Isolated Word Recognition by Recursive HMM Parameter Estimation Algorithm

Automatic speech recognition (ASR) technologies enable humans to communicate with computers. Isolated word (IWR) is an important part of many known ASR systems. Minimizing the error rate in cases incremental learning a unique challenge for developing on-line system. This paper focuses on IWR using recursive hidden Markov model (HMM) multivariate parameter estimation algorithm. The maximum likelihood method was used estimate unknown parameters model, and algorithm adapted EM HMMs derived. resulting among its counterparts because state transition probabilities calculation. It obtains more accurate estimates compared other algorithms this type. In our experiment, implemented several datasets IWR. Thus, convergence results are discussed work.