We have been focusing on applying speech technologies to pronunciation learning. In our previous study [1], a stressed syllable detector was implemented by using stressed syllable HMMs and unstressed oens. And using the detector internally, several systemswere implemented [2]. However, their development did not necessarily require the use of HMMs as an acoustic modeling method. In this paper, an HMM-based method, a DTW-based method, and a human strategy only iwth visual inspection were compared in terms of their performance in judging whether two utterances of a word have the same stress pattern, e.g. record and record. Here, one utterance was given by a Japanese learner and the other one was done by a native speaker. Experiments showed that HMMs gave us the higher performance than DTW and even human strategies. This result strongly supports the use of HMMs as an acoustic modeling method in the stressed syllable detector development.
展开▼