/*  REVISION INFORMATION  *****************************************************
#
# Revision 1.8  1995/04/17  19:51:19  len
# Temporary fix to frication balance.
#
# Revision 1.7  1995/03/21  04:52:35  len
# Now compiles FAT.  Also adjusted mono and stereo output volume to match
# approximately the output volume of the DSP.
#
# Revision 1.6  1995/03/04  05:55:54  len
# Changed controlRate parameter to a float.
#
# Revision 1.5  1995/03/02  04:33:03  len
# Added amplitude scaling to input of vocal tract and throat, to keep the
# software TRM in line with the DSP version.
#
# Revision 1.4  1994/11/24  05:24:09  len
# Added Hi/Low output sample rate switch.
#
# Revision 1.3  1994/10/20  21:20:16  len
# Changed nose and mouth aperture filter coefficients, so now specified as
# Hz values (which scale appropriately as the tube length changes), rather
# than arbitrary coefficient values (which don't scale).
#
# Revision 1.2  1994/08/05  03:12:48  len
# Resectioned tube so that it more closely conforms the the DRM proportions.
# Also changed frication injection so now allowed from S3 to S10.
#
# Revision 1.1.1.1  1994/07/07  03:48:52  len
# Initial archived version.
#

******************************************************************************/


VERSION:		v1	Original version
			v2	Resectioned version
			v3	Normalized aperture coefficients
			v4	Added Hi/Low output sample rate switch
			v5	Added scaling for vocal tract & throat
			v6	Changed controlRate to float
			v7	Compiles FAT, adjusted output volume
			v8	Temporary fix to frication balance

******************************************************************************/


HISTORY:

April 17th, 1995:	Adjusted the vocal tract volume scaling so that the
			frication is less loud.  This is a temporary fix to
			try to match the output of the dsp synthesizer---
			a proper fix is still needed.

March 20th, 1995:	Now compiles FAT (i.e. for the m68040, and for Intel).
			Output .snd files are always produced in big-endian
			format.  Also adjusted the mono and stereo output
			volumes so that they more or less match the output
			produced by the DSP (i.e. roughly 1/4 of the maximum
			amplitude).

March 3rd, 1995:	Changed the controlRate parameter to a float, keeping
			the software TRM in line with the DSP version.

March 1st, 1995:	Added scaling (2^(-5)) on input to the vocal tract
			& throat, to keep the software TRM in line with the
			DSP version.  Scaling should be added to the frication
			for both versions, so that parameter input can use
			a larger range.

November 23rd, 1994:	Added the Hi/Low output sample rate switch to the
			utterance rate parameters.  This allows either
			a 22050 or 44100 output sample rate (although other
			rates are possible, but not directly playable using
			NeXT computers).  This change keeps the software TRM
			in line with the DSP version.

October 20th, 1994:	Changed nose and mouth aperture filter coefficients,
			so now specified as Hz values (which scale
			appropriately as the tube length changes), rather
			than arbitrary coefficient values (which don't scale).

August 4th, 1994:	Resectioned the tube to make it conform as much as
			possible with DRM sectioning.  There are now 10
			sections in the oropharyngeal cavity (i.e. from the
			glottis to the mouth), grouped into 8 regions.  Also,
			frication can now from S3 to S10 (specified as 0 - 7).

			|-R1-|-R2-|-R3-|----R4---|----R5---|-R6-|-R7-|-R8--|
			|-S1-|-S2-|-S3-|-S4-|-S5-|-S6-|-S7-|-S8-|-S9-|-S10-|

			|----|----|-F0-|-F1-|-F2-|-F3-|-F4-|-F5-|-F6-|-F7--|


July 6th, 1994:		Initial archive of the software (non-real-time)
			version of the Tube Resonance Model for the production
			of speech.  This version has been harmonized with the
			DSP version.