Function that fits a wave

I need to create a function that describes an audio wave. I have tried the fit() function with sine and gaussian but they don't have sufficient detail. Can you recomend me some technique that generates a function that more accurately describes this wave?

The following is my code:

clear all;

[y,Fs] = audioread('raindrop.wav');

sound(y,Fs);

x = (1:1:length(y))';

f = fit(x,y,'gauss8');

C = coeffvalues(f);

plot(f,x,y)

%the following shows me just the fit

F = C(01)*exp(-((x-C(02))/C(03)).^2) + C(04)*exp(-((x-C(05))/C(06)).^2) + ...

C(07)*exp(-((x-C(08))/C(09)).^2) + C(10)*exp(-((x-C(11))/C(12)).^2) + ...

C(13)*exp(-((x-C(14))/C(15)).^2) + C(16)*exp(-((x-C(17))/C(18)).^2) + ...

C(19)*exp(-((x-C(20))/C(21)).^2) + C(22)*exp(-((x-C(23))/C(24)).^2);

plot(x,F)

sound(F,Fs)

[Mathieu NOE]

hello

maybe some usefull links :

How to fit a multiple gaussian in a curve with a multiple peaks - (mathworks.com)

Fitting Gaussian to a curve with multiple peaks - (mathworks.com)

Fit Multiple Gaussians - File Exchange - MATLAB Central (mathworks.com)

[Rowan Humphreys]

Thanks, I will look into that :)

Answer 1

[Pooja Kumari]

Dear Rowan,

It is my understanding that you are facing issue with creating a function to describe an audio wave with more detail than a simple sine or Gaussian function. I am assuming that you want to learn more about how to describe an audio wave. You can consider any of the below approach for describing an audio wave:

• Using Audio System Toolbox: For the design and analysis of speech, acoustic, and audio processing systems, Audio System Toolbox offers a variety of features and tools., you can refer to the below documentation for more information: https://www.mathworks.com/help/audio/getting-started-with-audio-system-toolbox.html?s_tid=CRUX_lftnav

You can also examine and comprehend the properties of the audio wave more thoroughly by making an app. You can refer to the below example on how to create an app to play and visualize audio file:

https://www.mathworks.com/help/audio/ug/create-an-app-to-play-and-visualize-audio-files.html

• Using Fourier analysis of audio signal: Performing a Fourier analysis on an audio signal involves applying a Fourier transform, such as the Fast Fourier Transform (FFT), to move the audio wave from the time domain to the frequency domain. This will give the amplitude and phase information for each frequency component present in the audio wave. You can refer to the documentation below for more information: https://www.mathworks.com/help/matlab/ref/fft.html

Regards,

Pooja Kumari