公开(公告)号：US20100121712A1

公开(公告)日：2010-05-13

申请号：US12269383

申请日：2008-11-12

申请人： Ben Shahshahani ,  Vanja Josifovski ,  Peiji Chen ,  Zengyan Zhang ,  Jianchang (JC) Mao ,  Ruofei (Bruce) Zhang ,  Sanjay Kshetramade

发明人： Ben Shahshahani ,  Vanja Josifovski ,  Peiji Chen ,  Zengyan Zhang ,  Jianchang (JC) Mao ,  Ruofei (Bruce) Zhang ,  Sanjay Kshetramade

IPC分类号： G06F17/30 ,  G06Q30/00

CPC分类号： G06Q10/0637 ,  G06Q30/02 ,  G06Q30/0256

摘要： A system for selecting advertisements for a web page. An advertisement request is received and the web page content is evaluated to identify a first advertisement query based on the content of the web page and a second advertisement query based on the content of the web page or a behavioral feature. A first set of advertisements is retrieved based on the first advertisement query and a second set of advertisements is retrieved based on the second advertisement query. The first set of advertisements is blended with the second set of advertisements to form a blended advertisement set. The blended advertisement set is provided for display with the web page.

摘要翻译： 一种用于选择网页的广告的系统。 接收广告请求，并且基于网页的内容和基于网页的内容或行为特征的第二广告查询来评估网页内容以识别第一广告查询。 基于第一广告查询检索第一组广告，并且基于第二广告查询来检索第二组广告。 第一组广告与第二组广告混合形成混合广告集。 混合广告组被提供用于与网页一起显示。

公开(公告)号：US20090248627A1

公开(公告)日：2009-10-01

申请号：US12056703

申请日：2008-03-27

申请人： Ben Shahshahani ,  Vanja Josifovski ,  Evgeniy Gabrilovich ,  Andrei Broder ,  Filip Radlinski

发明人： Ben Shahshahani ,  Vanja Josifovski ,  Evgeniy Gabrilovich ,  Andrei Broder ,  Filip Radlinski

IPC分类号： G06F17/30

CPC分类号： G06Q30/02 ,  G06F16/90335 ,  G06F16/951

摘要： A system and method are disclosed for rewriting queries. The queries may be rewritten into a bid phrase for identifying search results and/or advertisements. The bid phrase may be a keyword that is purchased for sponsored searching. A mapping between potential queries and bid phrases may be generated. The mapping may be referenced upon receiving a search query for identifying a query rewrite with a bid phrase for that search query. The mapping may be generated in preprocessing.

摘要翻译： 公开了用于重写查询的系统和方法。 查询可以被重写成用于识别搜索结果和/或广告的出价短语。 出价短语可以是为赞助搜索而购买的关键字。 可能会生成潜在查询和出价短语之间的映射。 可以在接收到用于识别具有该搜索查询的出价短语的查询重写的搜索查询时引用该映射。 可以在预处理中生成映射。

公开(公告)号：US07555430B2

公开(公告)日：2009-06-30

申请号：US11398402

申请日：2006-04-04

申请人： Hy Murveit ,  Ashvin Kannan ,  Ben Shahshahani ,  Chris Leggetter ,  Katherine Knill

发明人： Hy Murveit ,  Ashvin Kannan ,  Ben Shahshahani ,  Chris Leggetter ,  Katherine Knill

IPC分类号： G10L11/06

CPC分类号： G10L15/08 ,  G10L2015/085

摘要： Method and apparatus for multi-pass speech recognition. An input device receives spoken input. A processor performs a first pass speech recognition technique on the spoken input and forms first pass results. The first pass results include a number of alternative speech expressions, each having an assigned score related to the certainty that the corresponding expression correctly matches the spoken input. The processor selectively performs a second pass speech recognition technique on the spoken input according to the first pass results. Preferably, the second pass attempts to correctly match the spoken input to only those expressions which were identified during the first pass. Otherwise, if one of the expressions identified by the first pass is assigned a score higher than a predetermined threshold (e.g., 95%), the second pass is not performed. Because the second pass is performed only when necessary, the invention recognizes speech with a faster average speed for a given accuracy in comparison to prior systems.

