公开(公告)号：US20090119102A1

公开(公告)日：2009-05-07

申请号：US11933739

申请日：2007-11-01

Applicant: Srinivas BANGALORE ,  Vivek Kumar RANGARAJAN SRIDHAR

Inventor： Srinivas BANGALORE ,  Vivek Kumar RANGARAJAN SRIDHAR

IPC: G10L15/00

CPC classification number: G10L15/1807 ,  G10L25/90

Abstract: Disclosed are a system and method for exploiting information in an utterance for dialog act tagging. An exemplary method includes receiving a user utterance, computing at periodic intervals at least one parameter in the user utterance, quantizing the at least one parameter at each periodic interval, approximating conditional probabilities using an n-gram over a sliding window over the periodic intervals and tagging the utterance as a dialog act based on the approximated conditional probabilities.

Abstract translation: 公开了一种用于对话行为标记的话语中利用信息的系统和方法。 一种示例性方法包括接收用户话语，以周期性间隔计算用户话语中的至少一个参数，在每个周期性间隔量化所述至少一个参数，使用在周期性间隔上的滑动窗口上的n-gram近似条件概率，以及 基于近似条件概率将话语标记为对话动作。

公开(公告)号：US20130144616A1

公开(公告)日：2013-06-06

申请号：US13311852

申请日：2011-12-06

Applicant: Srinivas BANGALORE

Inventor： Srinivas BANGALORE

IPC: G10L21/02 ,  G10L15/00 ,  G10L11/00

CPC classification number: G10L15/075 ,  G10L15/1822 ,  G10L15/197 ,  G10L15/20 ,  G10L15/25 ,  G10L17/22 ,  G10L25/84 ,  G10L2015/223 ,  G10L2015/228

Abstract: Disclosed herein are systems, methods, and non-transitory computer-readable storage media for processing speech. A system configured to practice the method monitors user utterances to generate a conversation context. Then the system receives a current user utterance independent of non-natural language input intended to trigger speech processing. The system compares the current user utterance to the conversation context to generate a context similarity score, and if the context similarity score is above a threshold, incorporates the current user utterance into the conversation context. If the context similarity score is below the threshold, the system discards the current user utterance. The system can compare the current user utterance to the conversation context based on an n-gram distribution, a perplexity score, and a perplexity threshold. Alternately, the system can use a task model to compare the current user utterance to the conversation context.

Abstract translation: 本文公开了用于处理语音的系统，方法和非暂时的计算机可读存储介质。 被配置为练习该方法的系统监视用户话语以产生会话上下文。 然后，系统接收与用于触发语音处理的非自然语言输入无关的当前用户话语。 该系统将当前用户话语与对话上下文进行比较，以产生上下文相似性得分，并且如果上下文相似性得分高于阈值，则将当前用户话语合并到对话上下文中。 如果上下文相似性得分低于阈值，则系统丢弃当前用户的话语。 该系统可以基于n-gram分布，困惑度分数和困惑阈值将当前用户话语与对话上下文进行比较。 或者，系统可以使用任务模型来将当前用户话语与对话上下文进行比较。

公开(公告)号：US20120072217A1

公开(公告)日：2012-03-22

申请号：US12884959

申请日：2010-09-17

Applicant: Srinivas BANGALORE ,  Junlan Feng ,  Michael Johnston ,  Taniya Mishra

Inventor： Srinivas BANGALORE ,  Junlan Feng ,  Michael Johnston ,  Taniya Mishra

IPC: G10L15/06

CPC classification number: G10L15/1807 ,  G10L25/54 ,  G10L25/63 ,  G10L2015/226 ,  G10L2015/227

Abstract: Disclosed herein are systems, methods, and non-transitory computer-readable storage media for approximating relevant responses to a user query with voice-enabled search. A system practicing the method receives a word lattice generated by an automatic speech recognizer based on a user speech and a prosodic analysis of the user speech, generates a reweighted word lattice based on the word lattice and the prosodic analysis, approximates based on the reweighted word lattice one or more relevant responses to the query, and presents to a user the responses to the query. The prosodic analysis examines metalinguistic information of the user speech and can identify the most salient subject matter of the speech, assess how confident a speaker is in the content of his or her speech, and identify the attitude, mood, emotion, sentiment, etc. of the speaker. Other information not described in the content of the speech can also be used.

