Cavern @0dd0de8[2026-03-22]

Commits

• 0dd0de8: Significantly increased imaging performance (VoidX)

Cavern 2.1 - The long-awaited update

Downloads

Cavern Filter Studio

The Cavern library is now ready to handle configuration files like Equalizer APO in a tree, modify them graphically, optimize, and export to different formats.

New audio formats supported

• Meridian Lossless Packing (TrueHD Atmos) with Truehdd
• Core Audio Format
• Dolby Atmos Master Format

Driver additions

• Added upmixers to the Demo Player

QuickEQ additions

• Added different measurement graphs with display settings
• Added MiniDSP Flex HTx, JL Audio TüN, Roon, Wavelet, and WiiM support
• Added automatic crossover point detection
• Added faded display of individual measurements in multi-mic mode
• Added export of measurement signal after corrections
• Added target device selector for all devices supported in the API
• Added messages for successful operations
• Added single measurement removal for QuickEQ Home
• Added measurement file export for QuickEQ Home
• Added spectogram and phase curves to QuickEQ Home
• Added an option to filter channels on QuickEQ Home graphs
• Added an option to remove mains hum
• Added an option to export curves as CSV files
• Added an option to always save the measurement next to the exported configurations

QuickEQ changes

• Completely redesigned Multi-sub handling
• Room curves are now configurable with bass rise, knee, and treble suppression
• Room gains are now automatic by default
• Sealing is burned into FIR filter exports when possible

Cavernize additions

• Added speaker virtualization
• Added AC-3 and Enhanced AC-3 target codecs
• Added --unsafe hidden argument to bypass some codec restrictions
• Added -track and --extract arguments to extract raw bitstreams
• Added 4.1.3 channel layout
• Now downloadable as an installer

API additions

• Added Acurus Muse, AU N-Band EQ, MiniDSP DDRC-88A, Monoprice Monolith HTP-1, Sony ES series, Tonewinner AT-series, Trinnov, and Yamaha RX-A series support
• Added block-by-block filtering of complete signals
• Added channel layout graphics drawer
• Added generic matrix mixers with phase shift support
• Added linear fading
• Added phase swapping for allpass filters
• Added pink noise generators
• Added polarity corrections
• Added Spleeter output folder support

API changes

• Heavy FFT optimizations
• Updated Samples to .NET 8

Cavern 2.0.2 (Performance²)

The last few commits have improved performance so much that it's worthy of an early release.

Cavern 2.0.1 (Performance)

The last few commits have improved performance so much that it's worthy of an early release.

Cavern 2.0 - Long-term Support update

Finalized feature set and API. From now on, function signatures are guaranteed to stay the same. Only new features will be added when neccessary (like new codecs), existing ones will not be modified or removed.

Library additions

• Added MKV container writing
• Added MP4 container reading
• Added LAF environment export
• Added crossover export for Equalizer APO, supporting all of Cavern's basic crossovers
• Added FIR/IIR filter set export for supported software/hardware, including CamillaDSP
• Merging of multiple E-AC-3 files for channel-based spatial mix storage
• Microphone support for WebGL

Cavernize additions

• Added a metadata viewer
• Added 4.1.1 channel layout
• Content grading
• New render target selector
• Option to swap side/rear channels
• Support opening webm/weba files

Cavern 1.6 - Happiest Halloween update

This is the largest update to Cavern yet, including the world's first open-source Dolby Digital Plus Atmos decoder on the PC.

Driver changes

• E-AC-3 + JOC (Dolby Digital+ Atmos) visualization and rendering
• ADM BWF visualization and rendering
• A welcome screen is shown on first launch
• Custom channel names which also appear in QuickEQ
• Speaker layout auto-detection on first launch
• File browser dialogs are now Mono library popups that look like a Windows file browser
• Larger, more accessible UI

QuickEQ additions

• Added Expert crossover for complete cutoff frequency and subwoofer matrixing control.
• Added MultEQ-X support
• Added one-click Equalizer APO export
• Added the Smoother curve which keeps the system's character and does not require a dedicated measurement microphone
• Added a custom gain standard, where channel gains can be offset between -6 and 6 dB with sliders or manually without bounds
• Multiple microphones can be selected for measurements, these measurements are automatically added to the Multi-mic panel
• Default calibration files are automatically applied on MiniDSP UMIK-1 and XTZ II Pro microphones
• Arbitrary measurement signals can be used
• Warm-up mode against microphone clicking on measurement start
• Warning messages when channel gains are too far apart
• Error message when no microphone is connected

QuickEQ changes

• Measurements are now exported in Limitless Audio Format, which can contain complete multi-mic measurements
• Minimum corrected frequency is now allowed down to 10 Hz
• Delays and polarities are calculated after simulation for convolution exports, making them superior to any other export
• Delays smaller than half the filter size are baked into convolution filters
• After overwhelming user feedback, Cavern is now the default gain standard instead of DCI
• Filter sizes other than powers of 2 are now supported for convolution exports
• Multi-file exports can be renamed, no more folder selection

API additions

• Decoding of Matroska (.mkv/.mka) streams
• Decoding of E-AC-3 (.ac3/.ec3) streams, including JOC (Dolby Digital+ Atmos) into objects
• Decoding of ADM BWF (.wav) streams into objects
• CavernAmp accelerates some features with native code on 64-bit Windows for an average of 2x speedup
• Ear canal simulations by angle and distance with the Distancer filter
• Optimized convolution for basic impulse responses called Spike convolver
• Screen-locked source flag
• Many new waveform and audio file handling functions

API changes

• File reading is separated to reading, decoding, and rendering
• Linearized virtualizer frequency responses
• Moved API to .NET 6 and made NuGet-ready

Cavern 1.5 - Nearly 5 years update

Additions

• Valley correction in QuickEQ
• Gain and Cavernize filters
• Phase, RT60, and imulse likelyness for impulse responses
• 24-bit support for known formats, WAV reader
• FRD calibration support

Changes

• Impulse responses have way less noise
• Versatile convolution generation
• Headphone virtualization is available outside Unity
• Many refactors

Cavern 1.4 - Standalone update

Additions

• Cavern is no longer bound to Unity, separated to module DLLs, CavernUnity.dll is only an adapter
• Completely separated rendering to Unity's audio thread
• New, abstract, and complex filters
• Sources can be filtered individually
• EQ curves
• Zero delay (minimum phase) or linear phase convolution EQ generation

Changes

• The center channel now has an echo in virtualization

Cavern 1.3 - HRIR and Cavernize Lite update

Additions

• HRIR-based headphone virtualizer
• Cavernize Lite is now open source and included in this repo
• Variable band smoothing
• Implemented more Unity functions
• Support for virtual, even streaming audio sources with forced playback
• Noise generator audio source

Changes

• Greatly optimized code
• Virtualization does not override echo settings for audio sources
• The default sample rate is now 48 kHz (which is the sample rate of the HRIR)
• New set of supported DCPs

Cavern 1.2 - QuickEQ update

Additions

• QuickEQ: advanced room correction toolkit with extremely quick and accurate measurements
• Audio spoofer: converts Unity's listeners and sources to Cavern components
• Log display debug window

Changes

• Reworked resampling
• All C# code is now managed