公开(公告)号：US09460716B1

公开(公告)日：2016-10-04

申请号：US13937408

申请日：2013-07-09

Applicant: Google Inc.

Inventor： Mark Edward Epstein ,  Martin Jansche

IPC: G10L15/22 ,  G10L15/07 ,  G10L15/065

CPC classification number: G10L15/065 ,  G10L15/063 ,  G10L15/07 ,  G10L15/26 ,  G10L2015/227

Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for acoustic model generation. One of the methods includes identifying one or more demographic characteristics for a user of a social networking site. The method includes receiving speech data from the user, the speech data associated with a user device. The method includes storing the speech data associated with demographic characteristics of the user and the user device.

Abstract translation: 方法，系统和装置，包括在计算机存储介质上编码的计算机程序，用于声学模型生成。 方法之一包括为社交网站的用户识别一个或多个人口特征。 所述方法包括从用户接收语音数据，与用户设备相关联的语音数据。 该方法包括存储与用户和用户设备的人口特征相关联的语音数据。

公开(公告)号：US09437189B2

公开(公告)日：2016-09-06

申请号：US14290090

申请日：2014-05-29

Applicant: Google Inc.

Inventor： Mark Edward Epstein ,  Lucy Vasserman

IPC: G10L15/00 ,  G10L15/183 ,  G10L15/06 ,  G06F9/44 ,  G10L15/18

CPC classification number: G10L15/183 ,  G06F8/10 ,  G10L15/063 ,  G10L15/1815

Abstract: Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for generating language models. In some implementations, data is accessed that indicates a set of classes corresponding to a concept. A first language model is generated in which a first class represents the concept. A second language model is generated in which second classes represent the concept. Output of the first language model and the second language model is obtained, and the outputs are evaluated. A class from the set of classes is selected based on evaluating the output of the first language model and the output of the second language model. In some implementations, the first class and the second class are selected from a parse tree or other data that indicates relationships among the classes in the set of classes.

Abstract translation: 方法，系统和装置，包括在计算机存储介质上编码的计算机程序，用于生成语言模型。 在一些实现中，访问指示与概念相对应的一组类的数据。 生成第一语言模型，其中第一类表示概念。 生成第二种语言模型，其中第二类表示概念。 获得第一语言模型和第二语言模型的输出，并对输出进行评估。 基于评估第一语言模型的输出和第二语言模型的输出来选择来自该组类的类。 在一些实现中，从解析树或指示该组类中的类之间的关系的其他数据中选择第一类和第二类。

公开(公告)号：US09286892B2

公开(公告)日：2016-03-15

申请号：US14242228

申请日：2014-04-01

Applicant: Google Inc.

Inventor： Pedro J. Moreno Mengibar ,  Mark Edward Epstein

IPC: G10L15/18 ,  G10L15/06 ,  G06F17/27

CPC classification number: G10L15/18 ,  G06F17/2705 ,  G06F17/2775 ,  G06F17/2785 ,  G10L15/063 ,  G10L15/1815

Abstract: Some implementations include a computer-implemented method. The method can include providing a training set of text samples to a semantic parser that associates text samples with actions. The method can include obtaining, for each of one or more of the text samples of the training set, data that indicates one or more domains that the semantic parser has associated with the text sample. For each of one or more domains, a subset of the text samples of the training set can be generated that the semantic parser has associated with the domain. Using the subset of text samples associated with the domain, a language model can be generated for one or more of the domain. Speech recognition can be performed on an utterance using the one or more language models that are generated for the one or more of the domains.

Abstract translation: 一些实现包括计算机实现的方法。 该方法可以包括向语义解析器提供将文本样本与动作相关联的文本样本的训练集。 该方法可以包括为训练集合的一个或多个文本样本中的每一个获取指示语义解析器与文本样本相关联的一个或多个域的数据。 对于一个或多个域中的每一个，可以生成训练集的文本样本的子集，语义解析器已经与域相关联。 使用与域相关联的文本样本的子集，可以为一个或多个域生成语言模型。 可以使用为一个或多个域生成的一个或多个语言模型在话语上执行语音识别。

公开(公告)号：US20150279360A1

公开(公告)日：2015-10-01

申请号：US14242228

申请日：2014-04-01

Applicant: Google Inc.