摘要翻译： 用于多遍语音识别的方法和装置。 输入设备接收口头输入。 处理器在口头输入端执行第一遍语音识别技术，并形成第一遍的结果。 第一遍结果包括多个替代语言表达，每个具有与相应表达正确匹配口语输入的确定性相关的分配分数。 处理器根据第一遍的结果选择性地在口头输入上执行第二遍语音识别技术。 优选地，第二遍尝试将口头输入正确地匹配到在第一遍期间被识别的那些表达。 否则，如果由第一遍识别的表达式中的一个被分配高于预定阈值（例如，95％）的分数，则不执行第二遍。 因为仅在必要时执行第二遍，所以与现有系统相比，本发明以给定精度更快的平均速度识别语音。

公开(公告)号：US08745081B2

公开(公告)日：2014-06-03

申请号：US13419091

申请日：2012-03-13

申请人： Srinivas Vadrevu ,  Su-Lin Wu ,  Ben Shahshahani

发明人： Srinivas Vadrevu ,  Su-Lin Wu ,  Ben Shahshahani

IPC分类号： G06F7/00 ,  G06F17/30

CPC分类号： G06F17/3053 ,  G06F17/30477 ,  G06F17/30705 ,  G06F17/30864 ,  G06F17/30867 ,  G06F17/3087

摘要： News search and browse experience is personalized based on user preferences. User attributes like a geographic location are obtained and news sources preferred by other users with attributes similar to those of a requesting user are identified. News sources that are popular across different user groups are eliminated and relevant news items from the remaining news sources are retrieved and presented to the requesting user.

摘要翻译： 新闻搜索和浏览体验基于用户偏好进行个性化。 获得诸如地理位置的用户属性，并且识别出具有与请求用户的属性相似的属性的其他用户优选的新闻源。 在不同用户组中流行的新闻来源被消除，并且来自剩余新闻来源的相关新闻项被检索并呈现给请求用户。

公开(公告)号：US07401017B2

公开(公告)日：2008-07-15

申请号：US11398327

申请日：2006-04-04

申请人： Hy Murveit ,  Ashvin Kannan ,  Ben Shahshahani ,  Chris Leggetter ,  Katherine Knill

发明人： Hy Murveit ,  Ashvin Kannan ,  Ben Shahshahani ,  Chris Leggetter ,  Katherine Knill

IPC分类号： G10L21/00

CPC分类号： G10L15/08 ,  G10L2015/085

摘要： Method and apparatus for multi-pass speech recognition. An input device receives spoken input. A processor performs a first pass speech recognition technique on the spoken input and forms first pass results. The first pass results include a number of alternative speech expressions, each having an assigned score related to the certainty that the corresponding expression correctly matches the spoken input. The processor selectively performs a second pass speech recognition technique on the spoken input according to the first pass results. Preferably, the second pass attempts to correctly match the spoken input to only those expressions which were identified during the first pass. Otherwise, if one of the expressions identified by the first pass is assigned a score higher than a predetermined threshold (e.g., 95%), the second pass is not performed. Because the second pass is performed only when necessary, the invention recognizes speech with a faster average speed for a given accuracy in comparison to prior systems. Alternately, the first pass results identify a characteristic of the spoken input. The characteristic can be the gender of the speaker or a type of telephone the speaker is calling from. In which case, the second pass speech recognition technique is selected from a plurality of speech recognition techniques according to the characteristic identified by the first pass. Because the selected second pass technique is specific to the characteristic of the spoken input, the second pass technique can perform speech recognition faster for a given accuracy than a technique which is not specific.

