[music-dsp] Frequency domain never ceases to amaze me...
rwd at btconnect.com
Mon Mar 3 03:50:01 EST 2003
In principle, any alteration to individual frequency bins (speaking in pvoc
terms) will do this; as the various components in the source sound are
represented by multiple frequency bins in the analysis (indeed, in most cases,
by all of them, because of spectral leakage, windowing etc). In practice,
masking and other conveniences mean that a lot of effects can sound remarkably
clean; though they would not stand up to scrutiny by those with Golden Ears.
And it is always possible to say "that's what I want it to sound like".
Andrea Gozzi wrote:
> But (naive question)... doesn't the phase component also implicitly contains
> fractional frequency component?
> (i'm thinking to the phase vocoder approach, where the estimated true
> frequency is detected by phase deviations)
> I'd expect that smoothing the phase component would bring out some small
> inharmonicity or chorusing (because of all the different partials) in the
> ifft'ed sound..
> Is this a correct reasoning?
> dupswapdrop -- the music-dsp mailing list and website: subscription info, FAQ, source code archive, list archive, book reviews, dsp links http://shoko.calarts.edu/musicdsp/
More information about the music-dsp