专利快速检索-快速检索全球专利，免费商用专利数据库-IPRDB

31. 再颁专利

USRE45289E1 Selective noise/channel/coding models and recognizers for automatic speech recognition 有权
标题翻译：选择性噪声/信道/编码模型和自动语音识别识别器
公开(公告)号：USRE45289E1
公开(公告)日：2014-12-09
申请号：US09978250
申请日：2001-10-17
申请人： Randy G. Goldberg , Kenneth H. Rosen , Richard M. Sachs , Joel A. Winthrop
发明人： Randy G. Goldberg , Kenneth H. Rosen , Richard M. Sachs , Joel A. Winthrop
IPC分类号： G10L15/20 , G10L15/26
CPC分类号： G10L15/20
摘要： An apparatus and method for the robust recognition of speech during a call in a noisy environment is presented. Specific background noise models are created to model various background noises which may interfere in the error free recognition of speech. These background noise models are then used to determine which noise characteristics a particular call has. Once a determination has been made of the background noise in any given call, speech recognition is carried out using the appropriate background noise model.
摘要翻译：提出了一种用于在嘈杂环境中的呼叫期间鲁棒识别语音的装置和方法。创建特定的背景噪声模型来模拟可能干扰语音的无错误识别的各种背景噪声。然后使用这些背景噪声模型来确定特定呼叫的哪个噪声特性。一旦确定了任何给定呼叫中的背景噪声，则使用适当的背景噪声模型进行语音识别。

32. 发明授权

US08682665B2 Concise dynamic grammars using N-best selection 有权
标题翻译：使用N最佳选择的简明动态语法
公开(公告)号：US08682665B2
公开(公告)日：2014-03-25
申请号：US13096431
申请日：2011-04-28
申请人： Deborah W. Brown , Randy G. Goldberg , Stephen Michael Marcus , Richard R. Rosinski
发明人： Deborah W. Brown , Randy G. Goldberg , Stephen Michael Marcus , Richard R. Rosinski
IPC分类号： G10L15/06
CPC分类号： G10L15/197 , G10L15/02 , G10L15/06 , G10L15/063 , G10L15/19 , G10L15/22 , G10L25/27 , G10L2015/228
摘要： A method and apparatus derive a dynamic grammar composed of a subset of a plurality of data elements that are each associated with one of a plurality of reference identifiers. The present invention generates a set of selection identifiers on the basis of a user-provided first input identifier and determines which of these selection identifiers are present in a set of pre-stored reference identifiers. The present invention creates a dynamic grammar that includes those data elements that are associated with those reference identifiers that are matched to any of the selection identifiers. Based on a user-provided second identifier and on the data elements of the dynamic grammar, the present invention selects one of the reference identifiers in the dynamic grammar.
摘要翻译：方法和装置导出由多个数据元素的子集组成的动态语法，每个数据元素与多个参考标识符之一相关联。本发明基于用户提供的第一输入标识符生成一组选择标识符，并且确定这些选择标识符中的哪一个存在于一组预先存储的参考标识符中。本发明创建动态语法，其包括与与任何选择标识符匹配的那些参考标识符相关联的那些数据元素。基于用户提供的第二标识符和动态语法的数据元素，本发明在动态语法中选择参考标识符之一。

33. 发明授权

US08117033B2 System and method for automatic verification of the understandability of speech 有权
标题翻译：自动验证语音的可理解性的系统和方法
公开(公告)号：US08117033B2
公开(公告)日：2012-02-14
申请号：US13204932
申请日：2011-08-08
申请人： Harvey S. Cohen , Randy G. Goldberg , Kenneth H. Rosen
发明人： Harvey S. Cohen , Randy G. Goldberg , Kenneth H. Rosen
IPC分类号： G10L17/00
CPC分类号： H04M3/2236 , G10L25/69 , H04M3/4936
摘要： Disclosed herein are systems, methods, and computer-readable storage media for processing a message received from a user to determine whether an estimate of intelligibility is below an intelligibility threshold. The method includes recognizing a portion of a user's message that contains the one or more expected utterances from a critical information list, calculating an estimate of intelligibility for the recognized portion of the user's message that contains the one or more expected utterances, and prompting the user to repeat at least the recognized portion of the user's message if the calculated estimate of intelligibility for the recognized portion of the user's message is below an intelligibility threshold. In one aspect, the method further includes prompting the user to repeat at least a portion of the message if any of a measured speech level and a measured signal-to-noise ratio of the user's message are determined to be below their respective thresholds.
摘要翻译：本文公开了用于处理从用户接收的消息的系统，方法和计算机可读存储介质，以确定可懂度的估计是否低于可懂度阈值。该方法包括从关键信息列表中识别包含一个或多个预期话语的用户消息的一部分，计算包含一个或多个预期话语的用户消息的识别部分的可懂度的估计，并提示用户如果所计算的用户消息的识别部分的可懂度估计低于可懂度阈值，则至少重复用户消息的识别部分。在一个方面，所述方法还包括提示用户重复所述消息的至少一部分，如果确定所述用户消息的测量语音水平和所测量的信噪比中的任一个低于其各自的阈值。

