Implementation of phonetic level speech recognition system for Punjabi language

This paper explains the implementation of the phonetic level speech recognition system for Punjabi language because it is a highly prosodic language. Here Hidden Markov Toolkit (HTK) is used. First step is data collection and nine hours data is collected in read speech mode. Second step is data preparation, in which hmmlist, grammar and dictionary files are created. Once the data is prepared, 75% and 25% of data is used for training and testing respectively. The experimental results show that the accuracy of the system increases from 49.95% to 59.38% as number of phonetic unit's increases from 30 to 34 and data increases from 3 hours to 6 hours respectively.