-
公开(公告)号：KR1020140080717A
公开(公告)日：2014-07-01
申请号：KR1020120146174
申请日：2012-12-14
Applicant: 한국전자통신연구원
CPC classification number: G06F17/30002
Abstract: The present invention relates to an apparatus for constructing a task graph in a dialog processing system based on the task graph. The apparatus includes a task graph management interface for providing a client environment that can correct a task graph; a library interface for providing a client environment that can correct a dialog library; a slot system management interface for providing a client environment that can correct a slot system; a task knowledge management interface for providing a client environment that can correct task knowledge; and a dialog system engine for reflecting a knowledge for controlling a corrected dialog flow of the task graph management interface, the library interface, the slot system management interface, and the task knowledge management interface.
Abstract translation: 本发明涉及一种基于任务图在对话处理系统中构建任务图的装置。 该装置包括用于提供可以校正任务图的客户端环境的任务图管理界面; 用于提供可以校正对话库的客户端环境的库接口; 插槽系统管理接口，用于提供可以校正插槽系统的客户端环境; 任务知识管理界面，用于提供可以纠正任务知识的客户端环境; 以及用于反映用于控制任务图管理接口，库接口，时隙系统管理接口和任务知识管理接口的校正对话流的知识的对话系统引擎。
-
公开(公告)号：KR1020140079598A
公开(公告)日：2014-06-27
申请号：KR1020120147536
申请日：2012-12-17
Applicant: 한국전자통신연구원
CPC classification number: G06F17/2854
Abstract: The present invention relates to an apparatus and a method for verifying a context which verify a dual meaning expression of an input sentence and use the verified expression as a context in interpretation or translation according to a user`s context. For said purpose, the present invention comprises: a dual meaning expression verification part for verifying the dual meaning expression in a first input sentence and a reversely translated sentence for the first input sentence by the input of a user; a context generation part for generating the verified dual meaning expression as the context; and a context control part for controlling the generated context to be applied in translating or interpreting the first input sentence or second input sentences inputted after the first input sentence is inputted. The present invention improves interpretation and translation quality of a bidirectional automatic interpretation and translation device and enhances the communication with a communication partner from a quality and a time perspective.
Abstract translation: 本发明涉及一种用于验证验证输入句子的双重含义表达的上下文的装置和方法，并且根据用户的上下文使用经验证的表达作为解释或翻译中的上下文。 为了所述目的，本发明包括：双重意义表达验证部分，用于通过用户的输入验证第一输入句中的双重含义表达和第一输入句的反向翻译句; 用于生成经验证的双重含义表达作为上下文的上下文生成部分; 以及上下文控制部分，用于控制在输入第一输入句子之后输入的第一输入句子或第二输入句子的翻译或解释中应用的生成上下文。 本发明改进了双向自动解释和翻译装置的解释和翻译质量，并且从质量和时间角度增强了与通信伙伴的通信。
-
公开(公告)号：KR1020140079543A
公开(公告)日：2014-06-27
申请号：KR1020120146908
申请日：2012-12-14
Applicant: 한국전자통신연구원
CPC classification number: G06F17/2836 , G06F17/289 , G10L15/26
Abstract: The present invention relates to an automatic interpreting and translating device to interpret and translate a source language into a target language. The automatic interpreting and translating device comprises: a voice recognizing unit to convert a voice of a source language to a text if the voice of the source language is inputted, and convert a voice of a target language to a text if the voice of the target language is inputted; and an automatic translating unit to translate the converted text of the source language to the target language to output a translation result of the target language if the voice of the source language is converted to the text of the source language by the voice recognizing unit, and correct the text of the converted target language to output a translation result of the target language if the voice of the target language is converted to the text of the target language by the voice recognizing unit. Therefore, the automatic interpreting and translating device performs the natural correction to the target language, thereby improving availability and performance of the device.
