量子智能算法及其在语音识别中的应用
|
摘要
量子信息学是一门新兴的交叉学科,它在信息领域中有着独特的功能,在提高运算速度、确保信息安全、增大信息容量和提高检测精度等方面可突破现有经典信息系统的极限。特别是近年来,基于量子并行计算的量子智能算法有效地降低了一些经典难解算法的计算复杂度。本文的研究的主要工作有:
首先,研究了遗传算法、量子遗传算法的特性,提出了一种改进的实数编码量子遗传算法。通过对目标函数进行性能比较测试,实验结果表明,改进的实数编码量子遗传算法比传统二进制编码的量子遗传算法和实数编码量子遗传算法[19]都具有更好的全局搜索效果。
其次,本文将实数编码的量子遗传算法的思想引入到传统的粒子群优化算法中,提出一种新颖的量子粒子群优化算法。实验结果表明,本文提出的量子粒子群优化算法性能要优于Sun等人提出的量子粒子群算法和普通的粒子群优化算法。
再次,本文同时将免疫克隆选择的思想和混沌变异引入到量子遗传算法中,提出了混沌量子免疫遗传算法。实验结果表明,本文提出的混沌量子免疫遗传算法搜寻性能要优于免疫克隆选择算法和普通的遗传算法。
最后,将上述量子智能算法应用于HMM的孤立词语音识别系统。采用将量子智能算法引入到HMM的初始化聚类分析中。实验表明,基于量子智能算法的HMM语音识别系统比传统的HMM语音识别系统具有更好的识别能力。
Quantum information science is a rising cross discipline. Due to its unique features in the information field, it may break the limitation of classic information system. Particularly in recent years, quantum intelligence algorithms based on the parallel quantum computation can effectively simplify computation complexity of some classic algorithms.
First of all, we improved the quantum genetic algorithm. Simulation results show that the improved quantum genetic algorithm performs obviously superior to the classic quantum genetic algorithm.
Secondly, a novel Quantum Particle Swarm Optimization (QPSO) algorithm is proposed, which combine Particle Swarm Optimization (PSO) with real coded quantum genetic algorithm. Simulation results demonstrate the superiority of the proposed QPSO.
Thirdly, this paper combined QGA with immune clone select algortithm and chaotic and proposed a new algorithm-Chaotic Immune Quantum Genetic Algorithm (CIQGA). Simulation results show the superiority of the proposed CIQGA.
Finally, all the proposed algorithms in this paper were applied to initialization of cluster analysis of Speech Recognition System based on Hidden Markov Models (HMM). The experiment results show that 2.5% error reduction is achieved on our proposed algorithms based speaker-independent digit speech recognition system compared to K-means clustering algorithm based speech recognition system.
首先,研究了遗传算法、量子遗传算法的特性,提出了一种改进的实数编码量子遗传算法。通过对目标函数进行性能比较测试,实验结果表明,改进的实数编码量子遗传算法比传统二进制编码的量子遗传算法和实数编码量子遗传算法[19]都具有更好的全局搜索效果。
其次,本文将实数编码的量子遗传算法的思想引入到传统的粒子群优化算法中,提出一种新颖的量子粒子群优化算法。实验结果表明,本文提出的量子粒子群优化算法性能要优于Sun等人提出的量子粒子群算法和普通的粒子群优化算法。
再次,本文同时将免疫克隆选择的思想和混沌变异引入到量子遗传算法中,提出了混沌量子免疫遗传算法。实验结果表明,本文提出的混沌量子免疫遗传算法搜寻性能要优于免疫克隆选择算法和普通的遗传算法。
最后,将上述量子智能算法应用于HMM的孤立词语音识别系统。采用将量子智能算法引入到HMM的初始化聚类分析中。实验表明,基于量子智能算法的HMM语音识别系统比传统的HMM语音识别系统具有更好的识别能力。
Quantum information science is a rising cross discipline. Due to its unique features in the information field, it may break the limitation of classic information system. Particularly in recent years, quantum intelligence algorithms based on the parallel quantum computation can effectively simplify computation complexity of some classic algorithms.
First of all, we improved the quantum genetic algorithm. Simulation results show that the improved quantum genetic algorithm performs obviously superior to the classic quantum genetic algorithm.
Secondly, a novel Quantum Particle Swarm Optimization (QPSO) algorithm is proposed, which combine Particle Swarm Optimization (PSO) with real coded quantum genetic algorithm. Simulation results demonstrate the superiority of the proposed QPSO.
