A collection of audio presets for EasyEffects (PipeWire).
Looking for PulseEffects presets? Switch to the
pulseffectsbranch and install manually.
Vendored from droidwayin/GentleDynamics (GPLv3 upstream; carried here as data, with attribution).
- GentleDynamics — 8-band Bark-scale multiband compressor for transparent dynamic-range restoration.
- GentleDynamics Feather Loudness — same chain plus a compressor for a gentle loudness lift.
- GentleDynamics Dialogue Clarity Engine — compressors + de-esser + expander to keep film/TV dialogue intelligible.
- Aurora Immersive — hand-tuned pure-DSP chain (EQ → bass enhancer → multiband → exciter → stereo widening → subtle reverb → crossfeed → limiter) for immersive headphone listening. It's a voiced preset, not a reference one — start ~6 dB below your usual volume.
- Cupertino Laptop Speakers — an original, from-scratch pure-DSP loudness + clarity chain (high-pass driver protection → EQ boxiness-cut & presence/air lift → psychoacoustic bass enhancer → 8-band multiband compressor → high-frequency exciter → stereo widening → brickwall limiter) hand-tuned to make built-in laptop speakers sound big, loud and refined, inspired by the tuning philosophy of MacBook speakers. Not derived from any measurement dataset — it's a heuristic engineering voicing. Aggressive and speaker-specific (not for headphones); start a few dB below your usual volume.
Original pure-DSP presets built entirely from EasyEffects' built-in plugins (no IR assets); parameters follow the EasyEffects 8.2.7 plugin schema. Each is a heuristic engineering voicing, not derived from a measurement dataset, and each deliberately exercises plugins the rest of the collection doesn't use.
- Night Listening —
loudness(ISO 226-2023 equal-loudness contour) → multiband compressor → limiter, so bass and treble stay perceptually balanced at low late-night volume. Set the Loudness volume control to your actual listening level. - Levelizer (EBU R128) —
autogainnormalizes any source toward a −16 LUFS integrated target (plus a safety limiter), taming wildly inconsistent track-to-track / video loudness. - Mono Sum (Accessibility) — collapses stereo to mono via Stereo Tools (stereo-base −1), so single-sided-deaf or one-earbud listeners never lose center-panned content.
- Tiny Speaker Rescue — high-pass driver protection → psychoacoustic bass enhancer (implies a sub-bass the driver can't physically produce) → crystalizer transient clarity → limiter, for phone / laptop / small Bluetooth speakers.
- Analog Warmth — high-frequency exciter plus a short, subtle reverb for gentle tonal warmth and air.
- Concert Hall — a large-room zita reverb for spacious, ambient music listening. An effect voicing, not a reference curve.
- Movie Dialogue Boost — presence-lift EQ (~2.5 kHz) → multiband compression → de-esser to keep film/TV dialogue intelligible over music and effects.
- Reference Transparency —
filter(18 Hz subsonic guard, Butterworth Q) →loudness(ISO 226-2023 equal-loudness compensation, defaulted to the −14 dBFS "K-14" home/critical-listening reference) →limiter(−1 dB ceiling, 8x oversampled for true-peak safety, essentially inaudible under normal use). No EQ curve, no exciter/enhancer/compressor/reverb — the philosophy is "the best processing is the one you don't hear": only two additions to a flat pass-through, both backed by measurement standards (Butterworth alignment, ISO 226) rather than taste, plus a safety net that should never audibly engage on well-mastered source material. Note: EasyEffects' Loudnessvolumefield directly sets the plugin's output level (it's not just a curve reference — per the app's own docs it "controls the loudness of the 1 kHz pure sine wave"), so this preset plays back noticeably quieter than a flat-gain preset by design; raisevolumetoward 0 dB (less correction, less attenuation) or turn up your output device to compensate.
- HeSuVi Virtualization Pack (43 presets) — true-stereo IRs adapted from HeSuVi's per-technology HRTF data: Atmos, DTS Headphone:X, GSX, CMSS-3D, Dolby Headphone, DS3D, DVS, Nahimic, Windows Sonic, Sound Blaster SBX (incl. SBX 100), OpenAL (incl. CIAIR & default HRTFs), Razer Surround, Flux HEar and more, with
+/++/+++intensity variants. See Impulse Responses. - EFOtech MLV Pack (6 presets) — crossfeed IRs from Joe0Bloggs/EFOtech_MLV (MIT). The number is the effect depth (256–22000); start low.
