公开(公告)号：WO2010003117A3

公开(公告)日：2010-11-18

申请号：PCT/US2009049613

申请日：2009-07-02

Applicant: GOOGLE INC ,  MACHEREY WOLFGANG ,  KUMAR SHANKAR ,  TROMBLE ROY W ,  OCH FRANZ JOSEF ,  THAYER IGNACIO E ,  USZKOREIT JAKOB

Inventor： MACHEREY WOLFGANG ,  KUMAR SHANKAR ,  TROMBLE ROY W ,  OCH FRANZ JOSEF ,  THAYER IGNACIO E ,  USZKOREIT JAKOB

IPC: G06F17/00 ,  G06F9/06 ,  G06F11/00 ,  G06F17/20

CPC classification number: G06F17/2818

Abstract: Methods, systems, and apparatus, including computer program products, for language translation are disclosed. In one implementation, a method is provided. The method includes accessing a hypothesis space; performing decoding on the hypothesis space to obtain a translation hypothesis that minimizes an expected error in classification calculated relative to an evidence space; and providing the obtained translation hypothesis for use by a user as a suggested translation in a target translation.

公开(公告)号：WO2009129315A1

公开(公告)日：2009-10-22

申请号：PCT/US2009/040677

申请日：2009-04-15

Applicant: MOBILE TECHNOLOGIES, LLC ,  LANE, Ian, R. ,  WAIBEL, Alexender

Inventor： LANE, Ian, R. ,  WAIBEL, Alexender

IPC: G10L15/18 ,  G06F17/28

CPC classification number: G06F17/2818 ,  G06F17/2735 ,  G06F17/28 ,  G10L13/00 ,  G10L15/063 ,  G10L15/1815 ,  G10L15/183 ,  G10L15/22 ,  G10L15/26

Abstract: A method and apparatus are provided for updating the vocabulary of a speech translation system for translating a first language into a second language including written and spoken words. The method includes adding a new word in the first language to a first recognition lexicon of the first language and associating a description with the new word, wherein the description contains pronunciation and word class information. The new word and description are then updated in a first machine translation module associated with the first language. The first machine translation module contains a first tagging module, a first translation model and a first language module, and is configured to translate the new word to a corresponding translated word in the second language. Optionally, the invention may be used for bidirectional or multi-directional translation.

Abstract translation: 提供了一种用于更新用于将第一语言翻译成包括书面和口头单词的第二语言的语音翻译系统的词汇表的方法和装置。 该方法包括将第一语言中的新单词添加到第一语言的第一识别词典，并将描述与新单词相关联，其中该描述包含发音和单词类信息。 然后在与第一语言相关联的第一机器翻译模块中更新新的单词和描述。 第一机器翻译模块包含第一标记模块，第一翻译模型和第一语言模块，并且被配置为将新单词翻译成第二语言的对应翻译单词。 可选地，本发明可以用于双向或多向转换。

公开(公告)号：WO2009032472A3

公开(公告)日：2009-07-02

申请号：PCT/US2008072553

申请日：2008-08-08

Applicant: MICROSOFT CORP

Inventor： ITAGAKI MASAKI ,  AIKAWA TAKAKO ,  HE XIAODONG

IPC: G06F17/28 ,  G06F17/26 ,  G06F17/27

CPC classification number: G06F17/2818

Abstract: A method of determining the consistency of training data for a machine translation system is disclosed. The method includes receiving a signal indicative of a source language corpus and a target language corpus. A textual string is extracted from the source language corpus. The textual string is aligned with the target language corpus to identify a translation for the textual string from the target language corpus. A consistency index is calculated based on a relationship between the textual string from the source language corpus and the translation. An indication of thethe consistency index is stored on a tangible medium.

Abstract translation: 公开了一种确定机器翻译系统的训练数据的一致性的方法。 该方法包括接收指示源语言语料库和目标语言语料库的信号。 从源语言语料库中提取文本字符串。 文本字符串与目标语言语料库对齐，以从目标语言语料库中标识文本字符串的翻译。 基于源语言语料库的文本字符串与翻译之间的关系计算一致性索引。 一致性指标的指示存储在有形介质上。

公开(公告)号：WO2008053265A1

公开(公告)日：2008-05-08

申请号：PCT/IB2006/003068

申请日：2006-10-31

Applicant: NOKIA CORPORATION ,  NOKIA, INC. ,  WANG, Hao, Ul ,  TANG, Yuezhong ,  HU, Rile ,  WANG, Xia, S.

