Piano and Live Electronics


    Apr 25 2008 | 8:23 pm
    Greetings to all!
    I was wondering if anyone know about any paper or any other information concerning music pieces for PIANO and LIVE ELECTRONICS. I am trying to make one, but I want some external ideas. Thanks in advance everyone for the help...
    panos

    • Apr 25 2008 | 8:55 pm
      By "live electronics" I assume you mean using the piano as
      a sounds source. Using the piano as a sound source is
      problematic because is it polyphonic and current pitch
      detection tools are monophonic. Writing a piece that does
      not use the pianos strength (polyphony) is not very interesting.
      You might as well use a different instrument.
      I think a good place to start is try to sculpt away parts of
      the incoming spectra, or creating new related spectra by
      sustaining/enhancing existing frequencies.
    • Apr 25 2008 | 9:45 pm
      If you intend to start with analyzing a live- Piano for further processing, there is an excellent external by Arshia Cont for analyzing and transcribing live-piano. You can find it at :
      Hope this helps,
      _A
    • Apr 26 2008 | 12:06 am
      You could do a lot worse than just doing a search for David Tutor.
      bt
      On Apr 25, 2008, at 1:23 PM, panos amelides wrote:
      >
      > Greetings to all!
      >
      > I was wondering if anyone know about any paper or any other
      > information concerning music pieces for PIANO and LIVE ELECTRONICS.
      > I am trying to make one, but I want some external ideas. Thanks in
      > advance everyone for the help...
      >
      > panos
      barry threw
      Media Art and Technology
      San Francisco, CA
      Work: 857-544-3967
      Email: bthrew (at) gmail (dot) com
      Web: www.barrythrew.com
    • Apr 26 2008 | 1:40 am
      It's spelled David Tudor.
      Sorry, phonetic F'up.
      Muir was on me like lightning. ;)
      bt
      On Apr 25, 2008, at 2:45 PM, Andreas Breitscheid wrote:
      >
      >
      > If you intend to start with analyzing a live- Piano for further
      > processing, there is an excellent external by Arshia Cont for
      > analyzing and transcribing live-piano. You can find it at :
      > http://cosmal.ucsd.edu/arshia/index.php?n=Main.Multipitch
      >
      > Hope this helps,
      > _A
      barry threw
      Media Art and Technology
      San Francisco, CA
      Work: 857-544-3967
      Email: bthrew (at) gmail (dot) com
      Web: www.barrythrew.com
    • Apr 26 2008 | 1:38 pm
      Quote: barry threw wrote on Fri, 25 April 2008 18:06
      ----------------------------------------------------
      You could do a lot worse than just doing a search for David Tutor.
      ----------------------------------------------------
      Very nice !
    • Apr 26 2008 | 3:08 pm
      I'm trying to make the waveshaper of the article :
      2/3; x >= 1
      f(x) = x - x3/3; -1 < x < 1
      -2/3; x
      I attached a patch showing what I come up with, on one side, audio arithmetic and on the other, [lookup~].
      If any one have idea to implement this waveshaper using [lookup~] (I'm pretty sure lookup~ is less expensive than audio arithmetic).
    • Apr 26 2008 | 7:03 pm
      On 26 Apr 2008, at 17:08, Bertrand Fraysse wrote:
      > If any one have idea to implement this waveshaper using [lookup~]
      you got it almost right.
      simply clip between -1 and 1.
      the waveshaper will do the rest.
      in your arithmetic approach you missed a cord from output of clip~ to
      the right side of the first multiplication.
      but it will still not work like that, i believe.
      v
      ps: there seems to be a long standing bug (or misunderstanding on my
      side) about lookup~'s table size.
      anyway, you should make sure your buffer is actually at least 513
      samples long, and compute the wavetable with 513 (0 to 512) rather
      than 0 to 511. (larger table sizes won't hurt, but improve your
      results).
    • Apr 26 2008 | 10:34 pm
    • Apr 26 2008 | 11:15 pm
      Hello,
      If you wanna get an idea of pitches from a chord, you can use
      transcribe~
      This can be found at the following address :
      This is using NMF (non-negative matrix factorization) with analysed
      templates in a text file. Each template corresponds to a component
      nmf will extract ("""demixing"""). Each component can be a note.
      Pitches are deduced this way.
      "Realtime Multiple Pitch Observation using Sparse Non-negative
      Constraints" can be found at the following address :
      best,
      O../////
    • Apr 26 2008 | 11:16 pm
      >
      >
      > ps: there seems to be a long standing bug (or misunderstanding on my side)
      > about lookup~'s table size.
      > anyway, you should make sure your buffer is actually at least 513 samples
      > long, and compute the wavetable with 513 (0 to 512) rather than 0 to 511.
      > (larger table sizes won't hurt, but improve your results).
      >
      As I understand it, with cycle~ you can enter 512 or 513 samples, the
      buffer is always 513 samples long with the last one giving you the option of
      having the wave interpolate over the first and last values or for you to
      enter a value manually. Perhaps lookup~ gives you a similar option if it's
      this length?
      pelado
      --
    • Apr 27 2008 | 10:47 am
      hello pelado,
      On 27 Apr 2008, at 01:16, pelado wrote:
      >
      >> ps: there seems to be a long standing bug (or misunderstanding on
      >> my side) about lookup~'s table size.
      >> anyway, you should make sure your buffer is actually at least 513
      >> samples long, and compute the wavetable with 513 (0 to 512) rather
      >> than 0 to 511. (larger table sizes won't hurt, but improve your
      >> results).
      >
      > As I understand it, with cycle~ you can enter 512 or 513 samples,
      > the buffer is always 513 samples long with the last one giving you
      > the option of having the wave interpolate over the first and last
      > values or for you to enter a value manually. Perhaps lookup~ gives
      > you a similar option if it's this length?
      the interpolation from last to first value makes sense if you want to
      use the wavetable as a lookup for an oscillator, i.e. if you want to
      loop that thing.
      in the waveshaping context, i can't see any reason why this would be
      desirable.
      i don't know what's really going on, but i only get correct results
      for odd table sizes.
      v
    • Feb 19 2009 | 12:48 am
      See message #165895