外文翻譯--基于仿生模式識(shí)別的非特定人連續(xù)語(yǔ)音識(shí)別系統(tǒng)-全文預(yù)覽
【正文】 that ? achieved hardly the same recognition rate as the Basic algorithm. In the mean time, the MWNs used in the works are much less than of the Basic algorithm. Table 2． Experiment results of BPR basic algorithm recognition method The first option choice recognition rate （ Test set） The first two option choice Recognition rate（ Test set） DTW % % HMM % % BPR Basic algorithm % % Experiments were also carried on to evaluate Continuous density hidden Markov models (CDHMM),Dynamic time warping(DTW) and Biomimetic pattern recognition(BPR) for speech recognition, emphasizing the performance of each method across decreasing amounts of training samples as well as requirement of train time. The CDHMM system was implemented with 5 states per and BaumWelch reestimation are used for training and recognition． The reference templates for DTW system are the 第 5 頁(yè) training samples themselves. Both the CDHMM and DTW technique are implemented using the programs in Ref.[11]． We give in Table 2 the experiment results parison of BPR Basic algorithm， Dynamic time warping (DTW)and Hidden Markov models (HMMs) method． The HMMs system was based on Continuous density hidden Markov models(CDHMMs),and was implemented with 5 states per name. VI． Conclusions and Acknowledgments In this paper, A mandarin continuous speech recognition system based on BPR is ,a training samples selection method is also used to reduce the works scales. As a new genera l purpose theoretical model of pattern Recognition， BPR could be used in speech recognition too, and the experiment results show that it achieved a higher performance than HMM s and DTW. References [1]WangShoujue,“Blomimetic (Topological) pattern recognitionA new model of pattern recognition theory and its application”， Acta Electronics Sinica,(inChinese)， ,2020． [2]WangShoujue， ChenXu， “Blomimetic (Topological) pattern recognitionA new model of pattern recognition theory and its application”, Neural Networks, of the International Joint Conference on Neural Networks,July 2024,2020． [3]WangShoujue,ZhaoXingtao,“Biomimetic pattern recognition theory and its applications”， Chinese Journal of Electronics， ， ， ,2020． [4]Xu Jian． LiWeijun et a1,“Architecture research and hardware implementation on simplified neural puting system for face identification”， Neuarf Networks， 2020． Proceedings of the Intern atonal Joint Conference on Neural Networks， ,July 2024 2020． [5]Wang Zhihai,Mo Huayi et al,“A method of biomimetic pattern recognition for face recognition”, Neural Networks, of the International Joint Conference on Neural Networks,，,2024 July 2020． [6]WangShoujue,WangLiyan et a1,“A General Purpose Neuron Processor with DigitalAnalog Processing”，Chinese Journal of Electornics,1994． [7]Wang Shoujue,LiZhaozhou et a1,“Discussion on the basic mathematical models of neurons in general purpose neuroputer”， Acta Electronics Sinica(in Chinese),2020． [8]WangShoujue,Wang Bainan,“Analysis and theory of highdimension space geometry of artificial neural works”,Acta Electronics Sinica (in Chinese)， ， ,2020． [9]WangShoujue,Xujian et a1,“Multicamera humanface personal identiifcation system based on the biomimetic pattern recognition”， Acta Electronics Sinica (in Chinese)， ,2020． [10]Ryszard Engelking,Dimension Theory,PWNPolish Scientiifc Publishers—Warszawa,1978． [11]QiangHe,YingHe,Matlab Porgramming,Tsinghua University Press， 2020． 第 6 頁(yè) 中文翻譯 ： 電子學(xué)報(bào) 2020 年 7 月 15 卷第 3 期 基于仿生模式識(shí)別的非特定人連續(xù)語(yǔ)音識(shí)別系統(tǒng) 王守覺(jué) 秦虹 （中國(guó)，北京 100083，中科院半導(dǎo)體研究所人工神經(jīng)網(wǎng)絡(luò)實(shí)驗(yàn)室） 摘要：在非特定人語(yǔ)音識(shí)別中，隱馬爾科夫模型（ HMMs）是使用最多的技術(shù)，但是它
畢業(yè)設(shè)計(jì)相關(guān)推薦
鄂ICP備17016276號(hào)-1