October 20th, 2026 — Audio, General
A new release, liquidsfz-0.4.2 is now available. There were some fixes in the file dialog in order to make it work with bundled Ardour installations (from a payed linux Ardour version). Also the LV2 plugin now shows a midi keyboard, so key switches become more intuitive to use.
An AI code review was done, which found quite a few issues that were not found during testing until now, and many of the easy to fix/most problematic issues are now corrected with this release. This is the first time I’ve done this for a project and tried to fix everything that the AI said would be problematic. That things were not found until now usually meant that they were unlikely to happen in real use cases most of the time, because otherwise we would have seen the problem before.
Sometimes you needed “bad” .sfz files to trigger the issue, so maybe it would never happen because nobody used liquidsfz with a “bad” .sfz file, but if they ever did, they could crash their DAW or stall the audio thread, so it is probably a good idea to fix it although it never happened until now.
However for other cases, even with “good” .sfz files we could trigger a crash or a deadlock sometimes, or have the audio thread use non-realtime safe behaviour (which could cause dropouts) or cause undefined behaviour (which may become a problem later on). That these are extremely unlikely doesn’t mean that we’re good, because we never want this to happen ever, and here the fact that the AI looks at the code directly and doesn’t just test means that we see problems that we previously were not aware of.
Of course, sometimes there are false positives, where AI sees problems that literally don’t matter and are merely theroretically problematic, for these it is better to keep the code as it is.
August 11th, 2026 — Audio, General
A new release of gst123, my commandline media player based on GStreamer is available. The main new “feature” is probably that it now can play videos without crashing on Wayland, although it still relies on Xwayland to do so. See release link for full list of changes.
April 20th, 2026 — Audio
I didn’t have time yet to blog about liquidsfz-0.4.1, which was released two weeks ago, so here is a quick overview of the bigger changes.
The .sfz parser was made more robust, which means that broken .sfz files (there are some files out there which load fine in sforzando but have questionable syntax) can be loaded with a best-effort strategy. These files now produce warnings instead of an error.
Three different problems were fixed that could in some situations cause audible clicks, so updating from a previous version is recommended. A few smaller fixes (and two new opcodes) improve compatibility with more .sfz files. Finally, some improvements were made to the LV2 plugin.
March 13th, 2026 — Audio, General
The main goal of liquidsfz is to implement a library that supports playing .sfz files and is easy to integrate into other projects. We also provide a JACK client and a LV2 plugin.
A new version, liquidsfz-0.4.0 is now available.
The release adds support for parametric equalizers and some other new opcodes. It implements some extended CCs and generators (like sample=*sine), as well as parsing and loading programs from AriaBank .bank.xml files. A custom UI for the LV2 plugin was added to be able to select AriaBank programs in the LV2 plugin.
For a full list of changes, see the github release.
January 10th, 2026 — Audio, General
A new version, SpectMorph 1.0.0-beta3 is available at www.spectmorph.org.
SpectMorph (CLAP/LV2/VST plugin, JACK) is able to morph between samples of musical instruments. A standard set of instruments is shipped with SpectMorph, and an instrument editor is available to create user defined instruments from user samples.
The new features of the 1.0.0 beta releases (compared to the latest stable version) are described in a YouTube Tutorial.
In the beta3 version, the instrument editor has a new pitch detection algorithm and support for mp3 files. Other than that, there were many smaller fixes, some of them addressing critical problems, so we recommend updating.
If you are interested in a detailed list of changes, you can look at the NEWS file.
July 30th, 2025 — General
Imagewmark is an open source solution for embedding invisible digital watermarks in images. The watermark is designed to survive common operations like lossy compression (jpg) or crop + resize. It is uses forward error correction for robustness, AES key dependent watermark encoding and supports blind decoding (the original image is not needed to detect the watermark).
The source code of the first public release is now available: Imagewmark 0.5.0 on Github
July 5th, 2024 — Audio, General
Anklang is a digital audio synthesis application for live creation and composition of music. It runs on Linux and is Free Software.
I’ve created a small YouTube Video for the latest 0.3.0 release. You can watch this if you are curious about how Anklang looks and feels like.
Besides some cleanups and bugfixes, the 0.3.0 release contains
- a new SFZ based sampler device called LiquidSFZ, which also supports Hydrogen drumkits
- a new Saturation device
- improvements to the BlepSynth device (our two oscillator subtractive synth)
- various UI improvements
The full list of changes and the release itself can be found on the GitHub Release Page.
June 27th, 2024 — Audio, General
A new version, SpectMorph 1.0.0-beta1 is available at www.spectmorph.org.
The new features of this release are described in a YouTube Tutorial.
SpectMorph (CLAP/LV2/VST plugin, JACK) is able to morph between samples of musical instruments. A standard set of instruments is shipped with SpectMorph, and an instrument editor is available to create user defined instruments from user samples.
The first lines of code for SpectMorph were written in 2009. I also worked full time on SpectMorph for my university thesis. There were many incremental improvements over the years, so at this point, the next stable release of SpectMorph will be 1.0.0. I’d like SpectMorph-1.0.0 to be as stable as possible, so if you test the beta1 version and have any issues with it, please let me know.
In this release, two formant preserving algorithms for repitching vocals were added to the WavSource operator. This means that for user defined vocal instruments, good sound quality over a large midi note range is possible with only a single sample as base. Also with these algorithms enabled, portamento will dynamically adjust the spectral envelope, which means that it sounds more natural.
There are a lot of instruments which have good sound quality only for a certain pitch range. For instance, a trumpet will sound great for high notes, but for low notes, a tenor trombone or bass trombone sounds better. The same is true for bass, tenor and soprano voices. A new KeyTrack operator was added to handle these cases by combining multiple instruments by midi note range. Note that this is not a simple key split, because slopes can be used to morph between the instruments for seamless transitions.
Finally, SpectMorph now supports arbitrary user defined curves. A new Envelope operator was added for modulation envelopes with user defined curves, and the LFO operator also supports them.
If you are interested in a detailed list of changes, you can look at the NEWS file.
If you are interested in even more details, some SpectMorph related issues were discussed in the regular Anklang/SpectMorph developer meetings.
June 9th, 2024 — Audio, General
This is a header only library for C++ which implements fast factor 2, 4 or 8 upsampling and downsampling based on SSE instructions. I’ve developed the code for my DSP code in Anklang and SpectMorph.
It might be useful for you if you have some DSP loop which needs to be oversampled to avoid aliasing.
This is the initial public announcement of the library, please let me know if you have questions or need clarification in the API documentation.
November 2nd, 2023 — Audio
A new version, SpectMorph 0.6.1 is available at www.spectmorph.org.
SpectMorph (CLAP/LV2/VST plugin, JACK) is able to morph between samples of musical instruments. A standard set of instruments is shipped with SpectMorph, and an instrument editor is available to create user defined instruments from user samples.
The instrument editor was improved, the main change is that the volume of the individual samples that are part of an instrument can now be adjusted individually in the new “volume editor” if automatic volume normalization doesn’t work as expected. There were also some smaller usability improvement for the instrument editor.
One limitation of the previous releases of SpectMorph was that there was a fixed set of 128 user defined instruments. To overcome this limitation user defined instruments are organized in banks now, so you can have an unlimited number of banks each containing 128 instruments.
New standard instruments were added, namely Bass Flute, Soprano Saxophone, Clarinet, Bass Clarinet, Tenor Trombone, Viola and Double Bass. The samples and meta information for standard instruments are now available on github.
The code is now hard RT capable, every call of malloc() in the DSP thread was refactored and removed. Finally, a few UI problems on macOS were fixed.
A full list of all changes can be found here.