Break audio file into separate matrices depending on silence breaks

I have an audio file that has some speech like this: Hi (pause) How are you (pause)

Using wavread, I can sample this audio and transfer the sample data into a single matrix, however, I want to break it into 2 matrices as follows:

A= [data from hi]

B= [data from how are you]

Any ideas? Thanks!

[Walter Roberson]

https://www.mathworks.com/matlabcentral/answers/?term=tag%3A%22silence+detection%22

[Image Analyst]

You'd get more help if you attach your audio file. Basically, threshold the signal and throw out really short low signal stretches and keep only the longer silent areas, like a half second or whatever. Then use the indexed of those non-silent areas to crop them out. Very very easy if you have the Image Processing Toolbox (do you have that). I'll make you a demo if you need it and if you supply the audio file.