Oh, brilliant. The colors make it more identifiable. ..the GGAB..I'm not a programmer though I had tried writing in C++. I still can't figure out how the voice can be inputed into a program. Do you have to input GGABA....? Thanks for sharing. This is the most exciting thing I find today.
There are many different audio file formats of varying complexity. One of the simplest ones to work with is the uncompressed .WAV file format. You can use iTunes or ffmpeg to convert audio file formats such as MP3, AAC, Apple Lossless, AIFF, etc. into .WAV files. The latter can then be parsed without too much difficulty. Google ".wav format" for details of the .WAV specification.
I wrote it in C. (The end-titles were generated by scripting Photoshop, but even there the script file was generated and executed from within the command-line C program.)
The second harmonic of a particular note that is played is often discernible an octave above the fundamental; and sometimes the third and fourth harmonics as well at an octave and a fifth and two octaves above the fundamental, respectively. The harmonics tend to be quite visible when a note is played/sung loudly or with a sharp attack. This may be what you are referring to.
Distinct from the above is the uncertainty that arises from the fact that the music is being sampled for a finite period in time. Since we are doing a discrete Fourier transform, this means that in the frequency domain, we only have data at discrete points separated by an interval (delta_f) equal to the reciprocal of the sampling period. This imposes a limit to the frequency resolving ability of our analysis, typically about 20 Hz or so.
You can of course improve the frequency resolution by increasing the sampling period (time), but doing so would make the display less responsive (more sluggish). This is, in fact, the classical (pun intended) analog of the (quantum) Heisenberg uncertainty principle!
but I just noticed nearly the whole keyboard is vibrating each note. maybe just GATE the display by suppressing any animation if the velocity/gain isn't above a certain threshold?
How can I do it with one of my piece?
It is great!
robertofiore 1 year ago
gouldberg ?
ChundomanX 1 year ago
@ChundomanX Glenn Gould + Goldberg Variationen = GOULDberg
Get it?
tauwilltriumph 11 months ago
@tauwilltriumph o.o
ChundomanX 10 months ago
Oh, brilliant. The colors make it more identifiable. ..the GGAB..I'm not a programmer though I had tried writing in C++. I still can't figure out how the voice can be inputed into a program. Do you have to input GGABA....? Thanks for sharing. This is the most exciting thing I find today.
droi74 2 years ago
There are many different audio file formats of varying complexity. One of the simplest ones to work with is the uncompressed .WAV file format. You can use iTunes or ffmpeg to convert audio file formats such as MP3, AAC, Apple Lossless, AIFF, etc. into .WAV files. The latter can then be parsed without too much difficulty. Google ".wav format" for details of the .WAV specification.
inwit 2 years ago
Thanks for the reply. I know wav file. What I don't know: 1. Do you convert a sould file into this visual form?
or 2. Do you convert the digital scores into this visual form?
droi74 2 years ago
1. sound file -> visual form.
inwit 2 years ago
Pretty cool. What application was this done in?
rambothisistrautman 2 years ago 2
I piano-rolled my own.
inwit 2 years ago
piano-rolled my own?
You mean you wrote the software yourself? in Flash maybe?
whiterabbitpress 2 years ago
I wrote it in C. (The end-titles were generated by scripting Photoshop, but even there the script file was generated and executed from within the command-line C program.)
inwit 2 years ago
why so so many notes vibrate en mass? harmonics?
whiterabbitpress 2 years ago
The second harmonic of a particular note that is played is often discernible an octave above the fundamental; and sometimes the third and fourth harmonics as well at an octave and a fifth and two octaves above the fundamental, respectively. The harmonics tend to be quite visible when a note is played/sung loudly or with a sharp attack. This may be what you are referring to.
inwit 2 years ago
Distinct from the above is the uncertainty that arises from the fact that the music is being sampled for a finite period in time. Since we are doing a discrete Fourier transform, this means that in the frequency domain, we only have data at discrete points separated by an interval (delta_f) equal to the reciprocal of the sampling period. This imposes a limit to the frequency resolving ability of our analysis, typically about 20 Hz or so.
inwit 2 years ago
You can of course improve the frequency resolution by increasing the sampling period (time), but doing so would make the display less responsive (more sluggish). This is, in fact, the classical (pun intended) analog of the (quantum) Heisenberg uncertainty principle!
inwit 2 years ago
but I just noticed nearly the whole keyboard is vibrating each note. maybe just GATE the display by suppressing any animation if the velocity/gain isn't above a certain threshold?
whiterabbitpress 2 years ago
"Perfection . . . does not interest me. What is important is to vibrate." —George Enescu
inwit 2 years ago 2
Soud has it's colour! I like it!! :)
ascsco 2 years ago 3
Very nice! That's an interesting concept.
faraz1729 2 years ago