Path: utzoo!utgpu!jarvis.csri.toronto.edu!mailrus!tut.cis.ohio-state.edu!unmvax!polyslo!ddurbin
From: ddurbin@polyslo.CalPoly.EDU (Daniel A. Durbin)
Newsgroups: comp.binaries.ibm.pc.d
Subject: Re: digitize.arc data
Summary: I built one last year
Keywords: digitize, voice
Message-ID: <9943@polyslo.CalPoly.EDU>
Date: 4 Apr 89 16:21:01 GMT
References: <1813@csd4.milw.wisc.edu>
Reply-To: ddurbin@polyslo.CalPoly.EDU (Daniel A. Durbin)
Distribution: na
Organization: Cal Poly State University -- San Luis Obispo
Lines: 25

I received a file called DIGITIZE.ARC last summer which contained PLAY.EXE
and RECORD.EXE with the same date and time stamps as those included in the
file posted in comp.binaries.pc but it had fewer files.  However, I believe
the circuit (from Alan Jones) was the same.

I built the circuit last summer which worked as claimed.  The only touchy
part of the circuit was the potentiometer.  I placed it between a couple of
20Kohm resistors to reduce the circuit's sensitivity to its adjustment.

Since then I have designed and built an 8 bit 4 khz digitizer which eventually
became a voice mail card for the IBM PC.  With the files created by the 8 bit
4 Khz A/D converter, I created a file in the format of the .voi files that
come with DIGITIZE.ARC.  I found that I could recompile PLAY.EXE with the
clock frequency reduced to 8 Khz (down from 16 Khz) and convert my files to
half the size as those made for 16 Khz and could get almost the same quality
of playback as those made for 16 Khz.

In summary, the circuit works - the potientiometer adjustment is too touchy -
and the clock frequency could be reduced to 8 Khz to halve the size of the
.voi files.

	Daniel Durbin___________________________________________________
	SysOp: Cygnus X-1 BBS		| CIS: 73447,1744
	(805) 541-8505 (data)		| GEnie: D.DURBIN
	EL major at PolySlo		| ddurbin@polyslo.CalPoly.EDU