Abstract translation: 本文公开了用于使用具有语音的搜索近似对用户查询的相关响应的系统，方法和非暂时的计算机可读存储介质。 实施该方法的系统基于用户语音和用户语音的韵律分析接收由自动语音识别器生成的单词格，基于单词格点和韵律分析生成重新加权的词格，基于重新加权的词近似 对查询进行一个或多个相关响应，并向用户呈现对查询的响应。 韵律分析检查用户言语的金属语言信息，可以确定演讲中最突出的主题，评估演讲者在演讲内容中的信心，并确定态度，情绪，情感等情绪。 的演讲者。 还可以使用在语音内容中未描述的其他信息。

公开(公告)号：US20100100515A1

公开(公告)日：2010-04-22

申请号：US12252732

申请日：2008-10-16

Applicant: Srinivas BANGALORE ,  David J. SMITH

Inventor： Srinivas BANGALORE ,  David J. SMITH

IPC: G06N5/02 ,  G06F17/21

CPC classification number: G06F17/248

Abstract: A method for monitoring edits to a template for responding to an incoming communication includes categorizing the incoming communication into a category associated with the template for a response to the incoming communication. The method also includes determining distances between the template and each of a set of responses based on the template, at a predetermined level of granularity. The method also includes coding the template in accordance with the determined distances and displaying the coded template. A method for extracting a new template based on responses to an existing template includes selecting factors that affect quantitative measures for preparing a response to the incoming communication. The method includes using a mathematical model of the factors to cluster a set of responses created based on the existing template into two clusters. The method further includes restricting a first cluster centroid to be the existing template and searching for a second cluster centroid for a second cluster.

Abstract translation: 用于监视对模板的编辑以响应传入通信的方法包括将传入通信分类为与模板相关联的类别以对于进入通信的响应。 该方法还包括以预定的粒度级确定基于模板的模板与一组响应中的每一个之间的距离。 该方法还包括根据确定的距离对模板进行编码并显示编码的模板。 基于对现有模板的响应来提取新模板的方法包括选择影响对进入通信的响应的定量测量的因素。 该方法包括使用这些因素的数学模型将基于现有模板创建的一组响应聚类成两个聚类。 该方法还包括将第一聚类中心限制为现有模板并搜索第二聚簇的第二聚类质心。

公开(公告)号：US20090192838A1

公开(公告)日：2009-07-30

申请号：US12019821

申请日：2008-01-25

Applicant: Srinivas BANGALORE ,  Mazin Gilbert

Inventor： Srinivas BANGALORE ,  Mazin Gilbert

IPC: G06Q10/00

CPC classification number: H04L51/02 ,  G06Q10/06 ,  G06Q10/06311 ,  G06Q10/063112 ,  G06Q30/0207 ,  G06Q30/0245

Abstract: A system and method disclosed for using and updating a database of template responses for a live agent in response to user communications. The method includes computing an average string distance between each response from a live agent and a template, use to generate the response, modifying the computed average string distance based on a customer satisfaction score associated with each response and selecting a response that minimizes the computed average string distance and maximizes customer satisfaction. Upon receiving a further communication on a certain issue, the system presents a prototype response that has been added to the template database to the live agent for use in generating a response to the further communication that reduces handling time and increases customer satisfaction.

