jcm@mtunb.ATT.COM (John McMillan) (12/19/89)
In article <668@nz8r.UUCP> gwr@nz8r.UUCP (Gregory W. Ratcliff) writes: > >If the method of data conversion is "discovered" for the Unixpc >voice board, I would sure like to hear about it. I _don't_ do Voice, but I hope/believe the following statements are close enough to correct to be useful: Of the three codings you can choose (16K, 24K, 64K), the 64K mode is very accessible: it is standard mu-law encoding which even _I_ have noticed (& ignored) being discussed in many places over the last few years. It should be your choice for Fourier- or Psycho- analysis. The other two codings are Sub-Band Code which has been written up in one of the Bell Labs journals, but is much less accessible and does involve proprietary information. I'd recommend that you familiarize yourself with mu-law, try out some tests using that interpretation, and then contact me for more info. NOTE: I will probably be re-incarnated at a different E-mail address by then, unless you're a quick learner! john mcmillan -- att!mtunb!jcm -- Closet U-pc hacker