Abstract translation: 本发明涉及一种解释和翻译源语言为目标语言的自动翻译和翻译装置。 自动解释和翻译装置包括：语音识别单元，如果输入源语言的语音，则将源语言的语音转换为文本，并且如果目标语音的语音被转换为目标语言的语音 输入语言; 以及自动翻译单元，用于将源语言的转换文本转换为目标语言，以便如果通过语音识别单元将源语言的语音转换为源语言的文本，则输出目标语言的翻译结果，以及 如果目标语言的语音被语音识别单元转换成目标语言的文本，则校正转换的目标语言的文本以输出目标语言的翻译结果。 因此，自动解释和翻译装置对目标语言进行自然校正，从而提高设备的可用性和性能。
-
公开(公告)号：KR1020140052183A
公开(公告)日：2014-05-07
申请号：KR1020120117399
申请日：2012-10-22
Applicant: 한국전자통신연구원
Abstract: Provided is a dialogue management system which includes a dialogue system data base where dialogue system data base knowledge for generating a system response sentence about a temporary speech sentence is constructed, a dialogue system engine which generates the system response sentence based on the dialogue system base knowledge when a speech sentence of the same words as the temporary speech sentence is inputted, and a user interface which reconstructs or edits the dialogue system base knowledge as dialogue system knowledge according to external command input when there is an error in the system response sentence and performs a test on user intension and language comprehension about the speech sentence.
Abstract translation: 提供了一种对话管理系统，其包括对话系统数据库，其中构建用于生成关于临时语音句子的系统响应语句的对话系统数据库知识;基于对话系统基础知识生成系统响应语句的对话系统引擎 当输入与临时语音句子相同的语音的语句时，以及用户接口，当系统响应语句中存在错误并根据外部命令输入重建或编辑对话系统基础知识作为对话系统知识时，执行 对言语句的用户意图和语言理解的考验。
-
公开(公告)号：KR1020140049150A
公开(公告)日：2014-04-25
申请号：KR1020120114535
申请日：2012-10-16
Applicant: 한국전자통신연구원
IPC: G06F17/28
CPC classification number: G06F17/2854 , G06F17/2818 , G06F17/2872
Abstract: Disclosed are a machine translation post-processing system based on user participation and a method thereof. According to the present invention, the machine translation post-processing system based on user participation includes: a machine translation engine which provides statistical machine translation (SMT) and statistically-enhanced rule-based machine translation (RBMT) using a dual-language corpus database for a source language sentence inputted by a user; a user interface which is used to receives corrections for machine translation results by the machine translation engine from the user; a post-processing rule generator which compares the final sentence corrected through the user interface and the sentence translated by the machine translation engine and generates post-processing rules according to correction details; and an application range re-adjuster which re-adjusts the application range of rules for dual-language corpora based on the post-processing rules generated by the post-processing rule generator. [Reference numerals] (100) Machine translation engine; (110) Dual-language corpus database; (120) Data on a source language sentence and a translated sentence; (130) Translation memory; (200) User interface; (210) Corrected final sentence; (300) Post-processing rule generator; (400) Application range re-adjuster; (500) Post-processing knowledge storage unit; (510) Translation knowledge; (600) Analysis/generation data API; (610) Rules manually written by a user