Inventor： WANG, Hao, Ul ,  TANG, Yuezhong ,  HU, Rile ,  WANG, Xia, S.

IPC: G06K9/20 ,  G06F17/28

CPC classification number: G06F17/2818

Abstract: Improved methods, apparatus and computer program products implementing an index-based search algorithm for use with a translation program are disclosed herein. In the method, a group of words (such as menu items) are selected for translation from one language to another. Then, an index for use in conducting search-and-match operations during language translation operations is constructed for the group of words. In constructing an index, the frequency of appearance of characters used in the words to be translated is calculated. Then, coverage calculations are made based on various selections of key characters. Next, key characters are chosen based on the coverage calculations. Then, a key character index is constructed using the key characters. In one embodiment, the key character index comprises for each word an identification of which key characters appear in the word, and where the key characters appear in the word. After the key character index is constructed, words not containing key characters are identified in a specific set. The key character index and specific set index are then used during search and match language translation operations.

Abstract translation: 本文公开了改进的实现与翻译程序一起使用的基于索引的搜索算法的方法，装置和计算机程序产品。 在该方法中，选择一组单词（例如菜单项）以从一种语言翻译成另一种语言。 然后，针对该组组构建用于在语言翻译操作期间进行搜索和匹配操作的索引。 在构建索引时，计算要翻译的单词中使用的字符的出现频率。 然后，根据关键字符的各种选择进行覆盖计算。 接下来，基于覆盖率计算来选择关键字符。 然后，使用关键字符构建关键字符索引。 在一个实施例中，关键字索引包括对于每个单词，在单词中出现哪些关键字符的标识以及该单词中出现关键字符的标识。 在构建关键字符索引之后，在特定集合中识别不包含关键字符的字。 然后在搜索和匹配语言翻译操作中使用关键字符索引和特定设置索引。

公开(公告)号：WO2003105023A2

公开(公告)日：2003-12-18

申请号：PCT/US2003/009571

申请日：2003-03-26

Applicant: UNIVERSITY OF SOUTHERN CALIFORNIA ,  SORICUT, Radu ,  MARCU, Daniel ,  KNIGHT, Kevin

Inventor： SORICUT, Radu ,  MARCU, Daniel ,  KNIGHT, Kevin

IPC: G06F17/28

CPC classification number: G06F17/2818 ,  G06F17/2827 ,  G06F17/2845

Abstract: A statistical machine translation (MT) system may use a large monolingual corpus to improve the accuracy of translated phrases/sentences. The MT system may produce a alternative translations and use the large monolingual corpus to (re)rank the alternative translations.

Abstract translation: 统计机器翻译（MT）系统可以使用大的单语语料库来提高翻译短语/句子的准确性。 MT系统可以产生替代翻译，并使用大单语语料库（重新）排列替代翻译。

公开(公告)号：WO03083709A2

公开(公告)日：2003-10-09

申请号：PCT/US0309749

申请日：2003-03-28

Applicant: UNIV SOUTHERN CALIFORNIA ,  KOEHN PHILIPP ,  KNIGHT KEVIN

Inventor： KOEHN PHILIPP ,  KNIGHT KEVIN

IPC: G06F17/27 ,  G06F17/28

CPC classification number: G06F17/2775 ,  G06F17/271 ,  G06F17/2818 ,  G06F17/2827

Abstract: A method includes detecting a syntactic chunk in a source string in a first language, assigning a syntactic label to the detected syntactic chunk in the source string, mapping the detected syntactic chunk in the source string to a syntactic chunk in a target string in a second language, said mapping based on the assigned syntactic label, and translating the source string into a possible translation in the second language.

Abstract translation: 一种方法包括以第一语言检测源字符串中的句法块，向源字符串中检测到的句法块分配句法标签，将源字符串中检测到的句法块映射到目标字符串中的句法块中， 语言，基于所分配的句法标签的所述映射，以及将源字符串转换为第二语言的可能翻译。

公开(公告)号：WO0129697A9

公开(公告)日：2002-08-08

申请号：PCT/US0041256

申请日：2000-10-17

Applicant: SONY ELECTRONICS INC

Inventor： DUAN LEI ,  FRANZ ALEXANDER ,  HORIGUCHI KEIKO

IPC: G06F17/27 ,  G06F17/28

CPC classification number: G06F17/277 ,  G06F17/2818

Abstract: A method and system for reducing lexical ambiguity in an input stream (302) are described. In one embodiment, the input stream (302) is broken into tokens (308). The tokens are used to create a connection graph comprising a number of paths. Each of the paths is assigned a cost (318). At least one best path is defined based upon a corresponding cost to generate an output graph (322). The generated output graph is provided to reduce lexical ambiguity.

