专利快速检索-快速检索全球专利，免费商用专利数据库-IPRDB

1. 发明专利

JP2014013399A Method and system for processing dictated information 审中-公开
标题翻译：处理信息的方法和系统
公开(公告)号：JP2014013399A
公开(公告)日：2014-01-23
申请号：JP2013165440
申请日：2013-08-08
申请人： Nuance Communications Austria Gmbh , ニュアンスコミュニケーションズオーストリアゲーエムベーハー
发明人： MEHMET M OEZ
IPC分类号： G10L15/22 , G10L15/00
CPC分类号： G10L15/265 , G06F17/243 , G10L15/26 , G16H10/20
摘要： PROBLEM TO BE SOLVED: To provide an improved method for processing dictated information.SOLUTION: A method for processing dictated information into a dynamic form comprises steps of: reading an electronic document having a document structure comprising at least one text field; presenting a user with at least a part of the electronic document; receiving speech input of a first section of speech; processing the speech input into dictated text in a speech recognition engine 6; associating the dictated text with a first text field; and adjusting display of the electronic document at least partly on the basis of the speech input and controlling the amount of the electronic document to be shown to the user.
摘要翻译：要解决的问题：提供一种用于处理规定信息的改进方法。解决方案：将指令信息处理成动态形式的方法包括以下步骤：读取具有至少一个文本字段的文档结构的电子文档; 向用户呈现所述电子文档的至少一部分; 接收第一语音的语音输入; 将语音输入处理成语音识别引擎6中的指定文本; 将规定的文本与第一文本字段相关联; 以及至少部分地基于语音输入来调整电子文档的显示，并且控制要显示给用户的电子文档的数量。

2. 发明专利

JP2013122599A Method and system for dynamic creation of context 审中-公开
标题翻译：方法和系统的动态创造的背景
公开(公告)号：JP2013122599A
公开(公告)日：2013-06-20
申请号：JP2013000242
申请日：2013-01-04
申请人： Nuance Communications Austria Gmbh , ニュアンスコミュニケーションズオーストリアゲーエムベーハー
发明人： GERHARD GROBAUER , MIKLOS PAPAI
IPC分类号： G10L15/22 , G06F17/21 , G10L15/00 , G10L15/10 , G10L15/18 , G10L15/183
CPC分类号： G10L15/26 , G10L15/183
摘要： PROBLEM TO BE SOLVED: To provide sub contexts of corresponding sections of text of a speech based document in a method and system for a speech recognition system, which comprises an electronic document being a speech based document comprising one or more sections of text recognized or translated from sections of speech.SOLUTION: Sections of speech are dictated by an author. The method for a speech recognition system comprises the steps of: dynamically creating and adapting sub contexts by a speech recognizer; and associating the sub context with the sections of text.
摘要翻译：要解决的问题：为了在语音识别系统的方法和系统中提供基于语音的文档的相应文本段的子上下文，其包括作为基于语音的文档的电子文档，该文档包括文本的一个或多个部分从语篇中识别或翻译。
解决方案：演讲部分由作者决定。语音识别系统的方法包括以下步骤：通过语音识别器动态创建和调整子语境; 并将子上下文与文本的部分相关联。版权所有（C）2013，JPO＆INPIT

