[CM] cml, bird, freq-skew

[Bill Schottstaedt]

> No, it seems to be the profile (or outline) of the 
> freq-variation. 

That's right.  It's a multiplier on the frequency envelope
in Hz, so at some point in more explicit code, like your
python example, you'll need to convert that value to a
phase increment, via (I haven't had my coffee yet...)
freqskew * 2 * pi / sampling-rate.  I hope that's not
upside down or something -- mus_hz_to_radians
in clm.c, probably.