Word dependent N-best search method
作者： Richard M. Schwartz , Stephen C. Austin
DOI:
关键词:
摘要: As a step in finding the one most likely word sequence spoken language system, an N-best search is conducted to find N sentence hypotheses. During search, theories are distinguished based only on previous word. At each state within word, total probability calculated for of few words. end score recorded theory, together with name sentence, recursive traceback performed derive list best sentences.
lens.org 参考文章(4)
R. Schwartz, Y.-L. Chow, The N-best algorithms: an efficient and exact procedure for finding the N most likely sentence hypotheses international conference on acoustics, speech, and signal processing. pp. 81- 84 ,(1990) , 10.1109/ICASSP.1990.115542
Lalit R. Bahl, Determination of phone weights for markov models in a speech recognition system Journal of the Acoustical Society of America. ,vol. 86, pp. 2053- 2053 ,(1985) , 10.1121/1.398494
F.K. Soong, E.-F. Huang, A tree-trellis based fast search for finding the N-best sentence hypotheses in continuous speech recognition international conference on acoustics, speech, and signal processing. pp. 705- 708 ,(1991) , 10.1109/ICASSP.1991.150437
Lalit R. Bahl, Frederick Jelinek, Apparatus and method for determining a likely word sequence from labels generated by an acoustic processor Journal of the Acoustical Society of America. ,vol. 87, pp. 927- 927 ,(1986) , 10.1121/1.398847