Releases: basilica-audio/Miserere
Release list
v0.3.0
Miserere v0.3.0 — preset system, factory presets, German localisation
Adds the suite-wide M2 preset system (PresetManager/PresetBar, ported from basilica-audio/nave's pilot implementation): factory and user presets, save/save-as/rename/delete, next/previous, single-file and zip-bank import/export, and a "Default"-preset resolution order. Ten factory presets ship in the box (Default, Classic Parallel Blend, Crush Forward, Silk Sandwich, Wide & Wet, Direct Channel Only, Gentle Bus, Rough Mix Glue, Whisper Thicken, Aggressive Rock Vocal — see docs/presets.md for each one's intent). The preset bar's frame strings (buttons, menus, dialogs, error messages) ship with a German translation (resources/i18n/de.txt); parameter/DSP terminology is never translated. 122 Catch2 tests green (20 new).
Assets: the Windows build is attached. The signed macOS build failed again at the "Import signing certificate" step (known organization-level signing-secret issue, not specific to this release — see run 29476155353, job Signed macOS release / 87549484862) and will be attached in a follow-up run once that secret is fixed.
v0.2.0
Miserere v0.2.0 — the parallel vocal template, rebuilt
v2 replaces v0.1.0's topology entirely: the dry vocal now passes through a bit-transparent direct path at unity by default, and everything else — a FET crush bus, a Passive EQ / Opto Leveler / Passive EQ "sandwich", a dual micro-pitch spread, and a single-repeat dark slap — is layered underneath it via four parallel return busses with true-zero faders, Mute and exclusive Audition. v0.1.0 processed the direct signal by default and treated it as one bus among four, which inverted the technique this plugin packages; v0.2.0 implements the documented 2010–2023-era parallel vocal workflow as it is actually described in public interviews. Parameter IDs are breaking (pre-1.0); a v0.1.0 session loads without crashing, with old IDs silently ignored.
Honesty note: the voicing throughout is research-derived — built from public documentation, manuals, interviews and community measurement (see docs/research-notes.md in the repo) — not measured against hardware units.
Factory presets arrive with the M2 preset system.
Assets: the Windows build is attached. The signed macOS build is pending an organization-level signing-secret fix and will be attached in a follow-up run.
Miserere v0.1.0
Miserere is the parallel vocal chain of the Basilica Audio suite — sacred-architecture DSP for heavy music: a direct chain plus three parallel busses, each with its own fader, packaging the classic "rough vocal template" workflow as a single unit. Existing channel strips run serial; Miserere's identity is true parallel routing.
This is the first tagged release (v0.1.0), covering the M1 milestone: core DSP and test coverage.
What's in 0.1.0
- Bus A — Direct: HPF, console-style 3-band EQ, FET compressor (4:1/8:1) with gain-reduction metering, split-band de-esser (4–9 kHz), tape saturation with pre/de-emphasis.
- Bus B — Opto sandwich: passive-style boost-only EQ (60/100 Hz low, 8/10/12/16 kHz high) into an opto leveler with a genuinely program-dependent two-stage release (60 → 600 ms with gain-reduction history), plus a post air shelf.
- Bus C — Smash: FET limiter in all-buttons character — ~20:1, program-dependent release shortening, mid-forward detector tilt, drive.
- Bus D — Slap: 60–180 ms fractional delay (default 110 ms), filtered tape-soft feedback loop, mono switch.
- Per-bus fader/mute and exclusive solo; busses A–C are sample-aligned (minimum-phase only) so parallel summing never combs; zero latency.
- 65 Catch2 tests: bit-transparent direct-bus null, parallel impulse alignment, compressor static curves, opto two-stage release, de-esser selectivity, slap timing and feedback stability, NaN/Inf recovery, oversized blocks, state round-trip. pluginval (strictness 10) + auval green in CI.
- Full user manual, engineering notes and ADR-0003 on the parallel-bus topology.
Honest status notes
- Voicing: the modules implement the documented behaviors of the classic hardware categories (generic, no brand emulation claims) — a measured-reference + by-ear voicing pass follows in M2 (#4).
- Presets: the suite-wide preset system (factory presets, user presets, default, import/export) is scheduled for M2 (#3).
- GUI: functional slider editor; the photoreal skeuomorphic interface lands in M3.
Formats
| Platform | Formats | Notes |
|---|---|---|
| macOS 11.0+ | AU, VST3, Standalone | universal (Apple Silicon + Intel) |
| Windows x64 | VST3, Standalone |
Install by copying AU/Miserere.component to /Library/Audio/Plug-Ins/Components and/or VST3/Miserere.vst3 to /Library/Audio/Plug-Ins/VST3 (macOS) or C:\Program Files\Common Files\VST3 (Windows). Rescan plugins in your DAW after installing.
Signed & notarized
The macOS binaries in this release are Developer-ID-signed, notarized by Apple and stapled — they install and open without Gatekeeper warnings. (Windows binaries remain unsigned for now; Authenticode signing is tracked as roadmap milestone M4.)