Robust Speech Recognition using Vocal Tract Normalization for Emotional Variation

This paper studied the training methods less affected by the emotional variation for the development of the robust speech recognition system. For this purpose, the effect of emot ional variations on the speech signal were studied using speech database containing various emotions. The performance of the speech recognition system trained by using the speech signal containing no emotion is deteriorated if the test speech signal contains the emotions because of the emotional difference between the test and training data. In thi s study, it is observed that vocal tract length of the speaker is affected by the emotional variation and this effect is one of the reasons that makes the performance of the speech recognition system worse. In this paper, vocal tract normalization method is used to develop the robust speech recognition system for emotional variations. Experimental resul ts from the isolated word recognition using HMM showed that the vocal tract normalization method reduced the er ror rate of the conventional recognition system by 41.9% when emotional test data was used.