rubyk

Just a few thoughts on openCL, LLVM and garbage collector from Lua.

New MidiIn object, Lua scripts can receive midi events.

I am glad to announce that the rubyk project has received 15’000 CHF funding from the Swiss Federal Office of Culture to write the interface !

This will allow us to start working on the project. I am still looking for more funding (budget for the full interface is about 34’000 CHF)... If you know a generous donator or want to help, please announce yourself to the mailing list.

If you have strong competence in interface programming and C++, I am looking for a second programmer to help me in this task. Preferably around Lausanne (CH), so we can work in the same place…

We have started working on an interface for rubyk. This article gives some ideas on the futur interface.

GLUT wants to be in control and hates being pushed in an obscure thread somewhere.

Turing object now fully supports Lua scripts with a pre-filter (custom bang method) and has good debugging facilities with table printout and dot graphs.

I just created a drawing with the mapping of midi note values and keyboard. Could be useful to anyone doing low-level midi work.

Recognize a vector from belonging to a set using the Euclidean distance to the mean value.

By creating a session with the “Midi setup” tool, you can use it as a destination for your midi.

Worked all night to go through a massive cleanup of the code, replacing all uses of buffers of doubles by Matrix objects.

There must be a bug somewhere.. PCA object & Svm object working hand in hand should produce better results then the tiring sound .. hey no-sound (got you) thing. My back hurts (too much taping with my foot).

Finally, I found the courage to implement Principal Component Analysis using BLAS and LAPACK libraries. The top row of the partition shows the recording from the recognition of a foot beat: we have a long way to go before we can really play music, but at least it recognizes up and down moves.

Appart from creating stupid skylines, my implementation of short-time Fourier transform is useless.

Tutorial on Hidden Markov Models. I wrote this text to help me understand the subject before implementing it for rubyk.

After the first filtering stage we need to further compress our data (reduce it’s dimensionality). For this purpose, we use Vector Quantization.

Fast Fourier Transform (change the signal from the time domain to the frequency domain).

Real-time signal recognition using Support Vector Machine !

Seeing the signals live really helps ! Plot plot plot.

Just a note to inform we have added new objects and changed the code to use virtuals instead of functors for the main method (bang).