Hi Nhat,
I understand that you want to identify speaker using MFCC Feature and KMean Algorithm. Unfortunately it will not be possible to provide the complete code.
However, 'mfcc' function can be used to extract MFCC feature of speech signal.
Please refer to the documentation of 'mfcc' to understand usage of 'mfcc' function:
Then using cluster data task, the features can be clustered and it also provides the code to do the same. Please refer to the following documentation for more information:
Please refer to following documentation for more information on k-means clustering function of MATLAB:
I hope this helps!