Abstract translation: 公开了一种基于用户参与的机器翻译后处理系统及其方法。 根据本发明，基于用户参与的机器翻译后处理系统包括：使用双语言语料库数据库提供统计机器翻译（SMT）和统计增强的基于规则的机器翻译（RBMT）的机器翻译引擎 用于由用户输入的源语句; 用户接口，用于由用户接收机器翻译引擎对机器翻译结果的校正; 后处理规则生成器，其将通过用户界面修正的最终句子与由机器翻译引擎翻译的句子进行比较，并根据校正细节生成后处理规则; 以及应用范围重新调整器，其基于由后处理规则生成器生成的后处理规则重新调整双语言语料库规则的应用范围。 （附图标记）（100）机器翻译引擎; （110）双语语料库; （120）源语言句和翻译句的数据; （130）翻译记忆; （200）用户界面; （210）更正最后一句; （300）后处理规则生成器; （400）应用范围重新调整; （500）后处理知识存储单元; （510）翻译知识; （600）分析/生成数据API; （610）用户手动写的规则
-
Patent Agency Ranking
公开(公告)号：KR1020140049148A
公开(公告)日：2014-04-25
申请号：KR1020120114533
申请日：2012-10-16
Applicant: 한국전자통신연구원
CPC classification number: G06F17/2755 , G06F17/30011
Abstract: Disclosed are a word class tagging method based on morpheme separation and a device thereof. According to one embodiment of the present invention, the word class tagging method includes: a step of performing morpheme separation for a predetermined raw corpus; a step of searching for one or more documents similar to input text based on the result of the morpheme separation; a step of performing domain adaptation of a morpheme separation model and a word class tagging model using the searched similar documents; and a step of tagging word classes to the input text using the morpheme separation model and the word class tagging model to which domain adaptation is performed. The searching step uses a search database which is built by processing a group of documents in which the raw corpus is separated by morpheme based on the result of the morpheme separation or uses a plurality of clusters acquired by clustering the documents in which the raw corpus is separated by morpheme based on the result of the morpheme separation. By doing so, even if the word spacing among morphemes of the raw corpus is not clear, it is possible to perform domain adaptation of morpheme separation and word class tagging and to remove limitation of linguistic expandability. [Reference numerals] (AA) Start; (BB) End; (S110) Perform morpheme separation for a large-volume raw corpus; (S120) Search for documents similar to input text based on the morpheme separation result; (S130) Perform domain adaptation of a morpheme separation model and a word class tagging model using the searched similar documents; (S140) Tag word classes to the input text using the morpheme separation model and the word class tagging model to which domain adaptation is performed
Abstract translation: 公开了基于语素分离的词类标记方法及其装置。 根据本发明的一个实施例，词类标记方法包括：对预定原始语料库执行语素分离的步骤; 基于语素分离的结果搜索类似于输入文本的一个或多个文档的步骤; 使用搜索到的类似文档执行语素分离模型和单词类标签模型的域适配的步骤; 以及使用语素分离模型和执行域适配的单词类标签模型将单词类标记到输入文本的步骤。 搜索步骤使用搜索数据库，该搜索数据库是通过处理基于语素分离的结果的原始语料库被语素分离的一组文档来构建的，或者使用通过聚集原始语料库 基于语素分离结果的语素分离。 通过这样做，即使原始语料的语素之间的词间距不清楚，也可以进行语素分离和词类标注的领域适应，并消除语言扩展性的限制。 （附图标记）（AA）开始; （BB）结束; （S110）对大容量原始语料库执行语素分离; （S120）基于语素分离结果搜索类似于输入文本的文档; （S130）使用搜索到的类似文档对语素分离模型和单词类标注模型进行域适应; （S140）使用语素分离模型和进行域自适应的单词分类模型对输入文本进行标记词类
-
公开(公告)号：KR1020130123037A
公开(公告)日：2013-11-12
申请号：KR1020120046137
申请日：2012-05-02
Applicant: 한국전자통신연구원
IPC: G06Q50/30
CPC classification number: G06F17/289 , G06F17/2863 , G06F17/2872