3. 发明专利

JP2011100169A Speech recognition method for activating hyperlink of internet page 审中-公开
标题翻译：用于激活互联网页面的高清晰度的语音识别方法
公开(公告)号：JP2011100169A
公开(公告)日：2011-05-19
申请号：JP2011038654
申请日：2011-02-24
申请人： Nuance Communications Austria Gmbh , ニュアンスコミュニケーションズオーストリアゲーエムベーハー
发明人： HOLLEY MARTIN , KUBESCH DIETER
IPC分类号： G10L15/00 , G10L15/28 , G06F17/30 , G10L15/06 , G10L15/08 , G10L15/22
CPC分类号： G10L15/22 , G06F17/30899
摘要： PROBLEM TO BE SOLVED: To provide an improved speech recognition method, an improved speech recognition device and an improved computer program product. SOLUTION: The method of activating a hyperlink of an Internet page is provided, the hyperlink including a first hypertext and a universal resource locator (URL). The method includes steps of: providing an extra hyperlink assigned to the hyperlink, the extra hyperlink including a second hypertext associated with the URL of the hyperlink, the second hypertext being capable of being recognized by automatic speech recognition; receiving a spoken command from a user; performing speech recognition on the spoken command; and if the spoken command matches the second hypertext, activating the hyperlink via the associated URL. COPYRIGHT: (C)2011,JPO&INPIT
摘要翻译：要解决的问题：提供改进的语音识别方法，改进的语音识别装置和改进的计算机程序产品。提供了激活因特网页面的超链接的方法，超链接包括第一超文本和通用资源定位符（URL）。该方法包括以下步骤：提供分配给该超链接的额外超链接，该额外超链接包括与超链接的URL相关联的第二超文本，该第二超文本能够通过自动语音识别被识别; 从用户接收口令命令; 对口语指挥进行语音识别; 并且如果口头命令与第二超文本匹配，则通过相关联的URL激活超链接。版权所有（C）2011，JPO＆INPIT

4. 发明专利

JP2013068970A Word identification method, word identification device, and computer-readable code 有权
标题翻译：字识别方法，字识别装置和计算机可读代码
公开(公告)号：JP2013068970A
公开(公告)日：2013-04-18
申请号：JP2012273922
申请日：2012-12-14
申请人： Nuance Communications Austria Gmbh , ニュアンスコミュニケーションズオーストリアゲーエムベーハー
发明人： SAFFEL ZSOLT
IPC分类号： G10L15/183 , G10L15/14 , G10L15/18 , G10L15/187 , G10L15/197
CPC分类号： G10L15/28 , G10L15/08 , G10L15/187
摘要： PROBLEM TO BE SOLVED: To recognize an infinite number of words in principle.SOLUTION: The present invention handles a system for speech recognition, for example, for recognizing words in a continuous speech. Disclosed is a speech recognition system capable of recognizing a great number of words, or an infinite number of words in principle. The speech recognition system includes a word recognition device for deriving the best path in a word graph, and words are allocated to the speech based upon a minimum path thereof. A phoneme language model is applied to respective words of the word graph to obtain a word score. Further, the present invention relates to a device and method which identify words from a speech block, and a computer-readable code for implementing the method.
摘要翻译：要解决的问题：原则上承认无限数量的单词。解决方案：本发明处理用于语音识别的系统，例如用于识别连续语音中的单词。公开了一种能够在原理上识别大量单词或无限数量的单词的语音识别系统。语音识别系统包括用于导出单词图形中的最佳路径的单词识别装置，并且基于其最小路径将词语分配给语音。将音素语言模型应用于单词图形的相应单词以获得单词分数。此外，本发明涉及一种识别来自语音块的单词的装置和方法，以及用于实现该方法的计算机可读代码。版权所有（C）2013，JPO＆INPIT