- Synthetic Spherical Crossfeed — a mathematically modelled true-stereo crossfeed (rigid-sphere head, Woodworth ITD), generated by
scripts/generate-synthetic-crossfeed.js. - Synthetic Binaural Room — an "advanced" model-based true-stereo kernel generated by
scripts/generate-synthetic-binaural-room.js. Fully closed-form, nothing ripped from vendor software: geometric per-ear delay/ITD from two virtual speakers at ±30°, a Brown & Duda (1998) angle-dependent structural head-shadow shelf, a pinna reflection bank (spectral notches near 8 kHz for externalization), first-order image-source early reflections from a modelled 4.2×3.6×2.6 m room, and a deterministic (seeded) exponentially-decaying diffuse tail (RT60 ≈ 260 ms, DRR ≈ 8 dB). Adds room/externalization over the drier Synthetic Spherical Crossfeed; measured ITD ≈ 250 µs, tail interaural coherence ≈ 0. Deterministic output (safe to commit/validate). - LibreAtmos — an approximation of the Motorola/Dolby DAP ("Dolby Audio Processing") mobile headphone chain (Music profile): EQ + bass enhancer + a short convolver coloration stage + limiter. The EQ/dynamics curve is hand-tuned from DAP's publicly documented behavior; provenance of the convolver kernel itself is unverified — it is currently a short 2-channel stereo IR (not the 4-channel true-stereo crossfeed format used by the other Convolver presets in this pack), so no HRTF/crossfeed spatialization is actually applied yet.
- MIT KEMAR HRTF (SOFA) — the MIT KEMAR measured HRTF (Gardner & Martin, 1995), loaded as
.sofawith no conversion. - ARI HRTF (SOFA) — a measured DTF HRTF of subject NH2 from the Acoustics Research Institute (CC BY-SA 3.0); a flatter alternative to MIT KEMAR.
- IRCAM LISTEN HRTF (Subject 1002) — the IRCAM LISTEN measured HRTF (subject IRC_1002, free for research), delivered as a 4-channel true-stereo
.irsderived from HeSuVi'sirc02kernel (front L/R pair, head-symmetric). Unlike the SOFA presets it convolves a fixed frontal image (no live azimuth/elevation), but it loads directly without libmysofa. - FLORA (Free/Libre Object Reality Audio) — 4 presets (Cinema, Music, and a "full marker" variant of each) built on two custom, self-authored true-stereo HRIR kernels (not derived from any published dataset or measurement corpus). The "marker-stripped" kernel is the default; the "full marker" variant keeps embedded calibration markers in the impulse response. Cinema and Music each apply a distinct hand-tuned EQ curve on top of the shared convolver stage.
The SOFA presets convolve the whole mix as one virtual source — change azimuth/elevation live in the Convolver's SOFA controls.
irs/ holds the convolution kernels (RIFF/WAVE .irs and AES69 .sofa) used by the Convolver-based presets — every kernel here is referenced by a preset. EasyEffects' Convolver only loads mono, 2-channel, or 4-channel "true stereo" (L, LR, RL, R) kernels, so raw multi-channel HeSuVi/HRIR files have to be adapted first — that's how the HeSuVi and EFOtech kernels here were built. They're HeSuVi/EFOtech-derived true-stereo IRs, the generated synthetic crossfeed, the two measured SOFA HRTFs, and LibreAtmos's short 2-channel coloration IR; each is trimmed of leading/trailing silence to cut Convolver CPU and latency.
More community IRs to experiment with: icy-comet/EasyEffects-presets-impulses.
bash -c "$(curl -fsSL https://raw.githubusercontent.com/M0Rf30/easyeffects-presets/main/install.sh)"Press Enter for the default. The on-screen menu is authoritative if this table drifts:
| # | Installs |
|---|---|
| 1 (default) | Everything (all presets + all IR/SOFA files) |
| 2 | HeSuVi virtualization pack (43 presets) |
| 3 | Synthetic Spherical Crossfeed |
| 4 | EFOtech MLV pack (6 presets) |
| 5 | MIT KEMAR HRTF (SOFA) |
| 6 | ARI HRTF (SOFA) |
| 7 | GentleDynamics pack (3 presets) |
| 8 | Aurora Immersive |
| 9 | Cupertino Laptop Speakers |
| 10 | IRCAM LISTEN HRTF (Subject 1002) |
| 11 | Utility & Effects pack (8 presets) |
| 12 | Synthetic Binaural Room |
| 13 | LibreAtmos |
| 14 | FLORA pack (4 presets) |
curl is required (sudo apt install curl on Ubuntu).
Copy the .json files into your EasyEffects output/ directory — ~/.local/share/easyeffects (native) or ~/.var/app/com.github.wwmm.easyeffects/data/easyeffects (Flatpak). For Convolver presets, also copy the referenced .irs/.sofa (see each preset's kernel-name) into the sibling irs/ directory.
Run ./scripts/validate-presets.sh before opening a PR — it checks that every .json parses and that every kernel-name resolves to a file in irs/. CI runs the same check.