Abstract translation: 描述用于减少输入流（302）中的词汇模糊度的方法和系统。 在一个实施例中，输入流（302）被分成令牌（308）。 令牌用于创建包括多个路径的连接图。 每个路径被分配成本（318）。 基于生成输出图（322）的相应成本来定义至少一个最佳路径。 提供生成的输出图以减少词汇模糊度。

公开(公告)号：WO02039318A1

公开(公告)日：2002-05-16

申请号：PCT/US2000/042120

申请日：2000-11-09

IPC: G06F17/28

CPC classification number: G06F17/2836 ,  G06F17/2818

Abstract: In one aspect, the invention relates to a method of allowing a user to view and modify a weighting associated with a translation of a source language string. The method includes displaying to the user the weighting associated with the translation of the source language string, the weighting for use by a translation engine in selecting the translation and allowing the usr to modify the weighting associated with the translation. In one embodiment, the method further includes allowing the user to reset the weighting back to a default value subsequent to user modification of the weighting.

Abstract translation: 一方面，本发明涉及允许用户查看和修改与源语言字符串的翻译相关联的加权的方法。 该方法包括向用户显示与源语言字符串的翻译相关联的加权，翻译引擎在选择翻译时使用的加权，并允许用户修改与翻译相关联的加权。 在一个实施例中，该方法还包括允许用户在加权的用户修改之后将加权重置为默认值。

公开(公告)号：WO0186489A2

公开(公告)日：2001-11-15

申请号：PCT/US0115380

申请日：2001-05-11

Applicant: UNIV SOUTHERN CALIFORNIA ,  MARCU DANIEL ,  KNIGHT KEVIN

Inventor： MARCU DANIEL ,  KNIGHT KEVIN

IPC: G06F17/20 ,  G06F17/27 ,  G06F17/28 ,  G06F17/30

CPC classification number: G06F17/271 ,  G06F17/2715 ,  G06F17/274 ,  G06F17/2785 ,  G06F17/279 ,  G06F17/2818 ,  G06F17/30719

Abstract: A discourse structure for an input text segment is determined by generating a set of one or more discourse parsing decision rules based on a training set, and determining a discourse structure for the input text segment by applying the generated set of discourse parsing decision rules to the input text segment. A tree structure is summarized by generating a set of one or more summarization decision rules based on a training set, and compressing the tree structure by applying the generated set of summarization decision rules to the tree structure. Alternatively, summarization is accomplished by parsing an input text segment to generate a parse tree for the input segment, generating a plurality of potential solutions, applying a statistical model to determine a probability of correctness for each of potential solution, and extracting one or more high-probability solutions based on the solutions' respective determined probabilities of correctness.

Abstract translation: 用于输入文本段的话语结构通过基于训练集合生成一组一个或多个话语解析决定规则来确定，并且通过将所生成的话语解析决策规则集合应用于所述语篇分析决策规则来确定所述输入文本段的话语结构 输入文字段 通过基于训练集合生成一组一个或多个摘要决策规则来总结树结构，并且通过将生成的汇总决策规则集应用于树结构来压缩树结构。 或者，通过解析输入文本段以生成用于输入段的解析树，生成多个潜在解决方案，应用统计模型以确定每个潜在解的准确概率并提取一个或多个高 基于解决方案各自确定的正确概率的概率解决方案。

公开(公告)号：WO2001029697A1

公开(公告)日：2001-04-26

申请号：PCT/US2000/041256

申请日：2000-10-17

Applicant: SONY ELECTRONICS, INC.

Inventor： DUAN, Lei ,  FRANZ, Alexander ,  HORIGUCHI, Keiko

IPC: G06F17/27

CPC classification number: G06F17/277 ,  G06F17/2818

Abstract: A method and system for reducing lexical ambiguity in an input stream (302) are described. In one embodiment, the input stream (302) is broken into tokens (308). The tokens are used to create a connection graph comprising a number of paths. Each of the paths is assigned a cost (318). At least one best path is defined based upon a corresponding cost to generate an output graph (322). The generated output graph is provided to reduce lexical ambiguity.

Abstract translation: 描述用于减少输入流（302）中的词汇模糊度的方法和系统。 在一个实施例中，输入流（302）被分成令牌（308）。 令牌用于创建包括多个路径的连接图。 每个路径被分配成本（318）。 基于生成输出图（322）的相应成本来定义至少一个最佳路径。 提供生成的输出图以减少词汇模糊度。