Algunas imágenes del xilofón etéreo

Me acaban de mandar algunas imagenes de la semana de la computación

No se ve demasiada gente usándolo, pero bueno, algo es algo =p

El xilofón etereo

Hace poco fue la semana de la computación y presenté el Xilofón etereo. Dado la repercución que tuvo, me pareció que sería interesante dejar un post sobre cómo funciona.

Todo lo que necesitas es una computadora con una webcam y algo de un color distintivo, por ejemplo, el verde.

El programa lee foto por foto lo que registra la camara, y descarta todo lo que no sea del color especificado. Por ejemplo, supongamos que estoy parado delante de la camara con algo verde, lo que el programa mostraría es esto:

El puntito verde del medio, da cuenta de que el programa está detectando el objeto. Lo que en realidad la computadora está viendo en este ejemplo es esto:

Mucho más sencillo, ¿no?. A esto que la computadora ve, le calcula su centro de masa - el punto que está en el medio - y si ese punto ingresa en las zonas de abajo del campo visual de la cámara, sonará una nota dependiendo del cuadradito al que entre.

¿Querés probarlo en tu compu? ¡Acá tenés el código!

The final step

At last it all seems to be working!!

I have launched the last experiment for my thesis, I ran a t-test and it gave me a nice p-value!

I don't want to publish results until I finish the experiment, but there are a lot of good news!!

Stay tuned and if you didn't do the experiment and you're a spanish speaker... what are you waiting for??

What is Cognitive Computational Music?

Since the first time I started this project, when someone asked me what was this about, I had to make an explanation of at least two sentences. It looked something like this:

"I have been reading a lot about music cognition. Then, using that cognitive theories I built statistical/generative models in the hope of getting interesting feedback to the original theories"

After some time passed, I started to reach a resumed version of that two sentences into only one sentence: the one you see at the top of this blog.

It is still to long as a short way to describe this project: It doesn't fit in a title like the one you see in this post =).

So now you know what Cognitive Computational Music is.

Lot of work

I lost my will to write here on the last months because I thought there were no great advances, however, I've realized it's not like that.

I've reached an experimental phase: I'm testing my models making people to rank them. The webpage is in spanish, it will be available in english soon. If you want to take a look here is a link: http://lacompositora.elsonidoq.com.ar.

The first experiment is finished and now I'm analyzing the data. One interesting thing, that also looks cool is a way I discovered to graph the probability distribution of a sucession of three notes (two melodic intervals) following Eugene Narmour's Implication-Realization theory.

He establishes some criteria on how we percieve melodic contours, and using that I made a probability distribution.

Within his theory he establishes that there are some criteria on how we organize the melodies. These criteria are cultural-independent, but also he says that breaking some of these rules have some cultural and aesthetic effects.

It's interesting how the probability distribution that my program learns when I serve as input different musical styles are kind of similar, but also different.

Here is one of Tom Jobin (A Felicidade), one of Scott Jopplin (Ellite Syncopations) and one of Beethoven (a dance) in that order. The Y axis indicate the length of the first interval, and the X axis indicates the length of the second interval.

Tuning up the parameters

Como pretendo ser Licenciado en breve, estuve jugando con los parametros de mi programin para ver que tipo de experimentos puedo hacer para la tesis.

Habia uno en particular que me habia quedado muy alto, y ese parametro regula el nivel de variabilidad que hay en la frase. Cuando lo baje, todo empezo a sonar mejor!!

Escuchad!!

Esta es la original esta es con solo

------------------------------------------------------------------------------------------------

Since I wanna be Lic. Pablo in a short time, I started to think what kind of experiments can I do, so I started to play with all the parameters and listen all sort of midi outcomes.

I had a parameter that controls the ammount of variability really high, when I decreased, it started to sound better.

This is the original, this is with a solo

Listen!

Ok, let's do it in english

First of all, I don't wont you to expect nice english writing here, I just can't do it =D.

I have been posting in this blog the evolution of my grad thesis. I haven't finished it yet, I'm working on it, so by 2010 I became Lic. Pablo hahaha.

If you can read spanish, you'll notice that I didn't tell much about how did a came with a program that composes music. That is because it is getting me really hard to write about it in spanish so I graduate myself, so imagine in english... no way!

Anyway, I will upload everything, and free the source code (written in nice Python =D) as soon as I finish the thesis.

Stay tuned!

Pablo