Inventor： Pedro J. Moreno Mengibar ,  Mark Edward Epstein

IPC: G10L15/18 ,  G06F17/27 ,  G10L15/06

CPC classification number: G10L15/18 ,  G06F17/2705 ,  G06F17/2775 ,  G06F17/2785 ,  G10L15/063 ,  G10L15/1815

Abstract: Some implementations include a computer-implemented method. The method can include providing a training set of text samples to a semantic parser that associates text samples with actions. The method can include obtaining, for each of one or more of the text samples of the training set, data that indicates one or more domains that the semantic parser has associated with the text sample. For each of one or more domains, a subset of the text samples of the training set can be generated that the semantic parser has associated with the domain. Using the subset of text samples associated with the domain, a language model can be generated for one or more of the domain. Speech recognition can be performed on an utterance using the one or more language models that are generated for the one or more of the domains.

Abstract translation: 一些实现包括计算机实现的方法。 该方法可以包括向语义解析器提供将文本样本与动作相关联的文本样本的训练集。 该方法可以包括为训练集合的一个或多个文本样本中的每一个获取指示语义解析器与文本样本相关联的一个或多个域的数据。 对于一个或多个域中的每一个，可以生成训练集的文本样本的子集，语义解析器已经与域相关联。 使用与域相关联的文本样本的子集，可以为一个或多个域生成语言模型。 可以使用为一个或多个域生成的一个或多个语言模型在话语上执行语音识别。

公开(公告)号：US20140074470A1

公开(公告)日：2014-03-13

申请号：US13948996

申请日：2013-07-23

Applicant: Google Inc.

Inventor： Martin Jansche ,  Mark Edward Epstein ,  Ciprian I. Chelba

IPC: G10L15/06

CPC classification number: G10L15/063 ,  G10L15/187 ,  G10L2015/025

Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for improved pronunciation. One of the methods includes receiving data that represents an audible pronunciation of the name of an individual from a user device. The method includes identifying one or more other users that are members of a social circle that the individual is a member. The method includes identifying one or more devices associated with the other users. The method also includes providing information that identifies the individual and the data representing the audible pronunciation to the one or more identified devices.

Abstract translation: 方法，系统和装置，包括在计算机存储介质上编码的计算机程序，用于改进发音。 其中一种方法包括从用户设备接收表示个人名称的听觉发音的数据。 该方法包括识别作为该个体是成员的社交圈的成员的一个或多个其他用户。 该方法包括识别与其他用户相关联的一个或多个设备。 该方法还包括提供将个体和表示可听见的发音的数据标识给一个或多个所识别的设备的信息。

公开(公告)号：US09747895B1

公开(公告)日：2017-08-29

申请号：US13936858

申请日：2013-07-08

Applicant: Google Inc.

Inventor： Martin Jansche ,  Mark Edward Epstein

IPC: G10L15/183 ,  G10L15/26 ,  G10L15/18 ,  G06F17/21

CPC classification number: G10L15/183

Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for building language models. One of the methods includes identifying a first group of one or more users associated with a user in a social network. The method includes identifying first linguistic information associated with the first group. The method includes generating a first language model based on the first linguistic information. The method includes identifying a second group of one or more users associated with the user. The method includes identifying second linguistic information associated with the second group. The method includes generating a second language model based on the second linguistic information. The method includes associating the first language model and the second language model with the user.

公开(公告)号：US20150287410A1

公开(公告)日：2015-10-08

申请号：US13844312

申请日：2013-03-15

Applicant: Google Inc.

Inventor： Pedro J. Moreno Mengibar ,  Mark Edward Epstein

IPC: G10L17/00

CPC classification number: G10L17/005 ,  G06Q30/0271 ,  G10L15/1822 ,  G10L25/60 ,  G10L25/63

Abstract: Systems, apparatus and method for speech and semantic parsing for content selection. In an aspect, a method includes selecting, for each of a plurality of voice query analyzers, an analyzer output parameter; generating a voice query model for voice queries, the voice query model including analysis fields, wherein each analysis field in at least a first portion of the analysis fields corresponds to a corresponding analyzer output parameter; receiving, from a plurality of content item providers, voice query selection data that describes analyzer output parameter values for the voice query model that satisfy selection criteria for the content item provider; and persisting the voice query selection data for the content item providers to a computer memory device; wherein the voice query analyzers include a semantic analyzer and a biometric analyzer.

