利索-全球专利检索

  1. 登录
  2. 注册

搜索结果

147 条记录 (耗时 0.023 秒)

    Speech recognition method
    1.
    发明授权
    Speech recognition method 失效
    语音识别方法

    公开(公告)号：EP0191354B1

    公开(公告)日：1990-04-04

    申请号：EP86101089.0

    申请日：1986-01-28

    申请人： International Business Machines Corporation

    发明人： Bahl, Lalit Rai ,  Desouza, Peter Vincent ,  Mercer, Robert Leroy

    IPC分类号： G10L5/06

    CPC分类号： G10L15/144 ,  G10L2015/025

    摘要： For improving the efficiency of a speech recognition system based on a vocabulary of statistical word models, initially coarse word preselections are made for utterances, and are marked in a training session to indicate whether the selections were correct or wrong. Furthermore, each utterance is matched against each word preselected for it to obtain a probability measure for that combination as in a usual reccgnition process. Based on this knowledge of the correct or erroneous result of the coarse selection plus the probability measure, a discriminant analysis is made to determine, for the whole population of word models, how each phone in each word model should be weighted so that an optimum discrimination between similar words is achieved. The weighting coefficients thus obtained are stored with the word models and are later used, during actual speech recognition, to weight the probabilistic contribution of each phone for the final word selection.

    Normalization of speech signals
    8.
    发明公开
    Normalization of speech signals 失效
    语音信号的标准化。

    公开(公告)号：EP0182989A1

    公开(公告)日：1986-06-04

    申请号：EP85111566.7

    申请日：1985-09-13

    申请人： International Business Machines Corporation

    发明人： Picheny, Michael Alan

    IPC分类号： G10L3/02

    CPC分类号： G10L15/02

    摘要： 0 A method and a system are disclosed for normalizing a speech signal prior to a speech recognition process (29). In a preparatory procedure, a sample interval of speech is separated (15) into thirty-one frequency bands (17) and an amplitude histogram is generated (19) for each band. From these histograms, the 5% percentile amplitude value P(05) and the 95% percentile amplitude value P(95) are extracted (21) for each band and these values are stored (23) for later reference. For actual normalization, the current speech signal is also divided (15) into the same frequency band as in the preparatory procedure, and consecutive input amplitude values A(in) of each frequency band are modified (27), using (25) the percentile values of the respective band, to obtain output values according to