5. 发明专利

JP2012234204A Voice recognition device having reference conversion means and method 有权
标题翻译：具有参考转换方式和方法的语音识别装置
公开(公告)号：JP2012234204A
公开(公告)日：2012-11-29
申请号：JP2012175359
申请日：2012-08-07
申请人： Nuance Communications Austria Gmbh , ニュアンスコミュニケーションズオーストリアゲーエムベーハー
发明人： HEINRICH F BARTOSIK
IPC分类号： G10L15/06 , G10L15/00 , G10L15/065 , G10L15/20 , G10L21/0216
CPC分类号： G10L15/065 , G10L21/0216
摘要： PROBLEM TO BE SOLVED: To provide a voice recognition device having reference conversion means and a method.SOLUTION: A voice recognition device 8 recognizes a voice receivable via plural receiving channels including a first receiving channel 21 and a second receiving channel 25. The voice recognition device includes storing means 41-43 each of which stores a piece of reference information featured with word pronunciation made by plural reference speakers independent from a speaker. The storing means further stores a first conversion configured to convert the reference information corresponding to the first receiving channel into a piece of reference information corresponding to the second receiving channel; and a user application means 37 that applies the reference information using the first conversion to the second receiving channel by using the voice information obtained from the user via the first receiving channel to apply to the reference information independent from the speaker to obtain first applied reference information, according to a voice feature of a user speaking through the first receiving channel.
摘要翻译：要解决的问题：提供一种具有参考转换装置和方法的语音识别装置。解决方案：语音识别装置8通过包括第一接收信道21和第二接收信道25的多个接收信道识别可接收的语音。语音识别装置包括存储装置41-43，每个存储装置存储一条参考信息具有独立于扬声器的由多个参考音箱制成的单词发音。存储装置还存储被配置为将与第一接收信道相对应的参考信息转换为与第二接收信道相对应的参考信息的第一转换; 以及用户应用装置37，其使用经由第一接收信道从用户获得的语音信息将第一转换的参考信息应用于第二接收信道，以独立于说话者应用于参考信息，以获得第一应用的参考信息根据通过第一接收信道说出的用户的语音特征。版权所有（C）2013，JPO＆INPIT

6. 发明专利

JP2011221544A Speech detection device having two switch-off criteria 审中-公开
标题翻译：具有两个开关标准的语音检测设备
公开(公告)号：JP2011221544A
公开(公告)日：2011-11-04
申请号：JP2011122008
申请日：2011-05-31
申请人： Nuance Communications Austria Gmbh , ニュアンスコミュニケーションズオーストリアゲーエムベーハー
发明人： HEINRICH BARTOSIK
IPC分类号： G10L13/00 , G10L15/20 , G10L25/78 , G10L25/87
CPC分类号： G10L25/87 , G10L2025/783
摘要： PROBLEM TO BE SOLVED: To detect a speech signal having a sufficiently good signal-to-noise ratio and determine a speech time slot.SOLUTION: A speech detection device includes: a switch-on threshold detector 11 which sends first detection information when the energy amount of a received signal exceeds a first energy threshold; a switch-off threshold detector 12 which sends second detection information when the energy amount of the received signal is smaller than a second energy threshold being smaller than the first energy threshold; and information processing means 13 which terminates the production of speech detection information featuring a speech time slot, in response to input of the second detection information during a first switch-off period. The information processing means is configured to additionally terminate the sending of speech detection information if the first detection information is not inputted during a second switch-off period and/or if the first detection information is not inputted during a third switch-off period, and the start of the third switch-off period is determined as a time point when the second detection information is inputted for the first time after input of the first detection information ceases.
摘要翻译：要解决的问题：检测具有足够好的信噪比并且确定语音时隙的语音信号。解决方案：语音检测装置包括：接通阈值检测器11，当接收信号的能量量超过第一能量阈值时，发送第一检测信息; 关闭阈值检测器12，当接收信号的能量量小于小于第一能量阈值的第二能量阈值时，发送第二检测信息; 以及信息处理装置13，其响应于在第一关闭时段期间输入第二检测信息而终止具有语音时隙的语音检测信息的产生。信息处理装置被配置为如果在第二关闭时段期间没有输入第一检测信息和/或如果在第三关闭期间没有输入第一检测信息，则附加地终止语音检测信息的发送，以及确定第三关闭时段的开始是在第一检测信息的输入停止之后第一次输入第二检测信息的时间点。版权所有（C）2012，JPO＆INPIT

