This paper presents a fuzzy system to speech classification. In our previous works, we had developed a method for speech segmentation. For speech classification, the universe of discourse is divided into two types, and each type is treated as a class. These are silence and speech. The hyper-rectangular fuzzy system (HRFS) is used to classify frames and integrate the rule-based approach. The energy and entropy can extract fuzzy classification rules. In our experiments, continuous reading-rate of Mandarin balanced sentences was using to illustrate the performance of the proposed speech classification system. The effectiveness of the proposed system is confirmed by the experimental results.
關聯:
2001中華民國第九屆模糊理論及其應用會議論文集=Proceedings of 2001 Ninth National Conference on Fuzzy Theory and Its Applications,4頁