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. 发明申请

US20080312921A1 SPEECH RECOGNITION UTILIZING MULTITUDE OF SPEECH FEATURES 审中-公开
标题翻译：语音识别利用多种语音特征
公开(公告)号：US20080312921A1
公开(公告)日：2008-12-18
申请号：US12195123
申请日：2008-08-20
申请人： Scott E. Axelrod , Sreeram Viswanath Balakrishnan , Stanley F. Chen , Yuging Gao , Rameah A. Gopinath , Hong-Kwang Kuo , Benoit Maison , David Nahamoo , Michael Alan Picheny , George A. Saon , Geoffrey G. Zweig
发明人： Scott E. Axelrod , Sreeram Viswanath Balakrishnan , Stanley F. Chen , Yuging Gao , Rameah A. Gopinath , Hong-Kwang Kuo , Benoit Maison , David Nahamoo , Michael Alan Picheny , George A. Saon , Geoffrey G. Zweig
IPC分类号： G10L15/00 , G10L15/04
CPC分类号： G10L15/063 , G10L15/02 , G10L15/14 , G10L2015/085
摘要： In a speech recognition system, the combination of a log-linear model with a multitude of speech features is provided to recognize unknown speech utterances. The speech recognition system models the posterior probability of linguistic units relevant to speech recognition using a log-linear model. The posterior model captures the probability of the linguistic unit given the observed speech features and the parameters of the posterior model. The posterior model may be determined using the probability of the word sequence hypotheses given a multitude of speech features. Log-linear models are used with features derived from sparse or incomplete data. The speech features that are utilized may include asynchronous, overlapping, and statistically non-independent speech features. Not all features used in training need to appear in testing/recognition.
摘要翻译：在语音识别系统中，提供了具有多个语音特征的对数线性模型的组合来识别未知语音语音。语音识别系统使用对数线性模型对与语音识别相关的语言单位的后验概率进行建模。后验模型捕获了语言单位给出观察到的语音特征和后验模型参数的概率。可以使用给定多个语音特征的单词序列假设的概率来确定后验模型。对数线性模型与来自稀疏或不完整数据的特征一起使用。所使用的语音特征可以包括异步，重叠和统计上非独立的语音特征。培训中使用的并非所有功能都需要出现在测试/识别中。

2. 发明授权

US07464031B2 Speech recognition utilizing multitude of speech features 失效
标题翻译：语音识别利用多种语音特征
公开(公告)号：US07464031B2
公开(公告)日：2008-12-09
申请号：US10724536
申请日：2003-11-28
申请人： Scott E. Axelrod , Sreeram Viswanath Balakrishnan , Stanley F. Chen , Yuging Gao , Ramesh A. Gopinath , Hong-Kwang Kuo , Benoit Maison , David Nahamoo , Michael Alan Picheny , George A. Saon , Geoffrey G. Zweig
发明人： Scott E. Axelrod , Sreeram Viswanath Balakrishnan , Stanley F. Chen , Yuging Gao , Ramesh A. Gopinath , Hong-Kwang Kuo , Benoit Maison , David Nahamoo , Michael Alan Picheny , George A. Saon , Geoffrey G. Zweig
IPC分类号： G10L15/00 , G10L15/20
CPC分类号： G10L15/063 , G10L15/02 , G10L15/14 , G10L2015/085
摘要： In a speech recognition system, the combination of a log-linear model with a multitude of speech features is provided to recognize unknown speech utterances. The speech recognition system models the posterior probability of linguistic units relevant to speech recognition using a log-linear model. The posterior model captures the probability of the linguistic unit given the observed speech features and the parameters of the posterior model. The posterior model may be determined using the probability of the word sequence hypotheses given a multitude of speech features. Log-linear models are used with features derived from sparse or incomplete data. The speech features that are utilized may include asynchronous, overlapping, and statistically non-independent speech features. Not all features used in training need to appear in testing/recognition.
摘要翻译：在语音识别系统中，提供了具有多个语音特征的对数线性模型的组合来识别未知语音语音。语音识别系统使用对数线性模型对与语音识别相关的语言单位的后验概率进行建模。后验模型捕获了语言单位给出观察到的语音特征和后验模型参数的概率。可以使用给定多个语音特征的单词序列假设的概率来确定后验模型。对数线性模型与来自稀疏或不完整数据的特征一起使用。所使用的语音特征可以包括异步，重叠和统计上非独立的语音特征。培训中使用的并非所有功能都需要出现在测试/识别中。

