickup, door, teleport, engine, blip, whoosh, land, shield), each with its own synthesis recipe. Generation is fully deterministic via CRC32-seeded RNG, making it suitable for reproducible builds and git-tracked audio pipelines.
A companion QC script validates every output file for correct format, duration, peak level, DC offset, silence, and click-free edges. Generated audio is royalty-free and commercially usable with no attribution required.