Speech Recognition using MFCC and k-NN algorithm in Matlab

I am working on speech recognition and I have 30 recordings for 5 spoken words from 5 different speakers. I have extracted the features using MFCC algorithm. Now I have the following questions:

1. Do I have to save these features in .mat file? Would that be in a single file? Because different samples may not have the same number of rows.
2. How to feed these into fitcknn() function in matlab? I looked through the documentation but still I am not clear on how to apply this to my data.

Thanks in advance.

Answer 1

[Walter Roberson]

You can save them all in one .mat file if you save them as different variables, or if you save them all inside one cell array.

[KS]

Thank you Walter for your answer. But what about the second part of my question? How do I use that function? Any ideas?

[Walter Roberson]

The examples on the documentation page show how to use it. However, I do not know anything about why to use it.