3. 发明申请

US20050119885A1 Speech recognition utilizing multitude of speech features 失效
标题翻译：语音识别利用多种语音特征
公开(公告)号：US20050119885A1
公开(公告)日：2005-06-02
申请号：US10724536
申请日：2003-11-28
申请人： Scott Axelrod , Sreeram Balakrishnan , Stanley Chen , Yuging Gao , Ramesh Gopinath , Hong-Kwang Kuo , Benoit Maison , David Nahamoo , Michael Picheny , George Saon , Geoffrey Zweig
发明人： Scott Axelrod , Sreeram Balakrishnan , Stanley Chen , Yuging Gao , Ramesh Gopinath , Hong-Kwang Kuo , Benoit Maison , David Nahamoo , Michael Picheny , George Saon , Geoffrey Zweig
IPC分类号： G10L15/10 , G10L15/00 , G10L15/02 , G10L15/06 , G10L15/14
CPC分类号： G10L15/063 , G10L15/02 , G10L15/14 , G10L2015/085
摘要： In a speech recognition system, the combination of a log-linear model with a multitude of speech features is provided to recognize unknown speech utterances. The speech recognition system models the posterior probability of linguistic units relevant to speech recognition using a log-linear model. The posterior model captures the probability of the linguistic unit given the observed speech features and the parameters of the posterior model. The posterior model may be determined using the probability of the word sequence hypotheses given a multitude of speech features. Log-linear models are used with features derived from sparse or incomplete data. The speech features that are utilized may include asynchronous, overlapping, and statistically non-independent speech features. Not all features used in training need to appear in testing/recognition.
摘要翻译：在语音识别系统中，提供了具有多个语音特征的对数线性模型的组合来识别未知语音语音。语音识别系统使用对数线性模型对与语音识别相关的语言单位的后验概率进行建模。后验模型捕获了语言单位给出观察到的语音特征和后验模型参数的概率。可以使用给定多个语音特征的单词序列假设的概率来确定后验模型。对数线性模型与来自稀疏或不完整数据的特征一起使用。所使用的语音特征可以包括异步，重叠和统计上非独立的语音特征。培训中使用的并非所有功能都需要出现在测试/识别中。

4. 发明授权

US07251599B2 Automatic construction of unique signatures and confusable sets for database access 有权
标题翻译：自动构建数据库访问的独特签名和混淆集
公开(公告)号：US07251599B2
公开(公告)日：2007-07-31
申请号：US10315411
申请日：2002-12-10
申请人： Benoit Maison , Geoffrey G. Zweig
发明人： Benoit Maison , Geoffrey G. Zweig
IPC分类号： G10L15/02
CPC分类号： G10L15/18
摘要： Methods and arrangements for facilitating database access in speech recognition. A plurality of possible subsequences corresponding to a database entry are ascertained, a record of such subsequences and their correspondence to database entries is created, and either or both of the following are carried out: unique signatures are ascertained via determining whether a subsequence corresponding to a given database entry does not also correspond to at least one other database entry; and/or multiple occurrences of a given subsequence are found, with corresponding database entries being grouped into a confusion set.
摘要翻译：在语音识别中促进数据库访问的方法和安排。确定对应于数据库条目的多个可能的子序列，创建这样的子序列的记录及其与数据库条目的对应关系，并执行以下任何一个或两者：唯一签名是通过确定对应于给定的数据库条目也不对应于至少一个其他数据库条目; 和/或发现给定子序列的多次出现，其中相应的数据库条目被分组成混淆集合。

5. 发明申请

US20080243507A1 NATURAL ERROR HANDLING IN SPEECH RECOGNITION 有权
标题翻译：语音识别中的自然错误处理
公开(公告)号：US20080243507A1
公开(公告)日：2008-10-02
申请号：US12135397
申请日：2008-06-09
申请人： Ramesh A. Gopinath , Benoit Maison , Brian C. Wu
发明人： Ramesh A. Gopinath , Benoit Maison , Brian C. Wu
IPC分类号： G10L15/18
CPC分类号： G10L15/22 , G10L2015/225
摘要： A user interface, and associated techniques, that permit a fast and efficient way of correcting speech recognition errors, or of diminishing their impact. The user may correct mistakes in a natural way, essentially by repeating the information that was incorrectly recognized previously. Such a mechanism closely approximates what human-to-human dialogue would be in similar circumstances. Such a system fully takes advantage of all the information provided by the user, and on its own estimates the quality of the recognition in order to determine the correct sequence of words in the fewest number of steps.
摘要翻译：用户界面和相关技术，可以快速有效地纠正语音识别错误或减少其影响。用户可以以自然的方式纠正错误，主要是通过重复先前未正确识别的信息。这种机制与近似于类似情况下的人与人之间的对话密切相关。这样的系统充分利用用户提供的所有信息，并且自己估计识别的质量，以便以最少的步数确定正确的单词序列。

