Installing with ubuntu fiesty was relatively easy. The biggest problems were not related to clam: getting the nvidia to work and I also had a mysterious problem with the audio that seemed to go away by itself one day. I also tried installing clam on my Gentoo desktop, but I didn't get it working. It could be due to the fact that I tried to install qt4 instead of just the libraries (gentoo doesn't like having more than one version of a given program at the same time) due to not reading the directions closely. The next step in getting it working was figuring out what jack is used for...
Playing with the vowel synthesizer
Meeting Xavier last week was very helpful. I found out that my idea to use glottal pulses as an input for the vowel synthesizer was easier than I thought. I didn't need to write any code and most of what I needed was already in the experimental vowel synthesizer made by David Garcia. Learning how to use Jack was the missing link that let me play the electroglottogram data (http://sail.usc.edu/~kazemzad/EGG/only_source/ , this is a measurement of the opening and closing of the vocal folds from the USC phonetics lab) in Xmms and route it to the tweaked vowel synthesizer (http://sail.usc.edu/~kazemzad/EGG/clam/experimentalVowelSynth_from_EGG.clamnetwork , here I disconnected the harmonic peak generator and rerouted the peaks from the SMSAnalysisCore0 processing to the vowel resonator). It sounds a bit more natural than the original, but this may be more b/c of the natural pitch and timing than the actual glottal pulse info (the EGG records the opening and closing, not the actual sound input to the vocal resonator).
Another thought for a possible improvement would be to reorient the f1 f2 space so it matches the typical vowel diagram (and maybe put such a diagram in the background). This will probably involve real programming...
Here I'll put planning info