Abstract: The present invention relates to an apparatus and method for providing a two-way automatic interpretation and translation service comprising: a first interpretation and translation unit for interpreting and translating a first language into a second language; a second interpretation and translation unit for interpreting and translating the second language into the first language; and a context information management unit for receiving conversational context and translation history information, and sharing and managing the conversational context and translation history information; wherein each of the first and second interpretation and translation units provides an interpretation service for receiving an input conversation in the first or second language in speech and outputting results of interpretation in speech in the second or first language, and a translation service for receiving an input conversation in the first or second language in text and outputting results of translation in text in the second or first language. [Reference numerals] (100) First interpretation and translation unit;(200) Second interpretation and translation unit;(300) Context information managing unit;(400) Two-way communication unit;(410) Context record managing unit;(420) Context translation information DB
Abstract translation: 本发明涉及一种用于提供双向自动解释和翻译服务的装置和方法，包括：用于将第一语言解释和翻译成第二语言的第一解释和翻译单元; 第二个翻译单位，用于将第二语言翻译成第一语言; 以及上下文信息管理单元，用于接收对话上下文和翻译历史信息，以及共享和管理会话语境和翻译历史信息; 其中第一和第二解释和翻译单元中的每一个提供用于在语音中接收第一或第二语言的输入会话的解释服务，并输出具有第二或第一语言的语音中的解释结果，以及用于接收输入的翻译服务 在第一语言或第二语言中的文本中的对话，并以第二语言或第一语言输出文本中的翻译结果。 （100）第一解释和翻译单元;（200）第二解释和翻译单元;（300）上下文信息管理单元;（400）双向通信单元;（410）上下文记录管理单元;（420） 上下文翻译信息DB
-
公开(公告)号：KR101301536B1
公开(公告)日：2013-09-04
申请号：KR1020090123135
申请日：2009-12-11
Applicant: 한국전자통신연구원
CPC classification number: G06F17/2881 , G06F17/2836
Abstract: 본 발명은 자동 번역 기능과 오류 보정 기능을 포함하는 외국어 작문 서비스에 관한 것으로, 특히 외국어 작문을 지원함에 있어서 외국어와 모국어의 혼용 입력을 허용하고, 단어/구/문장 단위의 자동 번역을 통해 외국어의 입력 부분과 모국어의 번역 결과의 결합문을 학습자에게 제공하며, 제공되는 결합문에 대한 오류 보정을 지원하는데 적합한 외국어 작문 서비스를 제공하고자 한다. 본 발명으로 인해, 보다 편리하고 정확한 외국어 작문 환경을 제공할 수 있으며, 외국어 학습을 위해 작문을 시도하는 학습자에 대해 학습 효율을 증대시킬 수 있다.
외국어 작문, 외국어 학습, 자동 번역, 오류 보정-
公开(公告)号：KR101301535B1
公开(公告)日：2013-09-04
申请号：KR1020090118671
申请日：2009-12-02
Applicant: 한국전자통신연구원
CPC classification number: G06F17/2818 , G06F17/2827 , G06F17/2881
Abstract: 본 발명은 하이브리드 기법으로 자동 번역하는 장치 및 방법에 관한 것으로, 본 발명은, 입력된 원시 언어 문장을 노드 단위로 일반화하고, 노드 단위로 일반화된 원시 언어 문장을 통계 기반 번역 지식을 이용하여 노드 표현으로 변환된 제 1 번역 결과를 생성하며, 제 1 번역 결과에 대해 패턴 기반 번역 지식을 이용하여 노드별 대역어 생성을 반복 수행한 후에, 제 1 번역 결과와 제 2 번역 결과를 통해 목적 언어 문장으로 출력함으로써, 통계 기반 번역 지식 및 패턴 기반 번역 지식을 이용한 하이브리드 기법으로 자동 번역의 모호성 문제를 해결할 수 있는 것이다.
자동 번역, 통계 기반 자동 번역, 패턴 기반 자동 번역-
公开(公告)号：KR101266361B1
公开(公告)日：2013-05-22
申请号：KR1020090085422
申请日：2009-09-10
Applicant: 한국전자통신연구원
CPC classification number: G06F17/2836
Abstract: 구조화된번역메모리기반의자동번역시스템및 이를이용한자동번영방법이제공된다. 이구조화된번역메모리기반의자동번역시스템및 이를이용한자동번역방법에서는, 문장단위이하의기 설정된언어패턴을변환, 삭제및 치환하는처리과정을통해상기기설정된언어패턴을부분번역패턴으로구조화하여번역메모리데이터베이스를구축하는번역메모리구축모듈과, 상기번역메모리데이터베이스를참조하여상기입력문을상기문장단위의번역을수행하는문장단위번역모듈및 상기입력문에포함된상기문장단위이하의언어패턴의구조를분석하고, 상기번역메모리데이터베이스를참조하여상기분석된언어패턴에대응하는상기부분번역패턴을검색하고, 검색된부분번역패턴을조합하여상기번역문을출력하는부분조합번역모듈을포함한다.
-
-
-
-
-
-
-
-
-
Search Results
194
records
(took 0.024 seconds)