Thirdly, this paper combined QGA with immune clone select algortithm and chaotic and proposed a new algorithm-Chaotic Immune Quantum Genetic Algorithm (CIQGA). Simulation results show the superiority of the proposed CIQGA.
Finally, all the proposed algorithms in this paper were applied to initialization of cluster analysis of Speech Recognition System based on Hidden Markov Models (HMM). The experiment results show that 2.5% error reduction is achieved on our proposed algorithms based speaker-independent digit speech recognition system compared to K-means clustering algorithm based speech recognition system.
引文
[1] Man K.F, Tang K.S, Kwong S. Genetic algorithms: concepts and applications in engineering design [J]. IEEE Transactions, Vol.43(5), 519-534, 1996
[2] Charyaratana N, Zalzala A.M.S. Recent developments in evolutionary and genetic algorithms: theory and applications[J].GALESIA 97 Second International Conference, 9,270-277, 1997
[3] Holland J H. Genetic algorithm and classifier systems:foundations and their applications[C]. International Conference on Genetic Algorithms,Hillsdale,NJ:Lawrence Erlbaum Associates, 82-89, 1987
[4] I.E.Telatar. Capacity of multi-antenna Gaussian channels. European Transactions on Telecommunications,Vol.10(6), 585–595, 1999
[5] Gruska J. Quantum Computing[M]. McGraw-Hill, 1999
[6]李承祖等编著.量子通信和量子计算[M].国防科技大学出版社,2000
[7]曾谨言著.量子力学[M]卷I(第三版).科学出版社,2000
[8]李承祖.量子通信和量子计算[M].北京国防科技大学出版社, 101-145, 255-349, 2000
[9]王小平,曹立明.遗传算法-理论、应用与软件实现[M].西安交通大学出版社,2002
[10]杨淑媛,焦李成,刘芳.量子进化算法[J].工程数学学报, Vol.23(2),235-246, 2006
[11] Shor P.W. Polynomial-Time Algorithms for Prime Factorization and Discrete logarithms on Quantum Computer[J]. SIAM Journal on Computing, Vol.26(5): 1484-1509, 1997
[12] Deutsch D. Quantum computational networks[C]. Proc. of the Roy.Soc. of Lon.A, Vol.425:73-90,1989
[13]杨淑媛,刘芳,焦李成.一种基于量子染色体的遗传算法[J].西安电子科技大学学报(自然科学报). Vol.31(1), 2004
[14] Han K.-H., Park K.-H., Lee C.-H., Kim J.-H. Parallel Quantum-inspired Genetic Algorithm for combinatorial Optimization Problem[C]. Proc. of Congress on Evolutionary Computation, 1422-1429, 2001
[15] Han K.-H., Kim J.-H. Quantum-Inspired Evolutionary Algorithm for a Class of Combinatorial Optimization[J]. IEEE Trans. on Evolutionary Computation , Vol. 6(6), 580-593, 2002
[16]李飞.量子神经网络及其在信号处理中的应用研究[D].南京邮电大学,98-114, 2005
[17]杨俊安,庄镇泉,史亮.多宇宙并行量子遗传算法[J].电子学报,Vol.32(6), 923-928, 2004
[18] Han K.-H., Kim J.-H. Quantum-Inspired Evolutionary Algorithm for a Class of Combinatorial Optimization[J]. IEEE Trans. on Evolutionary Computation, Vol.6(6): 580-593, 2002
[19] Rui Zhang, Hui Gao. Real-coded quantum evolutinary algorithm for complex function with high dimension[C]. Proc.of IEEE International Conference on Mechatronics and Automation, 2974-2979, 2007
[20]陈辉,张家树,张超.实数编码混沌量子遗传算法[J].控制与决策, Vol.20(11):1300-1303, 2005
[21] KENNEDY J, EBERHART R C. Particle swarms optimization[C]. International Conference on Neural Networks, USA: 4:1942-1948, 1995
