公开(公告)号：WO2015009586A3

公开(公告)日：2015-01-22

申请号：PCT/US2014/046457

申请日：2014-07-14

Applicant: MICROSOFT CORPORATION

Inventor： SINGH, Prabhdeep ,  GANJAM, Kris ,  GULWANI, Sumit ,  MARRON, Mark ,  JU, Yun-Cheng ,  CHAKRABARTI, Kaushik

IPC: G06F17/24

Abstract: Described herein are various technologies pertaining to performing an operation relative to tabular data based upon voice input. An ASR system includes a language model that is customized based upon content of the tabular data. The ASR system receives a voice signal that is representative of speech of a user. The ASR system creates a transcription of the voice signal based upon the ASR being customized with the content of the tabular data. The operation relative to the tabular data is performed based upon the transcription of the voice signal.

公开(公告)号：WO2014189987A1

公开(公告)日：2014-11-27

申请号：PCT/US2014/038867

申请日：2014-05-21

Applicant: MICROSOFT CORPORATION

Inventor： ZEIGLER, Andrew Stephen ,  KIM, Michael H. ,  BENSON, Rodger ,  SARIN, Raman ,  JU, Yun-Cheng

IPC: G06F3/16

CPC classification number: G10L21/10 ,  G06F3/167 ,  G10L15/08 ,  G10L15/19 ,  G10L15/22 ,  G10L2015/223

Abstract: A disambiguation process for a voice interface for web pages or other documents. The process identifies interactive elements such as links, obtains one or more phrases of each interactive element, such as link text, title text and alternative text for images, and adds the phrases to a grammar which is used for speech recognition. A group of interactive elements are identified as potential best matches to a voice command when there is no single, clear best match. The disambiguation process modifies a display of the document to provide unique labels for each interactive element in the group, and the user is prompted to provide a subsequent spoke command to identify one of the unique labels. The selected unique label is identified and a click event is generated for the corresponding interactive element.

Abstract translation: 用于网页或其他文档的语音界面的消歧过程。 该过程识别诸如链接的交互元素，获得每个交互元素的一个或多个短语，例如链接文本，标题文本和用于图像的备选文本，并且将短语添加到用于语音识别的语法中。 当没有单一，明确的最佳匹配时，一组交互式元素被识别为与语音命令的潜在最佳匹配。 消歧过程修改文档的显示，为组中的每个交互元素提供唯一的标签，并且提示用户提供后续的辐条命令来识别唯一标签之一。 识别所选择的唯一标签，并为相应的交互式元素生成点击事件。

公开(公告)号：WO2014189988A1

公开(公告)日：2014-11-27

申请号：PCT/US2014/038868

申请日：2014-05-21

Applicant: MICROSOFT CORPORATION

Inventor： ZEIGLER, Andrew Stephen ,  KIM, Michael H. ,  BENSON, Rodger ,  SARIN, Raman ,  JU, Yun-Cheng

IPC: G10L15/22

CPC classification number: G10L15/26 ,  G10L15/22 ,  G10L2015/228

Abstract: A voice interface for web pages or other documents identifies interactive elements such as links, obtains one or more phrases of each interactive element, such as link text, title text and alternative text for images, and adds the phrases to a grammar which is used for speech recognition. A click event is generated for an interactive element having a phrase which is a best match for the voice command of a user. In one aspect, the phrases of currently-displayed elements of the document are used for speech recognition. In another aspect, phrases which are not displayed, such as title text and alternative text for images, are used in the grammar. In another aspect, updates to the document are detected and the grammar is updated accordingly so that the grammar is synchronized with the current state of the document.

Abstract translation: 用于网页或其他文档的语音界面标识诸如链接的交互元素，获得每个交互元素的一个或多个短语，例如链接文本，标题文本和用于图像的备选文本，并将短语添加到用于 语音识别。 为具有与用户的语音命令最佳匹配的短语的交互式元素生成点击事件。 在一个方面，文档中当前显示的元素的短语用于语音识别。 在另一方面，在语法中使用未显示的短语，例如标题文本和用于图像的备选文本。 在另一方面，检测到文档的更新并相应地更新语法，使得语法与文档的当前状态同步。

公开(公告)号：WO2014130745A2

公开(公告)日：2014-08-28

申请号：PCT/US2014/017521

申请日：2014-02-20

Applicant: MICROSOFT CORPORATION

Inventor： JU, Yun-Cheng ,  PHILIPOSE, Matthai ,  HAN, Seungyeop

IPC: G10L15/22

CPC classification number: G10L15/063 ,  G06F17/2785 ,  G10L15/02 ,  G10L15/065 ,  G10L15/22 ,  G10L2015/223 ,  G10L2015/225