6. 发明授权

US07386454B2 Natural error handling in speech recognition 有权
标题翻译：语音识别中的自然错误处理
公开(公告)号：US07386454B2
公开(公告)日：2008-06-10
申请号：US10210704
申请日：2002-07-31
申请人： Ramesh A. Gopinath , Benoit Maison , Brian C. Wu
发明人： Ramesh A. Gopinath , Benoit Maison , Brian C. Wu
IPC分类号： G10L11/00 , G10L21/00
CPC分类号： G10L15/22 , G10L2015/225
摘要： A user interface, and associated techniques, that permit a fast and efficient way of correcting speech recognition errors, or of diminishing their impact. The user may correct mistakes in a natural way, essentially by repeating the information that was incorrectly recognized previously. Such a mechanism closely approximates what human-to-human dialogue would be in similar circumstances. Such a system fully takes advantage of all the information provided by the user, and on its own estimates the quality of the recognition in order to determine the correct sequence of words in the fewest number of steps.
摘要翻译：用户界面和相关技术，可以快速有效地纠正语音识别错误或减少其影响。用户可以以自然的方式纠正错误，主要是通过重复先前未正确识别的信息。这种机制与近似于类似情况下的人与人之间的对话密切相关。这样的系统充分利用用户提供的所有信息，并且自己估计识别的质量，以便以最少的步数确定正确的单词序列。

7. 发明授权

US07206387B2 Resource allocation for voice processing applications 有权
标题翻译：语音处理应用的资源分配
公开(公告)号：US07206387B2
公开(公告)日：2007-04-17
申请号：US10645051
申请日：2003-08-21
申请人： Ea-Ee Jan , Benoit Maison , Andrzei Sakrajda
发明人： Ea-Ee Jan , Benoit Maison , Andrzei Sakrajda
IPC分类号： H04M1/64
CPC分类号： H04M3/50 , H04L67/1002 , H04L2012/6443 , H04L2012/6481 , H04M2201/39 , H04M2201/40 , H04M2201/41
摘要： A voice processing system is provided in which sets of engines running on a plurality of servers are configured differently from one another. The sets of engines may be configured to achieve different trade-offs between performance of a task and resources required to perform the task. In the voice processing system, a task routing server is provided that assigns different sets of sub-tasks to different sets of task engines. The number of engines used to perform a task and the number of engines in each set are adjusted. By adjusting the parameters settings for the set of engines based on the type of application, the particular requirements of the application, or the nature and importance of the subtasks, for example, advantages such as improvement of resource utilization and the hardware and software costs reduction may be obtained.
摘要翻译：提供语音处理系统，其中在多个服务器上运行的引擎组彼此不同地配置。发动机组可以被配置为在执行任务和执行任务所需的资源之间实现不同的权衡。在语音处理系统中，提供了将不同的子任务集分配给不同的任务引擎组的任务路由服务器。调整用于执行任务的引擎数量和每组中的引擎数量。通过基于应用程序的类型，应用程序的特定要求或子任务的性质和重要性调整引擎集的参数设置，例如资源利用率的提高以及降低硬件和软件成本的优点可以获得。