Abstract translation: 公开了用于响应于用户通信使用和更新活动代理的模板响应的数据库的系统和方法。 该方法包括计算来自活动代理和模板的每个响应之间的平均字符串距离，用于生成响应，基于与每个响应相关联的客户满意度得分修改所计算的平均字符串距离，并且选择使计算出的平均值最小化的响应 弦距，并最大化客户满意度。 在一个特定的问题上接收到进一步的通信之后，系统呈现已经被添加到模板数据库中的实时代理的原型响应，用于产生对进一步通信的响应，这减少了处理时间并提高了客户满意度。

公开(公告)号：US20130144594A1

公开(公告)日：2013-06-06

申请号：US13311836

申请日：2011-12-06

Applicant: Srinivas BANGALORE ,  Kateryna Kuksenok

Inventor： Srinivas BANGALORE ,  Kateryna Kuksenok

IPC: G06F17/28

CPC classification number: G06F17/2854 ,  G06F3/04842 ,  G06F17/2836 ,  G06F17/289

Abstract: Disclosed herein are systems, methods, and non-transitory computer-readable storage media for presenting a machine translation and alternative translations to a user, where a selection of any particular alternative translation results in the re-ranking of the remaining alternatives. The system then presents these re-ranked alternatives to the user, who can continue proofing the machine translation using the re-ranked alternatives or by typing an improved translation. This process continues until the user indicates that the current portion of the translation is complete, at which point the system moves to the next portion.

Abstract translation: 本文公开了用于向用户呈现机器翻译和替代翻译的系统，方法和非暂时的计算机可读存储介质，其中任何特定替代翻译的选择导致其余替代方案的重新排序。 然后，该系统将这些重新排列的替代品呈现给用户，他们可以使用重新排列的替代品或通过输入改进的翻译来继续打印机器翻译。 该过程继续，直到用户指示翻译的当前部分完成，在该点系统移动到下一部分。

公开(公告)号：US20120253799A1

公开(公告)日：2012-10-04

申请号：US13072920

申请日：2011-03-28

Applicant: Srinivas BANGALORE ,  Robert Bell ,  Diamantino Antonio Caseiro ,  Mazin Gilbert ,  Patrick Haffner

Inventor： Srinivas BANGALORE ,  Robert Bell ,  Diamantino Antonio Caseiro ,  Mazin Gilbert ,  Patrick Haffner

IPC: G10L15/00

CPC classification number: G10L15/183 ,  G10L15/06 ,  G10L15/065 ,  G10L15/22 ,  G10L15/30 ,  G10L2015/0635 ,  G10L2015/0636 ,  G10L2015/228

Abstract: Disclosed herein are systems, methods, and non-transitory computer-readable storage media for generating domain-specific speech recognition models for a domain of interest by combining and tuning existing speech recognition models when a speech recognizer does not have access to a speech recognition model for that domain of interest and when available domain-specific data is below a minimum desired threshold to create a new domain-specific speech recognition model. A system configured to practice the method identifies a speech recognition domain and combines a set of speech recognition models, each speech recognition model of the set of speech recognition models being from a respective speech recognition domain. The system receives an amount of data specific to the speech recognition domain, wherein the amount of data is less than a minimum threshold to create a new domain-specific model, and tunes the combined speech recognition model for the speech recognition domain based on the data.

Abstract translation: 本文公开的是当语音识别器不能访问语音识别模型时通过组合和调整现有语音识别模型来生成针对感兴趣域的域特定语音识别模型的系统，方法和非暂时计算机可读存储介质 对于该感兴趣的领域，并且当可用的域特定数据低于创建新的特定领域语音识别模型的最小期望阈值时。 配置为练习该方法的系统识别语音识别域并组合一组语音识别模型，语音识别模型集合的每个语音识别模型来自相应的语音识别域。 系统接收特定于语音识别域的数据量，其中数据量小于最小阈值以创建新的域特定模型，并且基于该数据调谐用于语音识别域的组合语音识别模型 。

公开(公告)号：US20120221331A1

公开(公告)日：2012-08-30

申请号：US13465659

申请日：2012-05-07

Applicant: Srinivas BANGALORE ,  Mazin G. Rahim ,  Junlan Feng

Inventor： Srinivas BANGALORE ,  Mazin G. Rahim ,  Junlan Feng

IPC: G10L15/26

CPC classification number: G06F17/243 ,  G06Q10/06311 ,  G06Q30/0201 ,  G06Q30/0635 ,  H04L12/6418 ,  H04M3/4938 ,  H04M7/0009 ,  H04M7/0036 ,  H04M2201/40 ,  H04M2201/60 ,  H04M2203/355

Abstract: A system and method provides a natural language interface to world-wide web content. Either in advance or dynamically, webpage content is parsed using a parsing algorithm. A person using a telephone interface can provide speech information, which is converted to text and used to automatically fill in input fields on a webpage form. The form is then submitted to a database search and a response is generated. Information contained on the responsive webpage is extracted and converted to speech via a text-to-speech engine and communicated to the person.