摘要翻译： 用于多遍语音识别的方法和装置。 输入设备接收口头输入。 处理器在口头输入端执行第一遍语音识别技术，并形成第一遍的结果。 第一遍结果包括多个替代语言表达，每个具有与相应表达正确匹配口语输入的确定性相关的分配分数。 处理器根据第一遍的结果选择性地在口头输入上执行第二遍语音识别技术。 优选地，第二遍尝试将口头输入正确地匹配到在第一遍期间被识别的那些表达。 否则，如果由第一遍识别的表达式中的一个被分配高于预定阈值（例如，95％）的分数，则不执行第二遍。 因为仅在必要时执行第二遍，所以与现有系统相比，本发明以给定精度更快的平均速度识别语音。 或者，第一遍结果确定了口头输入的特征。 特征可以是扬声器的性别或扬声器所呼叫的电话类型。 在这种情况下，根据由第一遍识别的特征，从多个语音识别技术中选择第二遍语音识别技术。 由于所选择的第二传递技术特定于口头输入的特征，所以与非特定的技术相比，第二遍技术可以比给定的精度更快地执行语音识别。

公开(公告)号：US07058573B1

公开(公告)日：2006-06-06

申请号：US09294847

申请日：1999-04-20

申请人： Hy Murveit ,  Ashvin Kannan ,  Ben Shahshahani ,  Chris Leggetter ,  Katherine Knill

发明人： Hy Murveit ,  Ashvin Kannan ,  Ben Shahshahani ,  Chris Leggetter ,  Katherine Knill

IPC分类号： G10L19/02

CPC分类号： G10L15/08 ,  G10L2015/085

摘要： Method and apparatus for multi-pass speech recognition. An input device receives spoken input. A processor performs a first pass speech recognition technique on the spoken input and forms first pass results. The first pass results include a number of alternative speech expressions, each having an assigned score related to the certainty that the corresponding expression correctly matches the spoken input. The processor selectively performs a second pass speech recognition technique on the spoken input according to the first pass results. Preferably, the second pass attempts to correctly match the spoken input to only those expressions which were identified during the first pass. Otherwise, if one of the expressions identified by the first pass is assigned a score higher than a predetermined threshold (e.g., 95%), the second pass is not performed. Because the second pass is performed only when necessary, the invention recognizes speech with a faster average speed for a given accuracy in comparison to prior systems. Alternately, the first pass results identify a characteristic of the spoken input. The characteristic can be the gender of the speaker or a type of telephone the speaker is calling from. In which case, the second pass speech recognition technique is selected from a plurality of speech recognition techniques according to the characteristic identified by the first pass. Because the selected second pass technique is specific to the characteristic of the spoken input, the second pass technique can perform speech recognition faster for a given accuracy than a technique which is not specific.

摘要翻译： 用于多遍语音识别的方法和装置。 输入设备接收口头输入。 处理器在口头输入端执行第一遍语音识别技术，并形成第一遍的结果。 第一遍结果包括多个替代语言表达，每个具有与相应表达正确匹配口语输入的确定性相关的分配分数。 处理器根据第一遍的结果选择性地在口头输入上执行第二遍语音识别技术。 优选地，第二遍尝试将口头输入正确地匹配到在第一遍期间被识别的那些表达。 否则，如果由第一遍识别的表达式中的一个被分配高于预定阈值（例如，95％）的分数，则不执行第二遍。 因为仅在必要时执行第二遍，所以与现有系统相比，本发明以给定精度更快的平均速度识别语音。 或者，第一遍结果识别口语输入的特征。 特征可以是扬声器的性别或扬声器所呼叫的电话类型。 在这种情况下，根据由第一遍识别的特征，从多个语音识别技术中选择第二遍语音识别技术。 由于所选择的第二传递技术特定于口头输入的特征，所以与非特定的技术相比，第二遍技术可以比给定的精度更快地执行语音识别。

公开(公告)号：US08880099B2

公开(公告)日：2014-11-04

申请号：US13271730

申请日：2011-10-12

申请人： Ben Shahshahani

发明人： Ben Shahshahani

IPC分类号： H04W24/00 ,  H04W60/00 ,  H04W4/20 ,  H04W4/02

CPC分类号： H04W60/00 ,  H04W4/02 ,  H04W4/20

摘要： Subject matter described pertains to methods, devices, and servers for transmitting and responding to a location declaration.

