基于禁止搜索的非線性時(shí)間匹配優(yōu)化算法
An algorithm for optimizing the nonlinear time alignment based on tabu approach
-
摘要: 動(dòng)態(tài)時(shí)間規(guī)整算法DTW(Dynamic Time Warping)作為一種非線性時(shí)間匹配技術(shù)已成功地應(yīng)用于語音識(shí)別系統(tǒng)中。DTW算法使用動(dòng)態(tài)規(guī)劃技術(shù)來搜索兩個(gè)時(shí)間序列的最優(yōu)規(guī)整路徑,雖然這種算法計(jì)算量小,運(yùn)算時(shí)間較短,但只是一種局部?jī)?yōu)化算法。禁止搜索TS(Tabu Search)算法是一種具有短期記憶的廣義啟發(fā)式全局搜索技術(shù),適用于解決許多非線性優(yōu)化問題。本文將該技術(shù)用于語音識(shí)別系統(tǒng)中,提出了基于禁止搜索的非線性時(shí)間規(guī)整的優(yōu)化算法TSTW,使得時(shí)間規(guī)整函數(shù)盡可能逼近全局最優(yōu)。仿真結(jié)果表明,TSTW比DTW有更高的識(shí)別率,且運(yùn)行時(shí)間比遺傳時(shí)間規(guī)整算法GTW大大減少。Abstract: Dynamic Time Warping(DTW) has been widely used in speech recognition systems as a nonlinear time alignment technique. It uses the dynamic programming technique to search the optimal warping path for two time sequences. Although this algorithm needs less computation and shorter training and searching time, it is a local optimization algorithm. The Tabu Search (TS) algorithm is the generalized heuristic global search technique with short-time memory, and suitable for solving many nonlinear optimization problems. This paper applies this technique to speech recognition systems, and presents a new algorithm for optimizing time warping based on TS approach, which makes time warping functions optimized globally. Simulation results show that TSTW has better time warping performance than DTW and GTW.
-
F. Itakura, Minimum prediction residual principle applied to speech recognition, IEEE Trans. on ASSP, 1975, ASSP-23(2), 67-72.[2]S.K. Wong, C. W. Chau, W. A. Halang, Genetic algorithm for optimizing the nonlinear time alignment of automatic speech recognition system, IEEE Trans. on Industrial Electronic, 1996,IE-43(5), 559-566.[3]H.F. Silverman, D. P. Morgan, The application of dynamic programming to connected speech recognition, IEEE ASSP Magazine, 1990, 38(7), 7-24.[4]L.R. Rabiner, S. E. Levinson, Isolated and connected word recognition-theory and selected applications, IEEE Trans. on Communications, 1981, COM-29(5), 621-658.[5]F. Glover, M. Laguna, Tabu Search, Kluwer Academic Publishers, London, 1997, I chapter. ［6］H. Sakoe, S. Chiba, Dynamic programming algorithm optimization for spoken word recognition,IEEE Trans. on ASSP, 1978, ASSP-26(2), 43-49. -
計(jì)量
- 文章訪問數(shù): 2151
- HTML全文瀏覽量: 108
- PDF下載量: 498
- 被引次數(shù): 0
下載:
