Method of speech recognition

专利名称：Method of speech recognition发明人：Taisuke Watanabe,Tatsuya Kimura申请号：US07/777713申请日：19911021公开号：US05220609A公开日：19930615

摘要：A method of speech recognition includes the steps of predetermining a seriesof parameters a.sub.1, a.sub.2, . . . , a.sub.J representative of standard patterns ofspeeches of preset words, where the letter J denotes a predetermined natural number;deriving parameters x.sub.i representative of data of respective frames of an input signal,the adscript i denotes a frame number; calculating similarities d.sub.j. sup. (i) between theparameters a.sub.j and the parameters x.sub.i, where j=1, 2, . . . , J; calculating parametersR.sub.j.sup.(i) for the respective preset words by referring to the following recurrenceformulas:

R.sub.1.sup.(i) =d.sub.1.sup.(i)

< P>R.sub.j.sup.(i) =d.sub.j.sup.(i) +opt(R.sub.j-1.sup.(i-h),R.sub.j-1. sup.(i-h-1),R.sub.j-1.sup.(i-h-2), . . . ,R.sub.j-1.sup.(i-h-m))

where j=2, 3, . . . , J; the character h and m denote preset positive integers; and thecharacter opt denotes an operator of selecting an optimal member of terms in relatedparentheses; and selecting one of the preset words as a recognition result, the selectedword corresponding to an optimal value of the parameters R.sub.j.sup.(i).

申请人：MATSUSHITA ELECTRIC INDUSTRIAL CO., LTD.

代理机构：Lowe, Price, LeBlanc & Becker

更多信息请下载全文后查看