发明名称 ACTIONS ON DIGITAL DOCUMENT ELEMENTS FROM VOICE
摘要 A set of one or more terms can be derived from the voiced user input. It can be determined that the set of one or more terms corresponds to a specific digital document in a computer system, and that the set of one or more terms corresponds to at least one computer-executable command. The determination that the set of one or more terms corresponds to the at least one command can include analyzing the set of one or more terms using a document-specific natural language translation data structure. The translation data structure can be a computer-readable data structure that is identified in the computer system with the specific digital document in the computer system. It can be determined that the set of one or more terms corresponds to an element in the document, and the at least one command can be executed on the element in the document.
申请公布号 US2016078019(A1) 申请公布日期 2016.03.17
申请号 US201414485540 申请日期 2014.09.12
申请人 Microsoft Corporation 发明人 Napieralski Phillip E.;Gandhi Vishal;Huang Zhen;Krut Denys;Huang Xiaodi;Xu Kaili
分类号 G06F17/28 主分类号 G06F17/28
代理机构 代理人
主权项 1. A computer-implemented method, comprising: loading a document-specific natural language translation data structure into a computer memory in a computer system, the translation data structure being a computer-readable data structure that is identified in the computer system with a specific digital document in the computer system; receiving natural language voiced user input into the computer system; deriving, via the computer system, a set of one or more terms from the voiced user input; determining, via the computer system, that the set of one or more terms corresponds to the specific digital document in the computer system; determining, via the computer system, that the set of one or more terms corresponds to at least one computer-executable command, the determination that the set of one or more terms corresponds to the at least one command comprising analyzing the set of one or more terms using the document-specific natural language translation data structure; determining, via the computer system, that the set of one or more terms corresponds to an element in the document; and executing the at least one command on the element in the document.
地址 Redmond WA US