34. 发明授权

US07996221B2 System and method for automatic verification of the understandability of speech 有权
标题翻译：自动验证语音的可理解性的系统和方法
公开(公告)号：US07996221B2
公开(公告)日：2011-08-09
申请号：US12644802
申请日：2009-12-22
申请人： Harvey S. Cohen , Randy G. Goldberg , Kenneth H. Rosen
发明人： Harvey S. Cohen , Randy G. Goldberg , Kenneth H. Rosen
IPC分类号： G10L17/00
CPC分类号： H04M3/2236 , G10L25/69 , H04M3/4936
摘要： Disclosed herein are systems, methods, and computer-readable storage media for processing a message received from a user to determine whether an estimate of intelligibility is below an intelligibility threshold. The method includes recognizing a portion of a user's message that contains the one or more expected utterances from a critical information list, calculating an estimate of intelligibility for the recognized portion of the user's message that contains the one or more expected utterances, and prompting the user to repeat at least the recognized portion of the user's message if the calculated estimate of intelligibility for the recognized portion of the user's message is below an intelligibility threshold. In one aspect, the method further includes prompting the user to repeat at least a portion of the message if any of a measured speech level and a measured signal-to-noise ratio of the user's message are determined to be below their respective thresholds.
摘要翻译：本文公开了用于处理从用户接收的消息的系统，方法和计算机可读存储介质，以确定可懂度的估计是否低于可懂度阈值。该方法包括从关键信息列表中识别包含一个或多个预期话语的用户消息的一部分，计算包含一个或多个预期话语的用户消息的识别部分的可懂度的估计，并提示用户如果所计算的用户消息的识别部分的可懂度估计低于可懂度阈值，则至少重复用户消息的识别部分。在一个方面，所述方法还包括提示用户重复所述消息的至少一部分，如果确定所述用户消息的测量语音水平和所测量的信噪比中的任一个低于其各自的阈值。

35. 发明授权

US07295982B1 System and method for automatic verification of the understandability of speech 有权
标题翻译：自动验证语音的可理解性的系统和方法
公开(公告)号：US07295982B1
公开(公告)日：2007-11-13
申请号：US09992988
申请日：2001-11-19
申请人： Harvey S. Cohen , Randy G. Goldberg , Kenneth H. Rosen
发明人： Harvey S. Cohen , Randy G. Goldberg , Kenneth H. Rosen
IPC分类号： G10L21/00
CPC分类号： H04M3/2236 , G10L25/69 , H04M3/4936
摘要： The present invention relates to a system and method for automatically verifying that a message received from a user is intelligible. In an exemplary embodiment, a message is received from the user. A speech level of the user's message may be measured and compared to a pre-determined speech level threshold to determine whether the measured speech level is below the pre-determined speech level threshold. A signal-to-noise ratio of the user's message may be measured and compared to a pre-determined signal-to-noise ratio threshold to determine whether the measured signal-to-noise ratio of the message is below the pre-determined signal-to-noise ratio threshold. An estimate of intelligibility for the user's message may be calculated and compared to an intelligibility threshold to determine whether the calculated estimate of intelligibility is below the intelligibility threshold. If any of the measured speech level, measured signal-to-noise ratio and calculated estimate of intelligibility of the user's message are determined to be below their respective thresholds, the user may be prompted to repeat at least a portion of the message.
摘要翻译：本发明涉及一种用于自动验证从用户接收的消息是可理解的系统和方法。在示例性实施例中，从用户接收到消息。可以测量用户消息的语音电平并将其与预定语音电平阈值进行比较，以确定所测量的语音电平是否低于预定语音电平阈值。可以测量用户消息的信噪比并将其与预定信噪比阈值进行比较，以确定所测量的消息的信噪比是否低于预定信号 - 信噪比阈值。可以计算用户消息的可懂度的估计，并将其与可懂度阈值进行比较，以确定所计算的可懂度的估计是否低于可懂度阈值。如果测量的语音水平，测量的信噪比和用户消息的可理解度的计算估计中的任何一个被确定为低于其各自的阈值，则可能提示用户重复消息的至少一部分。