[22] ENGELBRECHT A P.. Cooperative learning in neural networks using particle swarm optimization[J]. South African Computer Journal, 11:84-89, 2000
[23] Sun J., Feng B., Xu WB.. Particle swarm optimization with particles having quantum behavior[C]. Congress on Evolution Computation, Piscataway NJ: 325-331, 2004
[24]黄席樾等.现代智能算法理论及应用[M].北京,科学出版社, 2005
[25]黄友锐.智能优化算法及其应用[M].北京,国防工业出版社, 2008
[26] Timmis J., Neal M.. Hunt J.Artificial immune system for data analysis[J]. Biosystem, 55,143-150, 2000
[27]张彤,王宏伟,王子才.变尺度混沌优化方法及其应用[J].控制与决策, Vol.12(4):613-615, 1997
[28]马明,张杰,王建宇等.语音识别中隐马尔可夫模型初值的估计[J].数据采集与处理,Vol.12(2):96-100, 1997
[29] Rabiner L R.. A tutorial Hidden Markov Model and selected applications in speech recognition[C]. Proc. of the IEEE, Vol.77(2):257-285, 1989
[30]胡航.语音信号处理[M].哈尔滨工业大学出版社, 2005
[31]姚天任.数字语音处理[M].华中理工大学出版社,371-383,1989
[32]苗苗,马海武. HMM在语音识别系统的应用[J].现代电子技术,16:64-66, 2006
[33]谢锦辉.HMM及其在语音处理中的应用[M].华中理工大学出版社,1995
[34]何强,何英. MATLAB扩展编程[M].清华大学出版社, 330-372,2002
[35]詹新明,黄南山,杨灿.语音识别技术研究进展[J].现代计算机(专业版), 9,2008
[36]朱淑琴.语音识别系统关键技术研究[硕士学位论文] [D].西安电子科技大学, 2004
[37]蔡妍.语音信号端点检测方法的研究[硕士学位论文][D].江南大学, 2008
[38]李景川,董慧颖.一种改进的基于短时能量的端点检测算法[J].沈阳理工大学学报, (06),2008
[39]赖玉霞,刘建平,杨国兴.基于遗传算法的K均值聚类分析[J].计算机工程,Vol.34(20):200-202. 2008
[40]王一平.用遗传算法改进HMM的语音识别算法研究[硕士学位论文][D].太原理工大学,2007
[41] Narayanna A., Moore M..Quantum-inspired Genetic Algorithms[C]. IEEE International Conference on Evolutional Evolution, 61-66, 1996
[42] Han K.-H., Kim J.-H..Genetic Quantum Algorithm and its Application to Combinatorial Optimization[C], IEEE Congress on Evolutionary Computation, 1354-1360, 2000
[43] Han K.-H., Park K.-H., Lee C.-H., Kim J.-H.. Parallel Quantum-inspired Genetic Algorithm for combinatorial Optimization Problem[C], IEEE Congress on Evolutionary Computation, 1422-1429, 2001
[44] Han K.-H., Kim J.-H.. Quantum-Inspired Evolutionary Algorithm for a Class of Combinatorial Optimization[C]. IEEE Congress on Evolutionary Computation, Vol.6(6): 580-593, 2002
[2] Charyaratana N, Zalzala A.M.S. Recent developments in evolutionary and genetic algorithms: theory and applications[J].GALESIA 97 Second International Conference, 9,270-277, 1997
[3] Holland J H. Genetic algorithm and classifier systems:foundations and their applications[C]. International Conference on Genetic Algorithms,Hillsdale,NJ:Lawrence Erlbaum Associates, 82-89, 1987
[4] I.E.Telatar. Capacity of multi-antenna Gaussian channels. European Transactions on Telecommunications,Vol.10(6), 585–595, 1999
[5] Gruska J. Quantum Computing[M]. McGraw-Hill, 1999
[6]李承祖等编著.量子通信和量子计算[M].国防科技大学出版社,2000
[7]曾谨言著.量子力学[M]卷I(第三版).科学出版社,2000
[8]李承祖.量子通信和量子计算[M].北京国防科技大学出版社, 101-145, 255-349, 2000
[9]王小平,曹立明.遗传算法-理论、应用与软件实现[M].西安交通大学出版社,2002
[10]杨淑媛,焦李成,刘芳.量子进化算法[J].工程数学学报, Vol.23(2),235-246, 2006
[11] Shor P.W. Polynomial-Time Algorithms for Prime Factorization and Discrete logarithms on Quantum Computer[J]. SIAM Journal on Computing, Vol.26(5): 1484-1509, 1997
[12] Deutsch D. Quantum computational networks[C]. Proc. of the Roy.Soc. of Lon.A, Vol.425:73-90,1989
[13]杨淑媛,刘芳,焦李成.一种基于量子染色体的遗传算法[J].西安电子科技大学学报(自然科学报). Vol.31(1), 2004