Abstract translation: 用于内容选择的语音和语义解析的系统，设备和方法。 一方面，一种方法包括为多个语音查询分析器中的每一个选择分析器输出参数; 生成用于语音查询的语音查询模型，所述语音查询模型包括分析字段，其中所述分析字段的至少第一部分中的每个分析字段对应于相应的分析器输出参数; 从多个内容项提供者接收描述满足所述内容项提供者的选择标准的语音查询模型的分析器输出参数值的语音查询选择数据; 并且将所述内容项提供者的语音查询选择数据持久化到计算机存储装置; 其中所述语音查询分析器包括语义分析器和生物测定分析器。

公开(公告)号：US09020809B1

公开(公告)日：2015-04-28

申请号：US13780804

申请日：2013-02-28

Applicant: Google Inc.

Inventor： Pedro J. Mengibar ,  Mark Edward Epstein ,  Fadi Biadsy

IPC: G06F17/27 ,  G06F17/28 ,  G06F17/20

CPC classification number: G10L15/063 ,  G06F17/2785 ,  G10L15/19 ,  G10L2015/0631

Abstract: A method includes accessing data specifying a set of actions, each action defining a user device operation and for each action: accessing a corresponding set of command sentences for the action, determining first n-grams in the set of command sentences that are semantically relevant for the action, determining second n-grams in the set of command sentences that are semantically irrelevant for the action, generating a training set of command sentences from the corresponding set of command sentences, the generating the training set of command sentences including removing each second n-gram from each sentence in the corresponding set of command sentences for the action, and generating a command model from the training set of command sentences configured to generate an action score for the action for an input sentence based on: first n-grams for the action, and second n-grams for the action that are also second n-grams for all other actions.

公开(公告)号：US09529898B2

公开(公告)日：2016-12-27

申请号：US14656027

申请日：2015-03-12

Applicant: Google Inc.

Inventor： Mark Edward Epstein ,  Vladislav Schogol

IPC: G06F17/30 ,  G06F17/27

CPC classification number: G06F17/30707 ,  G06F17/2715 ,  G06F17/2775

Abstract: This document describes, among other things, a computer-implemented method. The method can include obtaining a plurality of text samples that each include one or more terms belonging to a first class of terms. The plurality of text samples can be classified into a plurality of groups of text samples. Each group of text samples can correspond to a different sub-class of terms. For each of the groups of text samples, a sub-class context model can be generated based on the text samples in the respective group of text samples. Particular ones of the sub-class context models that are determined to be similar can be merged to generate a hierarchical set of context models. Further, the method can include selecting particular ones of the context models and generating a class-based language model based on the selected context models.

Abstract translation: 本文档尤其描述了计算机实现的方法。 该方法可以包括获得多个文本样本，每个文本样本包括属于第一类术语的一个或多个术语。 多个文本样本可以分为多组文本样本。 每组文本样本可以对应于不同的子类的术语。 对于每组文本样本，可以基于相应文本样本组中的文本样本生成子类上下文模型。 被确定为相似的子类上下文模型中的特定的上下文模型可以被合并以生成上下文模型的分层集合。 此外，该方法可以包括选择上下文模型中的特定模型，并且基于所选择的上下文模型生成基于类的语言模型。

公开(公告)号：US09146919B2

公开(公告)日：2015-09-29

申请号：US13830969

申请日：2013-03-14

Applicant: Google Inc.

Inventor： Mark Edward Epstein ,  Pedro J. Mengibar

IPC: G06F17/28 ,  G06F17/27

CPC classification number: G06F17/289 ,  G06F17/278 ,  G06F17/28

Abstract: Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for training recognition canonical representations corresponding to named-entity phrases in a second natural language based on translating a set of allowable expressions with canonical representations from a first natural language, which may be generated by expanding a context-free grammar for the allowable expressions for the first natural language.

Abstract translation: 方法，系统和装置，包括在计算机存储介质上编码的计算机程序，用于训练对应于第二自然语言中的命名实体短语的识别规范表示，其基于从第一自然语言的规范表示转换一组可允许表达， 这可以通过扩展第一自然语言的允许表达式的上下文无关语法来产生。