7. 发明专利

JP2011204249A Text segmentation and label assignment with user interaction by topic specific language model and topic specific label statistics 审中-公开
标题翻译：用主题特定语言模型和主题特定标签统计的用户交互的文本分类和标签分配
公开(公告)号：JP2011204249A
公开(公告)日：2011-10-13
申请号：JP2011103843
申请日：2011-05-06
申请人： Nuance Communications Austria Gmbh , ニュアンスコミュニケーションズオーストリアゲーエムベーハー
发明人： PETERS JOCHEN , MATUSOV EVGENY , MEYER CARSTEN , KLAKOW DIETRICH
IPC分类号： G06F17/21 , G06F17/27 , G06F17/30
CPC分类号： G06F17/21 , G06F17/27 , G06F17/2765
摘要： PROBLEM TO BE SOLVED: To provide a method, a computer program product, a system and a user interface for implementing the text segmentation and label assignment of an unstructured text.SOLUTION: The performed segmentation and assignment are provided to a user for a general review. Additionally, alternative segmentations and label assignments are provided to the user. The user can select alternative segmentations and alternative labels and can enter a user defined segmentation and a user defined label. In response to modification introduced by the user, a plurality of different actions are initiated by incorporating the re-segmentation and re-labelling of successive parts of the document or the entire document. Furthermore, the method comprises steps for a learning functionality, logging and analyzing user introduced modification for the adaptation of the method to user's preference and the further training of a statistical model.
摘要翻译：要解决的问题：提供一种用于实现非结构化文本的文本分段和标签分配的方法，计算机程序产品，系统和用户界面。解决方案：将执行的分割和分配提供给用户进行一般审查。另外，为用户提供了可选的分段和标签分配。用户可以选择替代的分段和替代标签，并且可以输入用户定义的分割和用户定义的标签。响应于用户引入的修改，通过结合文档或整个文档的连续部分的重新分割和重新标记来启动多个不同的动作。此外，该方法包括用于学习功能，记录和分析用户引入的修改以适应用户偏好的方法和进一步训练统计模型的步骤。

8. 发明专利

JP2012238017A Speech recognition method with substitution command 审中-公开
标题翻译：具有替代命令的语音识别方法
公开(公告)号：JP2012238017A
公开(公告)日：2012-12-06
申请号：JP2012165310
申请日：2012-07-26
申请人： Nuance Communications Austria Gmbh , ニュアンスコミュニケーションズオーストリアゲーエムベーハー
发明人： HERIBERT WUTTE
IPC分类号： G10L15/22 , G10L15/00 , G10L15/10
CPC分类号： G10L15/22
摘要： PROBLEM TO BE SOLVED: To provide a speech recognition method with a substitution command.SOLUTION: A speech recognition method of editing an element of misrecognized text information in first text information accompanied by a four-word speech input command recognized with second text information includes a step of recognizing the first text information and second text information, and a step of checking on a coincidence of a word at a first position of the second text information with an element of a first speech input command in the four-word speech input command, checking a coincidence of a word at a third position or another position of the second text information with an element of a third speech input command of the four-word speech input command, and replacing an element of the misrecognized text information with an element of new text information that the second text information includes as an element of a fourth speech input command of the four-word speech input command.
摘要翻译：要解决的问题：提供具有替代命令的语音识别方法。解决方案：用第二文本信息识别的四字语音输入命令伴随着第一文本信息编辑误识别文本信息的元素的语音识别方法包括识别第一文本信息和第二文本信息的步骤，以及在第一文本信息的第一位置处的单词与四字语音输入命令中的第一语音输入命令的元素一致的步骤，检查在第三位置或另一位置处的单词的一致性所述第二文本信息与所述四字语音输入命令的第三语音输入命令的元素相关联，并且用所述第二文本信息所包含的新文本信息的元素替换所述错误识别的文本信息的元素作为所述四字语音输入命令的元素四字语音输入命令的第四语音输入命令。版权所有（C）2013，JPO＆INPIT

