专利名称：System and method for personalization of
acoustic models for automatic speechrecognition
发明人：Andrej Ljolje,Diamantino Antonio
Caseiro,Alistair D. Conkie
申请号：US15595131申请日：20170515公开号：US09837072B2公开日：20171205
专利附图：
摘要：Disclosed herein are methods, systems, and computer-readable storage media
for automatic speech recognition. The method includes selecting a speaker independentmodel, and selecting a quantity of speaker dependent models, the quantity of speakerdependent models being based on available computing resources, the selected modelsincluding the speaker independent model and the quantity of speaker dependentmodels. The method also includes recognizing an utterance using each of the selectedmodels in parallel, and selecting a dominant speech model from the selected modelsbased on recognition accuracy using the group of selected models. The system includes aprocessor and modules configured to control the processor to perform the method. Thecomputer-readable storage medium includes instructions for causing a computing deviceto perform the steps of the method.
申请人：Nuance Communications, Inc.
地址：Burlington MA US
国籍：US
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