How to write simple music program with MATLAB

Some people say ：MATLAB Besides having children , The others will .
Can it be used to create music ？
The answer is Tolerable
First we need to know what we hear DO RE MI, How did it come about ？ What is the difference between ？
Music as “ wave ” A kind of , Naturally, the superposition of sine waves can be used to simulate , Different frequencies can produce notes of different frequencies , So we just use MATLAB Generate a series of sine functions , Plus MATLAB Self contained **sound()** Function .
Music is mainly composed of tones and beats , So the music function needs two inputs , I.e. tone （tone） And beat （rythm）
Now the function module is given ：
Fs Is the sampling frequency , General default 8192Hz, That is, the number of samples per second ;
linspace(a,b,c) The function is used to uniformly generate b individual a To c Array between ;
freqs The array of is the frequency corresponding to different timbres , For the sake of simplicity, only commonly used ;
Be careful ： The playback frequency should be consistent with the sampling frequency ！！！

function y = gen_wave( tone, rythm )
%UNTITLED2  Music function   Corresponding music2
%    tone   pat 
    Fs = 8192;
    freqs = [523, 587, 659, 698, 783, 880, 988];
    x = linspace(0, 2 * pi * rythm, floor(Fs * rythm));
    y = sin(freqs(tone) * x);
end

Create a new main function and execute the following code to issue DO The sound .

Fs = 8192; 
y=[];
y = gen_wave(1,1);
sound(y,Fs);

After execution, use **plot()** Function to draw a sound image , Here's the picture

You can see very dense , After the local magnification is ：

First, you can see that after zooming in , Values are discontinuous , Then the sound is constant , This is not quite consistent with our reality , The piano should be at its maximum when it is just pressed , It will change with time , The sound fades away , So we are y Is multiplied by a function , Attenuate it , The code is as follows ：

function y = gen_wave( tone, rythm )

%    tone   pat 
    Fs = 8192;
    freqs = [523, 587, 659, 698, 783, 880, 988];
    x = linspace(0, 2 * pi * rythm, floor(Fs * rythm));
   y = sin(freqs(tone) * x) .*(1- x/(rythm * 2 *pi));
    
end

Rerun DO The sound , And draw an image ：

Zoom in

It can be seen that this is oscillation attenuation , Of course, it can also be multiplied by a function of exponential decay , The sound will be more real .
Then we can find a complete music score , such as 《 Little star 》

According to the simplified music, we can generate an array of music representatives of this song , The code is as follows ：

Fs = 8192;
y=[];
music = [1,1,5,5,6,6,5,...
         4,4,3,3,2,2,1,...
         5,5,4,4,3,3,2,...
         5,5,4,4,3,3,2,...
         1,1,5,5,6,6,5,...
         4,4,3,3,2,2,1];
music_length = length(music(:));

y1 = gen_wave(1,1);
y2 = gen_wave(2,1);
y3 = gen_wave(3,1);
y4 = gen_wave(4,1);
y5 = gen_wave(5,1);
y6 = gen_wave(6,1);
y7 = gen_wave(7,1);

for i = 1:1:music_length
    if music(i) == 1
        y=[y,y1];
        
    elseif music(i) == 2
        y=[y,y2];
        
    elseif music(i) == 3
        y=[y,y3];
        
    elseif music(i) == 4
        y=[y,y4];
        
    elseif music(i) == 5
        y=[y,y5];
        
    elseif music(i) == 6
        y=[y,y6];
        
    elseif music(i) == 7
        y=[y,y7];    
    end
end

sound(y, Fs);

By traversing the music array , Executing subroutines , We can generate an array of scales , Reuse sound() Just play it .

If it feels useful , Give me some advice. ,O(∩_∩)O ha-ha ~