Abstract: A development system is described for facilitating the development of a spoken natural language (SNL) interface. The development system receives seed templates from a developer, each of which provides a command phrasing that can be used to invoke a function, when spoken by an end user. The development system then uses one or more development resources, such as a crowdsourcing system and a paraphrasing system, to provide additional templates. This yields an extended set of templates. A generation system then generates one or more models based on the extended set of templates. A user device may install the model(s) for use in interpreting commands spoken by an end user. When the user device recognizes a command, it may automatically invoke a function associated with that command. Overall, the development system provides an easy-to-use tool for producing an SNL interface.

Abstract translation: 描述开发系统以促进口语自然语言（SNL）界面的开发。 开发系统从开发人员那里接收种子模板，每个开发人员在最终用户说出时都会提供一个可用于调用函数的命令语句。 开发系统然后使用一个或多个开发资源（例如众包系统和释义系统）来提供额外的模板。 这产生了一组扩展的模板。 生成系统然后基于扩展的模板集合生成一个或多个模型。 用户设备可以安装模型以用于解释由最终用户说出的命令。 当用户设备识别命令时，它可以自动调用与该命令相关的功能。 总的来说，开发系统为生成SNL接口提供了一个易于使用的工具。

公开(公告)号：WO2012173902A3

公开(公告)日：2012-12-20

申请号：PCT/US2012/041798

申请日：2012-06-10

Applicant: MICROSOFT CORPORATION

Inventor： JU, Yun-Cheng ,  TASHEV, Ivan J. ,  LI, Xiao ,  HAWKINS, Dax ,  SOEMO, Thomas ,  KIM, Michael H.

IPC: G10L15/08 ,  G10L15/04

Abstract: A multimedia system configured to receive user input in the form of a spelled character sequence is provided. In one implementation, a spell mode is initiated, and a user spells a character sequence. The multimedia system performs spelling recognition and recognizes a sequence of character representations having a possible ambiguity resulting from any user and/or system errors. The sequence of character representations with the possible ambiguity yields multiple search keys. The multimedia system performs a fuzzy pattern search by scoring each target item from a finite dataset of target items based on the multiple search keys. One or more relevant items are ranked and presented to the user for selection, each relevant item being a target item that exceeds a relevancy threshold. The user selects the indented character sequence from the one or more relevant items.

公开(公告)号：WO2009023564A1

公开(公告)日：2009-02-19

申请号：PCT/US2008/072620

申请日：2008-08-08

Applicant: MICROSOFT CORPORATION

Inventor： TASHEV, Ivan ,  SELTZER, Michael, Lewis ,  JU, Yun-Cheng ,  ACERO, Alex

IPC: G06Q50/00

CPC classification number: G01C21/3644 ,  G01C21/3679

Abstract: The presentation of location information to a user that is distracted by traveling can result in the user quickly forgetting, or never even comprehending, key parts of the location information, such as the street number. Identification can be made of intersections and points of interest near the user's destination, which can then be provided instead of, or in addition to, the address, thereby increasing user comprehension and retention, especially when distracted. Map data can be parsed into addresses, intersections and points of interest databases. These databases can be accessed to identify proximate intersections and points of interest, which can then be filtered and subsequently ranked to identify one intersection, one point of interest, or both, that can be presented to the user to aid the user in comprehending and retaining the location information even when distracted.

Abstract translation: 通过旅行分散给用户的位置信息的呈现可能导致用户快速地忘记甚至不理解诸如街道号码的位置信息的关键部分。 识别可以由用户目的地附近的交叉点和兴趣点组成，然后可以提供地址，也可以除了地址之外，还可以提供用户的理解和保留，特别是在分心时。 地图数据可以解析为地址，交叉点和兴趣点数据库。 可以访问这些数据库以识别最近的交叉点和兴趣点，然后可以对这些数据进行过滤并随后进行排序以识别一个交点，一个兴趣点或二者，可以呈现给用户以帮助用户理解和保留 位置信息即使分心。

公开(公告)号：WO2009009312A2

公开(公告)日：2009-01-15

申请号：PCT/US2008/068487

申请日：2008-06-27

Applicant: MICROSOFT CORPORATION

Inventor： JU, Yun-Cheng ,  WANG, Ye-Yi

IPC: G06F17/30

CPC classification number: G06F17/30646

Abstract: A directory assistance system includes a directory database and a search engine. The search engine is configured to search the directory database for a first set of residential listings based on at least one first search term. A second search term is received that is related to a cohabitant of the listing to be found. At least one search result is selected that satisfies the second search term.

