what speech processing features are available in MATLAB package? Also, suggest some algorithms for the same.

I am making a projects which requires images to be associated with tags. These tags will be obtained as audio input. The final goal is to identify the image each time a user inputs an associated tag.

The user input will be in audio form and will be compared with saved tags to identify the corresponding image. The user can be anyone. So, we are talking about different pitch, intensity etc. So, the main problem is the matching of input audio data with saved data. Suggest algorithms or functions available.