摘要翻译： 所描述的主题涉及用于发送和响应位置声明的方法，设备和服务器。

公开(公告)号：US20130246452A1

公开(公告)日：2013-09-19

申请号：US13419091

申请日：2012-03-13

申请人： Srinivas Vadrevu ,  Su-Lin Wu ,  Ben Shahshahani

发明人： Srinivas Vadrevu ,  Su-Lin Wu ,  Ben Shahshahani

IPC分类号： G06F17/30

CPC分类号： G06F17/3053 ,  G06F17/30477 ,  G06F17/30705 ,  G06F17/30864 ,  G06F17/30867 ,  G06F17/3087

摘要： News search and browse experience is personalized based on user preferences. User attributes like a geographic location are obtained and news sources preferred by other users with attributes similar to those of a requesting user are identified. News sources that are popular across different user groups are eliminated and relevant news items from the remaining news sources are retrieved and presented to the requesting user.

摘要翻译： 新闻搜索和浏览体验基于用户偏好进行个性化。 获得诸如地理位置的用户属性，并且识别出具有与请求用户的属性相似的属性的其他用户优选的新闻源。 在不同用户组中流行的新闻来源被消除，并且来自剩余新闻来源的相关新闻项被检索并呈现给请求用户。

公开(公告)号：US20100030647A1

公开(公告)日：2010-02-04

申请号：US12183221

申请日：2008-07-31

申请人： Ben Shahshahani

发明人： Ben Shahshahani

IPC分类号： G06Q30/00

CPC分类号： G06Q30/02 ,  G06Q30/0269

摘要： Advertisements are selected for display on search results pages or web pages displaying requested content. A request associated with a user is received for an advertisement, such as for display on a search results page or a content-displaying web page. Profile information associated with the user is identified. An inverted index is queried based on the identified profile information. The inverted index includes a list of user attributes and a list of at least one advertisement and corresponding weight factor that correspond to each listed user attribute. A plurality of advertisements is determined in the inverted index based on the query. The determined advertisements are filtered to generate an advertisement presentation configuration that includes at least one of the identified advertisements.

摘要翻译： 选择广告以在显示所请求的内容的搜索结果页面或网页上显示。 接收与用户相关联的请求用于广告，例如用于在搜索结果页面或内容显示网页上显示。 识别与用户相关联的简档信息。 基于识别的简档信息查询反向索引。 反向索引包括用户属性的列表和与每个列出的用户属性对应的至少一个广告和对应的权重因子的列表。 基于该查询，在反向索引中确定多个广告。 所确定的广告被过滤以生成包括所识别的广告中的至少一个的广告呈现配置。

公开(公告)号：US07206389B1

公开(公告)日：2007-04-17

申请号：US10753590

申请日：2004-01-07

申请人： Benoit Dumoulin ,  Dominic Lavoie ,  Real Tremblay ,  Ben Shahshahani ,  Remi Ken-Sho Kwan

发明人： Benoit Dumoulin ,  Dominic Lavoie ,  Real Tremblay ,  Ben Shahshahani ,  Remi Ken-Sho Kwan

IPC分类号： H04M1/64

CPC分类号： G10L15/1822 ,  H04M1/271 ,  H04M3/42 ,  H04M2201/40

摘要： A computerized method is provided for electronically directing a call to a class, such that an utterance spoken by a speaker and received by a call-routing system is classified by the call-routing system as being associated with the class, such that the call-routing system includes a speech-recognition module, a feature-extraction module, and a classification module. The method includes extracting features from recognized speech; weighting elements of a feature vector with respective speech-recognition scores, wherein each weighting element is associated with one of the features; ranking classes to which the features are associated; and electronically directing the call to a highest-ranking class.

摘要翻译： 提供了一种用于将呼叫电子地引导到类的计算机化方法，使得由呼叫路由系统接收的话音讲话由呼叫路由系统分类为与该类相关联，使得呼叫 - 路由系统包括语音识别模块，特征提取模块和分类模块。 该方法包括从识别的语音中提取特征; 具有相应语音识别分数的特征向量的加权元素，其中每个加权元素与特征之一相关联; 功能相关联的排名级别; 并将电话以电子方式引导到最高排名的班级。