Abstract translation: 系统和方法为世界各地的Web内容提供了一种自然语言界面。 提前或动态地，使用解析算法解析网页内容。 使用电话接口的人可以提供语音信息，其被转换成文本并用于自动填写网页表单上的输入字段。 然后将表单提交到数据库搜索，并生成响应。 包含在响应网页上的信息被提取并经由文本到语音引擎转换成语音，并传达给该人。

公开(公告)号：US20120150531A1

公开(公告)日：2012-06-14

申请号：US12963126

申请日：2010-12-08

Applicant: Srinivas BANGALORE ,  Sumit Chopra

Inventor： Srinivas BANGALORE ,  Sumit Chopra

IPC: G06F17/27

CPC classification number: G06F17/28

Abstract: Disclosed herein are systems, methods, and non-transitory computer-readable storage media for learning latent representations for natural language tasks. A system configured to practice the method analyzes, for a first natural language processing task, a first natural language corpus to generate a latent representation for words in the first corpus. Then the system analyzes, for a second natural language processing task, a second natural language corpus having a target word, and predicts a label for the target word based on the latent representation. In one variation, the target word is one or more word such as a rare word and/or a word not encountered in the first natural language corpus. The system can optionally assigning the label to the target word. The system can operate according to a connectionist model that includes a learnable linear mapping that maps each word in the first corpus to a low dimensional latent space.

Abstract translation: 本文公开了用于学习自然语言任务的潜在表示的系统，方法和非暂时的计算机可读存储介质。 一种被配置为练习该方法的系统，分析第一自然语言处理任务中的第一自然语言语料库以产生第一语料库中的单词的潜在表示。 然后，系统针对第二自然语言处理任务分析具有目标词的第二自然语言语料库，并且基于潜在表示来预测目标词的标签。 在一个变体中，目标词是一个或多个单词，例如在第一自然语言语料库中不遇到的罕见单词和/或单词。 系统可以选择将标签分配给目标字。 该系统可以根据连接主义模型来操作，该连接主义模型包括将第一语料库中的每个单词映射到低维空间的可学习的线性映射。

公开(公告)号：US20120089605A1

公开(公告)日：2012-04-12

申请号：US12901075

申请日：2010-10-08

Applicant: Srinivas BANGALORE ,  Junlan FENG ,  Michael James Robert JOHNSTON ,  Taniya MISHRA

Inventor： Srinivas BANGALORE ,  Junlan FENG ,  Michael James Robert JOHNSTON ,  Taniya MISHRA

IPC: G06F17/30

CPC classification number: G06F17/30976 ,  G06F17/30345 ,  G06F17/30997

Abstract: Delivering targeted content includes collecting, via at least one tangible processor, user activity data for users during a specified time period. questions asked by the users during the specified time period are extracted from the user activity data, via the at least one tangible processor, and stored in user profiles for the users. The user profiles are clustered, via the at least one tangible processor, based on the questions asked. Targeted content is delivered, via the at least one tangible processor, to a subset of the users based on the clustering.

Abstract translation: 提供目标内容包括通过至少一个有形处理器在指定的时间段内收集用户的用户活动数据。 通过至少一个有形处理器从用户活动数据中提取在指定时间段内由用户询问的问题，并存储在用户的用户简档中。 基于所提出的问题，通过至少一个有形处理器对用户简档进行聚类。 基于聚类，经由至少一个有形处理器将目标内容传递给用户的子集。