Is there a better way to write my downsampling function?

I have an audio data I read as a csv and to reduce computation I downsample my data before going with further processing. My code runs my datareduced function 4 times to downsample the data 81:1, but I'm wondering if there's a more efficient/shorter way to write this up in matlab

data = csvread( "VoiceTest1.csv");

% downsample data (81:1) to make processing quicker
reducedData = data;
for i = 1:4
  reducedData = reduce(reducedData);
end


function dataReduced = reduce(d)
  x1 = d(1:3:end);  %slice the x0 and x1's
  x2 = d(2:3:end);  %slice the x0 and x1's
  x3 = d(3:3:end);  %slice the x0 and x1's

  smooth = conv([1/2 1/2], [1/2 1/2]);

  s = min([size(x1,1)   size(x2,1)  size(x3,1)]);
  x1 = x1(1:s);
  x2 = x2(1:s);
  x3 = x3(1:s);
  
  dataReduced = x1.*smooth(1) + x2.*smooth(2) + x3.*smooth(3);
end

Is there any way to make this function better in matlab?

Any suggestions? Thanks

Answer 1

[jibrahim]

Hi Adriana,

You should be able to use one of many resampling functions in Signal Processing Toolbox:

https://www.mathworks.com/help/signal/ug/resampling.html

Take a look at resample in particular for your use case.