基于模糊神經(jīng)網(wǎng)絡(luò)的聲母識(shí)別
CONSONANT RECOGNITION BASED ON FUZZY NEURAL NETWORK
-
摘要: 模板匹配法技術(shù)是漢語聲母識(shí)別中較為成功的算法,但它的缺陷影響了其恢復(fù)錯(cuò)誤、改善識(shí)別性能。神經(jīng)網(wǎng)絡(luò)(NN)和模糊系統(tǒng)的結(jié)合,保留了雙方的優(yōu)點(diǎn),充分利用了模糊神經(jīng)網(wǎng)良好的容錯(cuò)性能、計(jì)算性能、分類性能和決策性能。本文重點(diǎn)研究了兩種基于模糊神經(jīng)網(wǎng)的聲母識(shí)別方案,通過對(duì)其結(jié)構(gòu)、識(shí)別率和特點(diǎn)的分析,可看出模糊神經(jīng)網(wǎng)的聲母識(shí)別性能明顯優(yōu)于模板匹配法,是更適于語音識(shí)別的網(wǎng)絡(luò)。
-
關(guān)鍵詞:
- 聲母識(shí)別; 模糊系統(tǒng); 神經(jīng)網(wǎng)絡(luò)
Abstract: Conventional template matching technique is a successfully used algorithm in Chinese consonant recognition, yet its disadvantage limits its recovery of error,improvement of performance. The hybrid system based on the combination of neural network(NN) and fuzzy system maintains their advantages and makes full use of error tolerence performance, calculation performance, classification performance and decision performance for fuzzy neural network. In this paper, two kinds of cosonant recognition schemes based on fuzzy neural network are studied in detail. From the discussion of their structure, recognition rate and characteristics, it can be seen that recognition performance of fuzzy neural network is superior to template matching scheme and thus is more suitable for speech recognition. -
Tatsuo MATSUOKA. Syllable Recognition Using Integrated Neural Networks. in Proceedings of the[2]IJCNN-89, Tokyo: 1989, June: 251-258.[3]Matsuoka T, Hamada H, Nakatsu R. A Study on Integrated Neural Networks for Syllable Recognition. Proc. ASJ Fall Meeting, Tokyo: 1988, October: 2-16.[4]吳立忠.漢語普通話輔音分類的研究:[碩士論文].華南理工大學(xué)無線電系,1986.[5]Waibel A, et al. Phoneme Recognition: Neural Networks vs Hidden Markov Models. in Proc. of ICASSP88, New York: 1988, April: 107-110.[6]Berenji H. A reiforcement larrg-based architecture for fuzzy logic control. Int1, J. Approx. Reason-[7]ing, 1992, 6(2): 261-292.[8]Huntsberger T, Ajjimarangsee P. Parallel self-organizing feature maps for unsupervised pattern recog-[9]nition. Int1. J. General Sys., 1990, 10(16): 261-292.[10]Aikawa K, et al. Automatic Generation of Consonant Discrimination Rules. in Proc, of ICASSP86, Tokyo: 1986, April: 2755-2758.[11]諸靜,等.模糊控制原理與應(yīng)用.北京:機(jī)械工業(yè)出版社,1995,194-312.[12]Carpenter G A, Grossberg S, Rosen D B. Fuzzy ART:Fast stable learning and categorization of analog pattern by an adaptive resonance system[J].Neural Networks.1991, 4(6):759-772 -
計(jì)量
- 文章訪問數(shù): 1966
- HTML全文瀏覽量: 105
- PDF下載量: 401
- 被引次數(shù): 0
下載:
