I am doing a project on near duplicate video detection.I want to compare the audio contents by creating a dictionary from reference videos and compare them against the query video. How to form a dictionary? How can i compare them?Please help

I am currently doing a project on near duplicate video detection based on audio content.I want to compare the audio contents by creating a dictionary from reference videos and compare them against the query video. I have extracted audio from videos in .wav format. I have to form a dictionary that contains the words extracted from the audio to compare them against the query audio.If the words in the dictionary match with the query audio then they are duplicate. I don't no how to form a dictionary? and How can i compare them in Matlab? Please help!!