1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
|
class:: IFFT
summary:: Inverse Fast Fourier Transform
related:: Classes/FFT, Guides/FFT-Overview
categories:: UGens>FFT
Description::
The inverse fast fourier transform converts from frequency content to a
signal.
The fast fourier transform analyzes the frequency content of a signal. The IFFT UGen converts this emphasis::frequency-domain:: information back into emphasis::time-domain:: audio data. Most often this is used as the end of a process which begins with link::Classes/FFT::, followed by frequency-domain processing using PV (phase-vocoder) UGens, followed by IFFT.
classmethods::
method::new, ar, kr
returns a time domain signal from converting the FFT frequency domain signal chain. The *new method is equivalent to the *ar message returns an audio rate signal.
argument::buffer
The FFT "chain" signal coming originally from an FFT UGen, perhaps via other PV UGens.
argument:: wintype
Defines how the data is windowed:
table::
## -1 || strong::rectangular:: windowing, simple but typically not recommended;
## 0 || (the default) strong::Sine:: windowing, typically recommended for phase-vocoder work;
## 1 || strong::Hann:: windowing, typically recommended for analysis work.
::
argument:: winsize
Can be used to account for zero-padding, in the same way as the link::Classes/FFT:: UGen.
returns::
The emphasis::time-domain:: audio signal.
discussion::
The IFFT UGen converts the FFT data in-place (in the original FFT buffer) and overlap-adds the result to produce a continuous signal at its output.
Examples::
code::
// without any modification, convert FFT chain (frequency domain signal) back to audio (time domain signal)
(
{ var in, chain;
in = WhiteNoise.ar;
chain = FFT(LocalBuf(2048), in);
IFFT.ar(chain) * -20.dbamp // inverse FFT
}.play;
)
::
|