36. 发明授权

US06208729B1 Method and apparatus for on-hold switching 有权
标题翻译：保持开关的方法和装置
公开(公告)号：US06208729B1
公开(公告)日：2001-03-27
申请号：US09174013
申请日：1998-10-16
申请人： Sanjay Agraharam , Randy G. Goldberg , Robert Edward Markowitz , Kenneth H. Rosen , David Hilton Shur
发明人： Sanjay Agraharam , Randy G. Goldberg , Robert Edward Markowitz , Kenneth H. Rosen , David Hilton Shur
IPC分类号： H04M300
CPC分类号： H04M3/428
摘要： The invention provides an on-hold switching device that permits a subscriber to be engaged in other activities through a telephone network while being placed on-hold by another party. When placed on-hold, the on-hold switching device disconnects the subscriber from the other party and connects the subscriber to the telephone network so that the subscriber may engage in other activities. The on-hold switching device monitors the signal bus connected to the other party to determine whether the on-hold condition is removed. When the on-hold condition is removed, the subscriber is reconnected to the other party. Thus, the on-hold switching device permits a subscriber to be engaged in other activities when placed on-hold by the other party.
摘要翻译：本发明提供了一种保持开关装置，其允许用户通过电话网络从事其他活动，同时被另一方置于保持状态。当保持时，保持开关设备将用户与另一方断开连接，并将用户连接到电话网络，使得用户可以从事其他活动。保持开关装置监视与另一方相连的信号总线，以确定是否取消保持状态。当保持状态被移除时，用户被重新连接到另一方。因此，保持开关装置允许用户在被另一方置于保持状态时从事其他活动。

37. 发明授权

US06205205B1 Method and system for affinity group telephone service 失效
标题翻译：亲和组电话服务的方法和系统
公开(公告)号：US06205205B1
公开(公告)日：2001-03-20
申请号：US09020931
申请日：1998-02-09
申请人： Randy G. Goldberg , Bruce Lowell Hanson , Amir M. Mane
发明人： Randy G. Goldberg , Bruce Lowell Hanson , Amir M. Mane
IPC分类号： H04M164
CPC分类号： H04M3/42017 , H04M3/42 , H04M2203/2044 , H04M2203/2061
摘要： A plurality of subscribers in a telecommunications network are divided into affinity groups. Subscribers in a particular affinity group have a common interest which is different from that of subscribers in other affinity groups. That interest may a common cultural heritage, a common language preference, a common vocational or avocational interest, or the like. Each respective affinity group has associated with it a customized set of call processing characteristics, such as a distinctive dial tone, a distinctive subscriber language preference, and/or an abbreviated dialing list of other subscribers in the group. In addition, each respective affinity group can have a special broadcast message sent to the some or all of the subscribers in the group.
摘要翻译：电信网络中的多个用户被分为亲和组。特定关联组中的订阅者具有与其他亲和组中的订户不同的共同兴趣。这种兴趣可能是共同的文化遗产，共同的语言偏好，普通的职业或职业兴趣等。每个相关联的组已经与它相关联的一组定制的呼叫处理特性集合，诸如区别拨号音，独特的用户语言偏好和/或组中的其他订户的缩写拨号列表。此外，每个相应的亲和组可以具有发送给组中的一些或所有订户的特殊广播消息。