8. 发明申请

US20090315829A1 Multi-User Pointing Apparaus and Method 有权
标题翻译：多用户指点Apparaus和方法
公开(公告)号：US20090315829A1
公开(公告)日：2009-12-24
申请号：US12309784
申请日：2006-08-02
申请人： Benoit Maison , Xavier Sohy , Olivier Pierre , Matthieu Laheux
发明人： Benoit Maison , Xavier Sohy , Olivier Pierre , Matthieu Laheux
IPC分类号： G06F3/033
CPC分类号： G06F3/03545 , G06F3/0386
摘要： An apparatus for interaction of a plurality of users with an application, comprising: a plurality of pointing devices (10) capable of emitting a signal (80; 110) (such as visible or infrared light, a laser or omnidirectional flashes), each pointing device comprising one or more triggers; a presentation device (20) (such as display, a wall, or any objects); and detection means (30) (camera, stereo camera, other) capable of detecting said signal (80; 110). The apparatus also comprises means (30, 40) for detecting trigger actions by said users, and recording the identity and time of said trigger actions (30, 40), means (40, 13) for enabling said pointing devices (10) to emit said signal (80; 110) and means (40) for enabling said detection means (30) to detect said signal (80, 110), and for linking said signal detection to a specific pointing device (10) trigger action.
摘要翻译：一种用于多个用户与应用的交互的装置，包括：能够发射信号（诸如可见光或红外光，激光或全向闪光）的多个指点设备（10），每个指向设备包括一个或多个触发器; 呈现装置（20）（诸如显示器，墙壁或任何物体）; 以及能够检测所述信号（80; 110）的检测装置（30）（照相机，立体相机等）。该装置还包括用于检测所述用户的触发动作并记录所述触发动作（30,40）的身份和时间的装置（30,40），用于使所述指示装置（10）发射的装置（40,13）所述信号（80; 110）和用于使所述检测装置（30）能够检测所述信号（80,110）以及用于将所述信号检测与特定指示装置（10）触发动作相关联的装置（40）。

9. 发明授权

US09465774B2 Optical character recognition system using multiple images and method of use 有权
标题翻译：光学字符识别系统采用多种图像和使用方法
公开(公告)号：US09465774B2
公开(公告)日：2016-10-11
申请号：US14659935
申请日：2015-03-17
申请人： Benoit Maison
发明人： Benoit Maison
IPC分类号： G06K9/18 , G06F17/21 , G06K9/00 , G06K9/32 , G06K9/62 , G06K9/68
CPC分类号： G06F17/21 , G06K9/00671 , G06K9/3258 , G06K9/6292 , G06K9/6807 , G06K2209/01
摘要： Disclosed is an improved OCR system wherein the same can be utilized for capturing and analyzing multiple images of a document to increase the efficiency and accuracy of digitizing printed texts on the document. Captured images are merged into a single set of character recognition results via a recognition method from multiple images, which include early fusion, late fusion, and hybrid fusion embodiments. The end product from each of the embodiments provides text and metadata that include recognized words. In late and hybrid fusion, words having confidence scores above a predetermined threshold are assembled together to form paragraphs to reconstruct a digital version of the document. In this way, the present invention utilizes multiple images that can be combined to aggregate information and achieve high accuracy when scanning and digitizing printed texts.
摘要翻译：公开了一种改进的OCR系统，其中相同的OCR系统可用于捕获和分析文档的多个图像，以提高在文档上数字化打印文本的效率和准确性。捕获的图像通过来自多个图像的识别方法合并成单个字符识别结果集合，其包括早期融合，后期融合和混合融合实施例。来自每个实施例的最终产品提供包括识别的单词的文本和元数据。在晚期和混合融合中，具有高于预定阈值的置信度得分的词组合在一起以形成段落以重建文档的数字版本。以这种方式，本发明利用可以组合的多个图像来聚合信息，并且当扫描和数字化印刷文本时实现高精度。

10. 发明申请

US20070179787A1 Update technique for speech recognition applications with uninterrupted (24X7) operation 失效
标题翻译：具有不间断（24X7）操作的语音识别应用的更新技术
公开(公告)号：US20070179787A1
公开(公告)日：2007-08-02
申请号：US11343622
申请日：2006-01-31
申请人： Ea-Ee Jan , Benoit Maison
发明人： Ea-Ee Jan , Benoit Maison
IPC分类号： G10L21/00
CPC分类号： G06F8/656
摘要： A method of updating an application in a processing system, the aforesaid method comprising the steps of: (a) starting operation of a first version of an application in the aforesaid processing system; (b) starting operation of a second version of the aforesaid application in the aforesaid processing system; and (c) incrementally reallocating processing system resources from the aforesaid first version of the aforesaid application to a second version of the aforesaid application, such that at most one application selected from the aforesaid first version and the aforesaid second version of the aforesaid application operates at full capacity at any time, and such that at least one of the aforesaid first version and the aforesaid second version of the aforesaid application is operating at all times.
摘要翻译：一种在处理系统中更新应用的方法，所述方法包括以下步骤：（a）在上述处理系统中启动应用的第一版本的操作; （b）在上述处理系统中开始执行上述申请的第二版本; 和（c）将上述申请的上述第一版本的处理系统资源逐步重新分配到上述申请的第二版本，使得从上述第一版本和上述第二版本中选出的至多一个应用程序在并且使得上述第一版本和上述第二版本中的至少一个正在任何时候运行。

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