SHOW:
|
|
- or go back to the newest paste.
| 1 | - | function y = stft(x, w, N, H, fs) |
| 1 | + | function y = stft(x, w, N, H, fs,t) |
| 2 | % Analysis/synthesis of a sound using the short-time Fourier transform | |
| 3 | % Authors: J. Bonada, X. Serra, X. Amatriain, A. Loscos | |
| 4 | % x: input sound, w: analysis window (odd size), N: FFT size, H: hop size | |
| 5 | % y: output sound | |
| 6 | % | |
| 7 | %-------------------------------------------------------------------------- | |
| 8 | % This source code is provided without any warranties as published in | |
| 9 | % DAFX book 2nd edition, copyright Wiley & Sons 2011, available at | |
| 10 | % http://www.dafx.de. It may be used for educational purposes and not | |
| 11 | % for commercial applications without further permission. | |
| 12 | %-------------------------------------------------------------------------- | |
| 13 | ||
| 14 | M = length(w); % analysis window size | |
| 15 | N2 = N/2+1; % size of positive spectrum | |
| 16 | soundlength = length(x); % length of input sound array | |
| 17 | hM = (M-1)/2; % half analysis window size | |
| 18 | pin = 1+hM; % initialize sound pointer in middle of analysis window | |
| 19 | pend = soundlength-hM; % last sample to start a frame | |
| 20 | fftbuffer = zeros(N,1); % initialize buffer for FFT | |
| 21 | yw = zeros(M,1); % initialize output sound frame | |
| 22 | y = zeros(soundlength,1); % initialize output array | |
| 23 | w = w/sum(w); % normalize analysis window | |
| 24 | k = [0:1:N]; | |
| 25 | freq = fs*k/N; | |
| 26 | - | t = -80 ; % threshold |
| 26 | + | %t = -80 ; % threshold |
| 27 | ||
| 28 | %while pin<pend | |
| 29 | %-----analysis-----% | |
| 30 | clf; | |
| 31 | xw = x(pin-hM:pin+hM).*w(1:M); % window the input sound | |
| 32 | plot(freq(1:M), abs(xw)); | |
| 33 | %pause | |
| 34 | fftbuffer(:) = 0; % reset buffer | |
| 35 | fftbuffer(1:(M+1)/2) = xw((M+1)/2:M); % zero-phase window in fftbuffer | |
| 36 | fftbuffer(N-(M-1)/2+1:N) = xw(1:(M-1)/2); | |
| 37 | X = fft(fftbuffer); % compute FFT | |
| 38 | mX = 20*log10(abs(X(1:N2))); % magnitude spectrum of positive frequencies | |
| 39 | ploc = 1 + find((mX(2:N2-1)>t).*(mX(2:N2-1)>mX(3:N2)).*(mX(2:N2-1)>mX(1:N2-2))); | |
| 40 | subplot(2,1,1); | |
| 41 | plot(freq(1:N2),mX); | |
| 42 | hold on; | |
| 43 | plot(freq(ploc),mX(ploc),'X'); | |
| 44 | hold off; | |
| 45 | pX = unwrap(angle(X(1:N2))); % unwrapped phase spect. of positive freq. | |
| 46 | ||
| 47 | subplot(2,1,2) | |
| 48 | plot(freq(1:N2),pX); | |
| 49 | hold on; | |
| 50 | plot(freq(ploc),pX(ploc),'X'); | |
| 51 | hold off; | |
| 52 | pmag=mX(ploc); | |
| 53 | pphase=pX(ploc); | |
| 54 | pmag; | |
| 55 | pphase; | |
| 56 | %-----synthesis-----% | |
| 57 | % Y = zeros(N,1); % initialize output spectrum | |
| 58 | % Y(1:N2) = 10.^(mX/20).*exp(i.*pX); % generate positive freq. | |
| 59 | % Y(N2+1:N) = 10.^(mX(N2-1:-1:2)/20).*exp(-i.*pX(N2-1:-1:2)); | |
| 60 | % % generate neg.freq. | |
| 61 | % fftbuffer = real(ifft(Y)); % inverse FFT | |
| 62 | % yw(1:(M-1)/2) = fftbuffer(N-(M-1)/2+1:N); % undo zero-phase window | |
| 63 | % yw((M+1)/2:M) = fftbuffer(1:(M+1)/2); | |
| 64 | % y(pin-hM:pin+hM) = y(pin-hM:pin+hM) + H*yw(1:M); % overlap-add | |
| 65 | % pin = pin+H; % advance sound pointer | |
| 66 | % end | |
| 67 | % wavwrite(y, fs,'out.wav'); | |
| 68 | - | end |
| 68 | + | end |
| 69 | ||
| 70 |
