On 10/22/13 3:16 AM, Robert Bielik wrote:
Hi Robert,
robert bristow-johnson skrev 2013-10-21 18:13:
this is interesting, but i am not so sure i agree with it. i've
always been going under the assumption that the grain size is 2N,
twice the length of the input period (and overlapping complementary
windows so that at a shift of 0 cents, there is perfect
reconstruction of the original). but i always thought that if
upshift, there would be more than 2 overlapping grains. for a
maximum of 1 octave up, i've used a maximum of 4 overlapping grains.
but i am *very* interested to find out if/that my previous M.O. is
wrong.
Can't really argue with you if it is wrong or not, I just remember
that that's how I used to do it. It *might* have to do with loudness,
i.e. that when upshifting, there would be a perceived increase in
loudness due to many overlapping grains, however my memory may fail me
here... :)
just to be clear, 2 overlapping grains mean, at most, unity gain because
they have complementary windowing to create the grains.
there *is* an increased energy or power in the upshifting. this
PSOLA/Lent/Hamon method is intended, i think, for pitch shifting of
vocals and in such a way that they don't sound like Alvin the Chipmunk
(upshifting) or Satan from Hell (sufficient downshifting). i think, in
the vocal model, higher notes *are* louder than lower notes, often in a
single phrase of singing.
but it's nothing that can't get adjusted with a smoothed
shift-interval-dependent gain. or it could be shift-interval, input
pitch and loudness, and even affected by a "brightness" timbre
measurement (smoothed highpass dB amplitude minus lowpass dB
amplitude). mix those parameters together in some reasonable manner to
get a loudness adjustment that might fix an undesired increase in loudness.
or just hook a compressor onto it, i dunno.
--
r b-j r...@audioimagination.com
"Imagination is more important than knowledge."
--
dupswapdrop -- the music-dsp mailing list and website:
subscription info, FAQ, source code archive, list archive, book reviews, dsp
links
http://music.columbia.edu/cmc/music-dsp
http://music.columbia.edu/mailman/listinfo/music-dsp
