codeclips

Some useful code clips for my work.

Recognize Sound

ライフゲームの世界

Recognizing Sounds (A Deep Learning Case Study)

You can download data set from here data set of Recognizing Sounds.

https://github.com/ybayle/awesome-deep-learning-music

https://qiita.com/eve_yk/items/07bc094538f2d50841f4

WinLen = 1;
SamplingFrequence = 44100;
SamplingLen = SamplingFrequence * WinLen;
clear AudioData, SamplingFrequence;
[chord1, SamplingFrequence] = audioread("chord1.wav", [1, SamplingLen]);
[fret_3rd, SamplingFrequence] = audioread("3rd_fret.wav", [1, SamplingLen]);

SamplingPeriod = 1/SamplingFrequence;

% Time = (0:length(chord1)-1) * SamplingPeriod;

chord1FFT = fft(chord1);
fret_3rdFFT = fft(fret_3rd);

chord1P2 = abs(chord1FFT/SamplingLen);
chord1P1 = chord1P2(1:SamplingLen/2+1);
chord1P1(2:end-1) = 2*chord1P1(2:end-1);

fret_3rdP2 = abs(fret_3rdFFT/SamplingLen);
fret_3rdP1 = fret_3rdP2(1:SamplingLen/2+1);
fret_3rdP1(2:end-1) = 2*fret_3rdP1(2:end-1);

Frequence = (SamplingFrequence/SamplingLen)*(0:(SamplingLen/2));

figure (1);
plot(Frequence, chord1P1);
figure (2);
plot(Frequence, fret_3rdP1);


title('Single-Sided Amplitude Spectrum of X(t)');
xlabel('f (Hz)');
ylabel('|P1(f)|');