Hi FXSound team and community,
I wanted to introduce a project I’ve been working on called AudioTune, because FXSound has become an important part of its technical foundation — and I thought there might even be some potential for collaboration in the future.
AudioTune is a hearing-personalized audio correction system.
The basic idea is to go beyond a conventional equalizer or a fixed headphone profile. AudioTune measures the user’s individual hearing across a wide frequency range and creates a personalized left/right hearing profile from those results.
That profile is then converted into a real-time DSP correction curve intended to compensate, as far as reasonably possible, for individual frequency-dependent hearing differences.
The project currently includes things such as:
-
Individual hearing measurements for the left and right ear
-
Automatic generation of personalized correction curves
-
Separate left/right DSP processing
-
Correction intensity adjustment
-
Fine-tuning on top of the measured correction
-
Presets and A/B comparison
-
Protection against excessive gain and clipping
-
Real-time visualization of the resulting correction
-
A Windows desktop application designed to make the whole process accessible without requiring knowledge of EQs or audio engineering
Where FXSound comes in
For the actual audio-processing side, I have integrated the FXSound core into AudioTune.
Instead of trying to reinvent an entire mature audio-processing stack, AudioTune essentially adds another layer on top: the hearing measurement, analysis, correction-generation logic and user experience determine what should be corrected, while FXSound provides an extremely useful foundation for actually processing the audio.
That combination has worked surprisingly well.
I really like the philosophy behind FXSound, particularly that high-quality system-wide audio processing can remain accessible rather than being locked behind expensive proprietary solutions.
We also got it running on Android
One experiment went a little further than I originally expected:
We have now managed to get the DSP approach running on Android using Shizuku.
This is still very much alpha software and there is plenty that needs to be cleaned up, tested and made robust. Android obviously presents a completely different audio environment compared with Windows.
Nevertheless, the important part is that the concept works: we can apply the processing system-wide on Android without requiring a traditionally rooted device.
That opens up some very interesting possibilities for a future mobile version of AudioTune — and potentially for FXSound-related technology on Android in general.
Why I’m posting here
Since FXSound is becoming such an important part of AudioTune, I didn’t want to simply build around it quietly without reaching out to the people behind it.
I would be very interested in talking to the FXSound developers about the project.
There may be areas where cooperation could make sense — whether that is technical exchange, improving the integration, discussing the Android work, or exploring whether some of the hearing-personalization ideas could be useful to FXSound itself.
I’m not coming with a predefined business proposal or expectation. I mainly wanted to show what I’m building, explain how FXSound is being used, and open the door for a conversation if the team finds the project interesting.
I’d also be very happy to share more technical details, screenshots, architecture information or results from the Android implementation if there is interest.
And of course: thank you for making FXSound available and continuing to develop it. It has allowed me to focus much more of my time on the hearing-correction side of AudioTune instead of having to build every part of the audio stack from scratch.
I’d love to hear what you think.
GitHub Link:
Best regards,
Sven
