Benno Senoner wrote: > years ago at a DSP university class I wrote a small app to demonstrate > the use of > the STFT (short term fourier transformation) which is what you mentioned. > Basically it was a FFT based EQ with arbitrary FFT size and overlapping > factor. > normally one uses an overlap-add method and applies FFT windowing to > smooth out > the edges (hanning, hamming, kaiser etc). with kaiser windows and an > overlapping of > 8 (I don't remember exaclty), you get pratically zero artifacts and the > impulse response > shows a SNR around 100dB or so. > I remember that I played the Titanic movie theme from Celine Dion and I > could almost > filter out single notes from the voice. > We could such a module to LS too.
What kind of latency did that system have? The idea here, I think, would be for it to be real-time enough to be able to play this as a synthesizer. Not that I'm any kind of an expert. I'm just trying to expand my education is all. Thank you for your help, and for letting me badger the list. :-) -- Darren ------------------------------------------------------------------------- This SF.net email is sponsored by: Microsoft Defy all challenges. Microsoft(R) Visual Studio 2008. http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/ _______________________________________________ Linuxsampler-devel mailing list Linuxsampler-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/linuxsampler-devel
