This thesis examines the impact of accent present in speech on the performance of speaker independent Automatic Speech Recognition (ASR) systems. The research is focused on validating the assumption that the presence of accent in speech can increase the error rate and that a speaker independent ASR engine, trained by a variety of accents, performs poorer than an engine that is trained for a particular accent, when tested by the same accent. It also proposes a method to lower the recognition error rate and measure the improvement by first determining the accent of the utterance and then applying the appropriate ASR engine from a bank of engines trained for different accents. The second part of this research work proposes a method for identifying accents by tracking the temporal evolution of speaker’s vocal tract. An LPC feature space is proposed to model the speaker’s vocal tract.