home *** CD-ROM | disk | FTP | other *** search
- Newsgroups: comp.dsp
- Path: sparky!uunet!psinntp!newstand.syr.edu!rodan.acs.syr.edu!mepeters
- From: mepeters@rodan.acs.syr.edu (Mark E. Peterson)
- Subject: Vocoders and FFTs
- Message-ID: <1992Dec14.235747.7470@newstand.syr.edu>
- Organization: Syracuse University, Syracuse, NY
- Date: Mon, 14 Dec 92 23:57:46 EST
- Lines: 12
-
- Hello. Reading through the posts on pitch shiftings I have heard nothing
- about using FFTs to reconstruct voice signals using different sets of
- frequencies. Is this sort of method used for vocoders? Would a 256 point
- FFT be enough to do the job if you just modulated 256 certain sine waves
- together according to amplitude? I would like to experiment with the
- sort of idea, but I'd like to learn if it's been done before. I want to
- do it in real time. I have the integer FFT posted a few days ago that
- can do a real-time FFT in 2.6 milliseconds.. That combined with averaging
- 256 sine waves could be accomplished barely in real time at 4Khz at least.
- Has anyone had or heard of any success with this?
-
-
-