How do I create a pretrained network for speech dataset

How do i create a pretrained network for speech dataset and save in .mat format. The "audioFeatureExtractor (Audio Toolbox)" object used to train the network, and normalization factors for the features.

Answer 1

[Varun Sai Alaparthi]

Hello Ehijas,

Please refer the following link for training a net on speech dataset

Train Speech Command Recognition Model Using Deep Learning - MATLAB & Simulink (mathworks.com)

After training the model you can save it as a .mat file using the following command

outputDir = "path_to_dir"; 
outputFile = fullfile(outputDir, "net.mat"); 
save(outputFile, "net"); 

In the case above, net is the name of the network in the MATLAB workspace, and we are saving it to the location path_to_dir/net.mat.

Or you can directly save it with this command

pretrainednet = net; 
save pretrainednet; 

I hope this information helps and please reach out for any further issues.

[EHIJAS KV]

thanks for the support