专利名称：Multiple models integration for multi-environment speech recognition
发明人：Rahim, Mazin G.申请号：EP98108805.7申请日：19980514公开号：EP0881625B1公开日：20050810
摘要：A speech recognition system which effectively recognizes unknown speech frommultiple acoustic environments includes a set of secondary models, each associated withone or more particular acoustic environments, integrated with a base set of recognitionmodels. The speech recognition system is trained by making a set of secondary models ina first stage of training, and integrating the set of secondary models with a base set ofrecognition models in a second stage of training.
申请人：AT & T CORP
地址：US
国籍：US
代理机构：Modiano, Guido, Dr.-Ing.
更多信息请下载全文后查看
因篇幅问题不能全部显示，请点此查看更多更全内容