[14] Han K.-H., Park K.-H., Lee C.-H., Kim J.-H. Parallel Quantum-inspired Genetic Algorithm for combinatorial Optimization Problem[C]. Proc. of Congress on Evolutionary Computation, 1422-1429, 2001
[15] Han K.-H., Kim J.-H. Quantum-Inspired Evolutionary Algorithm for a Class of Combinatorial Optimization[J]. IEEE Trans. on Evolutionary Computation , Vol. 6(6), 580-593, 2002
[16]李飞.量子神经网络及其在信号处理中的应用研究[D].南京邮电大学,98-114, 2005
[17]杨俊安,庄镇泉,史亮.多宇宙并行量子遗传算法[J].电子学报,Vol.32(6), 923-928, 2004
[18] Han K.-H., Kim J.-H. Quantum-Inspired Evolutionary Algorithm for a Class of Combinatorial Optimization[J]. IEEE Trans. on Evolutionary Computation, Vol.6(6): 580-593, 2002
[19] Rui Zhang, Hui Gao. Real-coded quantum evolutinary algorithm for complex function with high dimension[C]. Proc.of IEEE International Conference on Mechatronics and Automation, 2974-2979, 2007
[20]陈辉,张家树,张超.实数编码混沌量子遗传算法[J].控制与决策, Vol.20(11):1300-1303, 2005
[21] KENNEDY J, EBERHART R C. Particle swarms optimization[C]. International Conference on Neural Networks, USA: 4:1942-1948, 1995
[22] ENGELBRECHT A P.. Cooperative learning in neural networks using particle swarm optimization[J]. South African Computer Journal, 11:84-89, 2000
[23] Sun J., Feng B., Xu WB.. Particle swarm optimization with particles having quantum behavior[C]. Congress on Evolution Computation, Piscataway NJ: 325-331, 2004
[24]黄席樾等.现代智能算法理论及应用[M].北京,科学出版社, 2005
[25]黄友锐.智能优化算法及其应用[M].北京,国防工业出版社, 2008
[26] Timmis J., Neal M.. Hunt J.Artificial immune system for data analysis[J]. Biosystem, 55,143-150, 2000
[27]张彤,王宏伟,王子才.变尺度混沌优化方法及其应用[J].控制与决策, Vol.12(4):613-615, 1997
[28]马明,张杰,王建宇等.语音识别中隐马尔可夫模型初值的估计[J].数据采集与处理,Vol.12(2):96-100, 1997
[29] Rabiner L R.. A tutorial Hidden Markov Model and selected applications in speech recognition[C]. Proc. of the IEEE, Vol.77(2):257-285, 1989
[30]胡航.语音信号处理[M].哈尔滨工业大学出版社, 2005
[31]姚天任.数字语音处理[M].华中理工大学出版社,371-383,1989
[32]苗苗,马海武. HMM在语音识别系统的应用[J].现代电子技术,16:64-66, 2006
[33]谢锦辉.HMM及其在语音处理中的应用[M].华中理工大学出版社,1995
[34]何强,何英. MATLAB扩展编程[M].清华大学出版社, 330-372,2002
[35]詹新明,黄南山,杨灿.语音识别技术研究进展[J].现代计算机(专业版), 9,2008
[36]朱淑琴.语音识别系统关键技术研究[硕士学位论文] [D].西安电子科技大学, 2004
[37]蔡妍.语音信号端点检测方法的研究[硕士学位论文][D].江南大学, 2008
[38]李景川,董慧颖.一种改进的基于短时能量的端点检测算法[J].沈阳理工大学学报, (06),2008
[39]赖玉霞,刘建平,杨国兴.基于遗传算法的K均值聚类分析[J].计算机工程,Vol.34(20):200-202. 2008
[40]王一平.用遗传算法改进HMM的语音识别算法研究[硕士学位论文][D].太原理工大学,2007
[41] Narayanna A., Moore M..Quantum-inspired Genetic Algorithms[C]. IEEE International Conference on Evolutional Evolution, 61-66, 1996
[42] Han K.-H., Kim J.-H..Genetic Quantum Algorithm and its Application to Combinatorial Optimization[C], IEEE Congress on Evolutionary Computation, 1354-1360, 2000
[43] Han K.-H., Park K.-H., Lee C.-H., Kim J.-H.. Parallel Quantum-inspired Genetic Algorithm for combinatorial Optimization Problem[C], IEEE Congress on Evolutionary Computation, 1422-1429, 2001
[44] Han K.-H., Kim J.-H.. Quantum-Inspired Evolutionary Algorithm for a Class of Combinatorial Optimization[C]. IEEE Congress on Evolutionary Computation, Vol.6(6): 580-593, 2002