Description

Scale the frequency components of a pv stream, resulting in pitch shift. Output amplitudes can be optionally modified in order to attempt formant preservation.

Syntax

fsig pvscale fsigin, kscal[, kkeepform, kgain, kcoefs]

Performance

fsig -- output pv stream

fsigin -- input pv stream

kscal -- scaling ratio.

kkeepform -- attempt to keep input signal formants; 0: do not keep formants; 1: keep formants using a liftered cepstrum method; 2: keep formants by using a true envelope method (defaults to 0).

kgain -- amplitude scaling (defaults to 1).

kcoefs -- number of cepstrum coefs used in formant preservation (defaults to 80).

The quality of the pitch shift will be improved with the use of a Hanning window in the pvoc analysis. Formant preservation method 1 is less intensive than method 2, which might not be suited to realtime use.

	Warning
	It is unsafe to use the same f-variable for both input and output of pvs opcodes. Using the same one might lead to undefined behavior on some opcodes. Use a different one on the left and right sides of the opcode.

Examples

asig  in                                 ; get the signal in

fsig  pvsanal   asig, 1024, 256, 1024, 1 ; analyse it
ftps  pvscale   fsig, 1.5, 1, 1          ; transpose it keeping formants
atps  pvsynth   ftps                     ; synthesise it

adp   delayr    0.1                      ; delay original signal
adel  deltapn   1024                     ; by 1024 samples
      delayw    asig
      out       atps + adel              ; add tranposed and original

The example above shows a vocal harmoniser. The delay is necessary to time-align the signals, as the analysis-synthesis process will imply a delay of 1024 samples between the analysis input and the synthesis output.

Here is an example of the use of the pvscale opcode. It uses the file pvscale.csd.

<CsoundSynthesizer>
<CsOptions>
-odac
</CsOptions>
<CsInstruments>

sr = 44100
ksmps = 16
nchnls = 1
0dbfs = 1

;; example written by joachim heintz 2009

instr 1
ifftsize	=		1024
ioverlap	=		ifftsize / 4
iwinsize	=		ifftsize
iwinshape	=		1; von-Hann window
Sfile		=		"fox.wav"
ain		soundin	Sfile
fftin		pvsanal	ain, ifftsize, ioverlap, iwinsize, iwinshape; fft-analysis of the audio-signal
fftblur	pvscale	fftin, p4, p5, p6; scale
aout		pvsynth	fftblur; resynthesis
		out		aout
endin

</CsInstruments>
<CsScore>
i 1 0 3 1 0 1; original sound
i 1 3 3 1.5 0 2; fifth higher without ...
i 1 6 3 1.5 1 2; ... and with different ...
i 1 9 3 1.5 2 5; ... kinds of formant preservation
e
</CsScore>
</CsoundSynthesizer>

See Also

pvsanal, pvsynth, pvsadsyn

Credits

New plugin in version 5

November 2004.