🚀 go-pugleaf

Does anybody know anything about CVSD (continuously variable slope
deltamodulation ) for encoding of speech??? I have a speech card that uses this
technique for recording and play-back, it produces very good quality speech
considering it is only running at 32 K bits sec.

However wish to play back the sound through an I.B.M PC internal speaker. I am
currently trying to write a program to do this, treating the data as a
bitstream sort of works but it is not very clear.

A shareware utility called REMAC produces "quite good" quality sound on the
I.B.M speaker and almost plays CVSD files.

I've tried different types of bit "averaging" on the data but this does not
seem to have much effect. I think some form of system modeling is required.

Would anybody have any bright ideas????

Any references to relevant books on the subject ????



Thanks in advance


Dougal