38. 发明授权

US6154579A Confusion matrix based method and system for correcting misrecognized words appearing in documents generated by an optical character recognition technique 失效
标题翻译：基于混淆矩阵的方法和系统，用于校正出现在由光学字符识别技术产生的文档中的误识别词
公开(公告)号：US6154579A
公开(公告)日：2000-11-28
申请号：US909199
申请日：1997-08-11
申请人： Randy G. Goldberg
发明人： Randy G. Goldberg
IPC分类号： G06K9/03 , G06K9/68 , G06K9/72
CPC分类号： G06K9/6857 , G06K9/03
摘要： A method and apparatus for correcting misrecognized words appearing in electronic documents that have been generated by scanning an original document in accordance with an optical character recognition ("OCR") technique. If an incorrect word is found in the electronic document, the present invention generates at least one reference word and selects the reference word that is the most likely correct replacement for the incorrect word. This selection is accomplished by performing a probabilistic determination that assigns to each reference word a replacement word recognition probability. The probabilistic determination is carried out on the basis of a pre-stored confusion matrix that stores a plurality of probability values. The confusion matrix is used to associate each character of recognized word in the electronic document with a corresponding character of a word in the original document on the basis of these probability values.
摘要翻译：一种用于校正通过根据光学字符识别（“OCR”）技术扫描原始文档而产生的电子文档中出现的误识别词的方法和装置。如果在电子文档中发现不正确的字，则本发明生成至少一个参考字，并且选择对该错误字最可能正确替换的参考字。该选择通过执行向每个参考字分配替换字识别概率的概率确定来实现。概率确定是基于存储多个概率值的预先存储的混淆矩阵进行的。混淆矩阵用于基于这些概率值将电子文档中的识别字的每个字符与原始文档中的单词的相应字符相关联。

39. 发明授权

US6049768A Speech recognition system with implicit checksum 失效
标题翻译：具有隐式校验和的语音识别系统
公开(公告)号：US6049768A
公开(公告)日：2000-04-11
申请号：US963381
申请日：1997-11-03
申请人： Randy G. Goldberg , John Bruce Harlow
发明人： Randy G. Goldberg , John Bruce Harlow
IPC分类号： G10L15/22 , G10L17/00
CPC分类号： G10L17/24 , G10L15/22
摘要： A user identifier is preassigned to a user, the user identifier being generated by retrieving an implicit checksum and calculating the user identifier (by, for example, using a hidden Markov model) using a checksum operation such as a modulo algorithm so that the user identifier conforms to the implicit checksum, the implicit checksum being based on a checksum parameter, such as the current date or month or a user's phone number. The preassigned user identifier is then received from the user, at and least one recognized identifier is received that is based on the received preassigned user identifier. A checksum is calculated for one recognized identifier using a checksum operation, and the implicit checksum is retrieved. The calculated checksum is compared with the implicit checksum, and the one recognized identifier is selected as the received preassigned user identifier if the calculated checksum equals the retrieved implicit checksum. The selected recognized identifier can be determined as the received user identifier by presenting the selection to the user and receiving confirmation.
摘要翻译：用户标识符被预先分配给用户，该用户标识符是通过检索隐式校验和并使用诸如模算法的校验和操作计算用户标识符（例如使用隐马尔可夫模型）来生成的，以便用户标识符符合隐式校验和，隐式校验和基于校验和参数，例如当前日期或月份或用户的电话号码。然后从用户接收预先分配的用户标识符，并且基于接收的预先分配的用户标识符接收至少一个识别的标识符。使用校验和操作为一个识别的标识符计算校验和，并检索隐式校验和。将计算的校验和与隐式校验和进行比较，如果计算的校验和等于检索到的隐式校验和，则将一个已识别的标识符选择为接收的预先分配的用户标识符。所选择的识别标识符可以通过向用户呈现选择并接收确认来确定为接收的用户标识符。

40. 发明授权

US5970446A Selective noise/channel/coding models and recognizers for automatic speech recognition 失效
标题翻译：选择性噪声/信道/编码模型和自动语音识别识别器
公开(公告)号：US5970446A
公开(公告)日：1999-10-19
申请号：US978527
申请日：1997-11-25
申请人： Randy G. Goldberg , Kenneth H. Rosen , Richard M. Sachs , Joel A. Winthrop, III
发明人： Randy G. Goldberg , Kenneth H. Rosen , Richard M. Sachs , Joel A. Winthrop, III
IPC分类号： G10L15/20 , G10L7/08
CPC分类号： G10L15/20
摘要： An apparatus and method for the robust recognition of speech during a call in a noisy environment is presented. Specific background noise models are created to model various background noises which may interfere in the error free recognition of speech. These background noise models are then used to determine which noise characteristics a particular call has. Once a determination has been made of the background noise in any given call, speech recognition is carried out using the appropriate background noise model.
摘要翻译：提出了一种用于在嘈杂环境中的呼叫期间鲁棒识别语音的装置和方法。创建特定的背景噪声模型来模拟可能干扰语音的无错误识别的各种背景噪声。然后使用这些背景噪声模型来确定特定呼叫的哪个噪声特性。一旦确定了任何给定呼叫中的背景噪声，则使用适当的背景噪声模型进行语音识别。

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