My (2 year old) son likes screaming at the OLPC "Measure" application, so I'd like to have something like that for all of our computers (i.e. something that displays audio as a simple view that lets you see your own voice).  SilleScope is the result.  At the moment it only has an ALSA source (i.e. Linux desktop only), but it's pretty simple wrt what it accepts as input (it just reads a stream of 8-bit unsigned ints) from stdin and dumps them into an OpenGL texture (which is a 1D ring-buffer, basically).  The rendering is done with an OpenGL shader, so can be customized pretty easily.

Usage:

    virtualenv sillescope-env
    source sillescope-env/bin/activate
    pip install sill-e-scope

    alsa-source | sillescope
    sine-source | sillescope 

Of course, since it's OpenGL, it likely won't work on the OLPC, but oh well.  Anyway, it is there should someone wish to play with it.  There's lots of basic things that should be written, such as options to allow you to choose ALSA sources, basic interactions (e.g. letting you zoom in, or display fewer samples/screen).

I would like to figure out why alsasound doesn't seem to return 16-bit samples when I ask it to :( .  But better would be figuring out how to do the read with an old /dev/pcm style read (I gather we lost that one with OSS), to eliminate the second dependency (I don't mind PyOpenGL, *I* already have that :) ).