Abstract translation: 目录辅助系统包括目录数据库和搜索引擎。 搜索引擎被配置为基于至少一个第一搜索项搜索目录数据库中的第一组住宅列表。 接收与待发现的列表的同居者相关的第二搜索词。 选择满足第二搜索项的至少一个搜索结果。

公开(公告)号：WO1998000834A1

公开(公告)日：1998-01-08

申请号：PCT/US1997011683

申请日：1997-06-27

Applicant: MICROSOFT CORPORATION

Inventor： MICROSOFT CORPORATION ,  HON, Hsiao-Wuen ,  HUANG, Xuedong, D. ,  HWANG, Mei-Yuh ,  JIANG, Li ,  JU, Yun-Cheng ,  MAHAJAN, Milind, V. ,  ROZAK, Michael, J.

IPC: G10L03/00

CPC classification number: G10L15/063 ,  G10L2015/0635

Abstract: A method and system for dynamically selecting words for training a speech recognition system. The speech recognition system models each phoneme using a hidden Markov model and represents each word as a sequence of phonemes. The training system ranks each phoneme for each frame according to the probability that the corresponding codeword will be spoken as part of the phoneme. The training system collects spoken utterances for which the corresponding word is known. The training system then aligns the codewords of each utterance with the phoneme that it is recognized to be part of. The training system then calculates an average rank for each phoneme using the aligned codewords for the aligned frames. Finally, the training system selects words for training that contain phonemes with a low rank.

Abstract translation: 一种用于动态选择用于训练语音识别系统的单词的方法和系统。 语音识别系统使用隐马尔科夫模型对每个音素进行建模，并将每个单词表示为一系列音素。 训练系统根据将相应的码字作为音素的一部分被说出的概率，对每个帧的每个音素进行排序。 训练系统收集对应词语已知的口语说话。 然后，训练系统将每个话语的码字与被认为是其一部分的音素对齐。 训练系统然后使用对齐的帧的对齐码字来计算每个音素的平均等级。 最后，训练系统选择含有低等级音素的训练词。

公开(公告)号：EP0907949B1

公开(公告)日：2001-10-31

申请号：EP97934052.8

申请日：1997-06-27

Applicant: MICROSOFT CORPORATION

Inventor： HON, Hsiao-Wuen ,  HUANG, Xuedong, D. ,  HWANG, Mei-Yuh ,  JIANG, Li ,  JU, Yun-Cheng ,  MAHAJAN, Milind, V. ,  ROZAK, Michael, J.

IPC: G10L15/06

CPC classification number: G10L15/063 ,  G10L2015/0635

Abstract: A method and system for dynamically selecting words for training a speech recognition system. The speech recognition system models each phoneme using a hidden Markov model and represents each word as a sequence of phonemes. The training system ranks each phoneme for each frame according to the probability that the corresponding codeword will be spoken as part of the phoneme. The training system collects spoken utterances for which the corresponding word is known. The training system then aligns the codewords of each utterance with the phoneme that it is recognized to be part of. The training system then calculates an average rank for each phoneme using the aligned codewords for the aligned frames. Finally, the training system selects words for training that contain phonemes with a low rank.

公开(公告)号：EP0907949A1

公开(公告)日：1999-04-14

申请号：EP97934052.0

申请日：1997-06-27

Applicant: MICROSOFT CORPORATION

Inventor： HON, Hsiao-Wuen ,  HUANG, Xuedong, D. ,  HWANG, Mei-Yuh ,  JIANG, Li ,  JU, Yun-Cheng ,  MAHAJAN, Milind, V. ,  ROZAK, Michael, J.

IPC: G10L15

CPC classification number: G10L15/063 ,  G10L2015/0635

Abstract: A method and system for dynamically selecting words for training a speech recognition system. The speech recognition system models each phoneme using a hidden Markov model and represents each word as a sequence of phonemes. The training system ranks each phoneme for each frame according to the probability that the corresponding codeword will be spoken as part of the phoneme. The training system collects spoken utterances for which the corresponding word is known. The training system then aligns the codewords of each utterance with the phoneme that it is recognized to be part of. The training system then calculates an average rank for each phoneme using the aligned codewords for the aligned frames. Finally, the training system selects words for training that contain phonemes with a low rank.