9. 发明专利

JP2012009046A Topic singular language model and text segment division and labeling using user dialog based on topic singular label statistic 有权
标题翻译：基于主题单标签统计的用户对话框主题语言模型和文本段分类和标签
公开(公告)号：JP2012009046A
公开(公告)日：2012-01-12
申请号：JP2011170125
申请日：2011-08-03
申请人： Nuance Communications Austria Gmbh , ニュアンスコミュニケーションズオーストリアゲーエムベーハー
发明人： JOCHEN PETERS , EVGENI MASOV , MAYER CARSTEN , KULAKOVA DIETRICH
IPC分类号： G06F17/21 , G06F17/27 , G06F17/30
CPC分类号： G06F17/21 , G06F17/27 , G06F17/2765
摘要： PROBLEM TO BE SOLVED: To provide a method for executing segment division and labeling of a non-structured text in response to determination of a user, and to provide a text segment division system and a user interface for the text segment division system.SOLUTION: Since segment division and labeling to be executed are subject to overall revising, alternative segment division and labeling are presented to a user. The user is able to select the alternative segment division and the alternative label, and to input segment division defined by the user and a label defined by the user. In response to correction introduced by the user, a plurality of different operations are started, in which re-segment division and re-labeling of a following section of a document or the whole document are included. A learning function of correction, log recording and analysis introduced by the user are provided in order to make this method adaptive to the taste of the user and to perform further training of a statistical model.
摘要翻译：要解决的问题：提供一种响应于用户的确定而执行非结构化文本的段划分和标记的方法，并且为文本段划分系统提供文本段划分系统和用户界面。
解决方案：由于要执行的段划分和标签需要进行总体修改，因此可以向用户提供替代段划分和标签。用户能够选择替代段划分和替代标签，并且输入由用户定义的段划分和由用户定义的标签。响应于用户引入的校正，开始多个不同的操作，其中包括文档或整个文档的后续部分的重新划分和重新标记。提供了用户引入的校正，日志记录和分析的学习功能，以使得该方法适应于用户的口味并进行统计模型的进一步训练。版权所有（C）2012，JPO＆INPIT

10. 发明专利

JP2011186491A Topic specific model for text formatting and speech recognition 有权
标题翻译：用于文本格式和语音识别的主题特定模型
公开(公告)号：JP2011186491A
公开(公告)日：2011-09-22
申请号：JP2011106732
申请日：2011-05-11
申请人： Nuance Communications Austria Gmbh , ニュアンスコミュニケーションズオーストリアゲーエムベーハー
发明人： PETERS JOCHEN , MATUSOV EVGENY , MEYER CARSTEN , KLAKOW DIETRICH
IPC分类号： G10L15/22 , G06F17/21 , G06F17/27 , G10L15/183 , G10L15/32
CPC分类号： G10L15/183 , G06F17/211 , G06F17/2715 , G10L15/32
摘要： PROBLEM TO BE SOLVED: To provide an improved method of speech recognition and text formatting by clearly using information specific to a context and/or structural information of a text or speech input. SOLUTION: A text document which may be obtained from a first speech recognition pass is subject to segmentation and to an assignment of topic specific models for each obtained section. Each model of the set of models provides statistic information about text processing or formatting rules and about language model probabilities. Furthermore, other properties of a speech recognition and/or formatting system (such as e.g. settings for the speaking rate) may be encoded in the statistical models. The models are generated on the basis of annotated training data and/or by manual coding. Based on the assignment of models to sections of text, an improved speech recognition and/or text formatting procedure are performed. COPYRIGHT: (C)2011,JPO&INPIT
摘要翻译：要解决的问题：通过清楚地使用特定于文本或语音输入的上下文和/或结构信息的信息来提供改进的语音识别和文本格式化方法。解决方案：可以从第一语音识别通过获得的文本文档被分割，并且对于每个获得的部分进行主题特定模型的分配。模型集合中的每个模型提供关于文本处理或格式规则以及关于语言模型概率的统计信息。此外，可以在统计模型中编码语音识别和/或格式化系统的其他属性（例如用于说话率的设置）。这些模型是根据注释的训练数据和/或手动编码生成的。基于将文本分配给文本部分，执行改进的语音识别和/或文本格式化过程。版权所有（C）2011，JPO＆INPIT

你已经成功收藏专利！

检索式保存成功!

IPRDB

热门服务

关于我们

友情链接

联系方式