Text this: Audio-visual based speech recognition using lip feature and hidden markov modelling /