-
公开(公告)号：KR1020090034578A
公开(公告)日：2009-04-08
申请号：KR1020070099887
申请日：2007-10-04
Applicant: 포항공과대학교 산학협력단
IPC: G06F17/20
CPC classification number: G06F17/2827 , G06F17/2818 , G10L2015/025
Abstract: A phoneme series-utterance sequence conversion method and an apparatus thereof in which the morpheme for the Korean text-to-speech system is provided to a phoneme series-utterance sequence transform system by using a sampling method. A training part(110) comprises phoneme series arranged with a phoneme series-utterance sequence sorter(111) and a translation rule extracting unit(112) extracting regulation from the utterance sequence. The phoneme series-utterance sequence sorter arranges the phoneme series and utterance sequence of the language corpus. The translation rule extracting unit extracts regulation from the phoneme series and the arranged utterance sequence. A transform unit(120) comprises a rule application part(121). A conversion rule tree(130) uses the enunciation generating of the transform unit. When the phoneme series and the given utterance sequence have the phoneme series-utterance sequence sorter, it determines whether any kind of phoneme is corresponded to any kind of utterance.
Abstract translation: 一种音素系列发声序列转换方法及其装置，其中通过使用采样方法将韩文文本到语音系统的语素提供给音素系列发声序列变换系统。 训练部（110）包括配置有音素系列话音序列分类器（111）的音素系列和从发音顺序提取音调的翻译规则提取单元（112）。 音素系列语音序列分拣机安排语言语料库的音素系列和语音序列。 翻译规则提取单元从音素系列中提取规则和排列的话语序列。 变换单元（120）包括规则应用部分（121）。 转换规则树（130）使用变换单元的发音生成。 当音素系列和给定的语音序列具有音素系列 - 话音序列分选器时，它确定任何种类的音素是否对应于任何种类的语音。
-
62.发明公开
公开(公告)号：KR1020080069077A
公开(公告)日：2008-07-25
申请号：KR1020070006707
申请日：2007-01-22
Applicant: 포항공과대학교 산학협력단
IPC: G06F17/28 , G10L15/187
CPC classification number: G06F17/2818 , G06F17/289 , G10L15/187
Abstract: An automatic speech interpretation system based on a statistical automatic interpretation mode, and interpretation processing and training methods thereof are provided to increase performance by performing pre treatment, post treatment, and rearrangement when a statistical automatic interpreter is applied. A statistical automatic interpreter(40) interprets a target language in an interpreted language by receiving a target language model(s6) and an interpretation model(s4). A pre-interpretation processing module(02) determines the final voice recognition and interpretation sentences by transferring a plurality of candidate voice recognition sentences to the statistical automatic interpreter without determining the voice recognition sentence after execution of a voice recognizer, and considering voice recognition and interpretation scores. A post-interpretation module(03) outputs the final interpretation result by arranging the candidate voice recognition sentences in order of a sum of the voice recognition and interpretation scores.
Abstract translation: 提供了一种基于统计自动解释模式的自动语音解释系统及其解释处理和训练方法，以在应用统计自动解释器时通过执行预处理，后处理和重新排列来提高性能。 统计自动解释器（40）通过接收目标语言模型（s6）和解释模型（s4）来解释解释语言中的目标语言。 预解释处理模块（02）通过将多个候选语音识别语句转移到统计自动解释器来确定最终语音识别和解释句子，而不需要在执行语音识别器之后确定语音识别语句，并且考虑语音识别和解释 分数。 后解读模块（03）通过按照语音识别和解释得分之和的顺序排列候选语音识别语句来输出最终解释结果。
-
公开(公告)号：KR1020080026782A
公开(公告)日：2008-03-26
申请号：KR1020060091790
申请日：2006-09-21
Applicant: 학교법인 포항공과대학교 , 포항공과대학교 산학협력단
Abstract: A method and an apparatus for comprehending spoken words by using an information extraction method are provided to comprehend essential elements selectively in comprehending the spoken words on the basis of a meaning structure suitable for each specific domain, thereby improving a degree of comprehension for the spoken words. A method for comprehending spoken words by using an information extraction method comprises the following steps of: standardizing the meaning structure of the spoken words previously(210); embodying the standardized meaning structure to be suitable for a specific domain(220); inputting spoken words recognized through a voice recognition unit(230); performing the natural language processing of the inputted spoken words(240); selecting quality with a specific meaning for determining the meaning structure by a result analyzed through the natural language processing(250); performing mechanical studying by using the selected quality(260); and comprehending the spoken words based on the meaning structure formed by determining corresponding elements configuring the meaning structure through the mechanical studying(270).
Abstract translation: 提供了一种通过使用信息提取方法来理解口语的方法和装置，用于在基于适合于每个特定领域的意义结构的基础上理解说出的单词的选择性的基本元素，从而提高口语单词的理解程度 。 一种通过使用信息提取方法来理解口语的方法包括以下步骤：对先前语音词汇的含义结构进行标准化（210）; 体现了适用于特定领域的标准化意义结构（220）; 输入通过语音识别单元（230）识别的口语单词; 对所输入的口语（240）进行自然语言处理; 通过自然语言处理（250）分析结果，选择具有意义的质量来确定意义结构; 使用所选质量进行机械学习（260）; 并基于通过机械学习确定构成意义结构的相应元素形成的意义结构来理解口语（270）。
-
公开(公告)号：KR1020080024752A
公开(公告)日：2008-03-19
申请号：KR1020060089264
申请日：2006-09-14
Applicant: 학교법인 포항공과대학교 , 포항공과대학교 산학협력단
Abstract: A device and a method for managing dialog of a chatting agent are provided to improve performance of a chatting robot continuously by providing an answer to the chatting robot based on dialog stacks in consideration of dialog context. A dialog analyzer(120) analyzes a speech sound(110) received from a user. A domain determiner(130) determines and verifies domains to determine the domain of the speech sound based on a speech sound analysis result. A dialog expert(140) provides a system speech sound(170) in response to the user speech sound by searching a dialog example database constructed from a dialog corpus. The domain determiner includes a language/semantic quality extractor(121,122) and a keyword extractor(123), and determines the domain by integrating and applying extracted qualities to a quality based probability model. The dialog expert includes a dialog stack(150) storing dialog example session information of the previous user speech sound and dialog/main flow analysis result, and a dialog example selector(160) indexing information such as semantic information of the previous user speech sound and scenario session information added from a scenario based dialog corpus.
Abstract translation: 提供了一种用于管理聊天代理的对话的设备和方法，以通过基于对话框叠加提供对聊天机器人的回答来连续地提高聊天机器人的性能。 对话分析器（120）分析从用户接收的语音（110）。 域确定器（130）基于语音分析结果来确定和验证域以确定语音的域。 对话专家（140）通过搜索由对话语料库构建的对话示例数据库来响应于用户语音声音提供系统语音（170）。 域确定器包括语言/语义质量提取器（121,122）和关键字提取器（123），并且通过将提取的质量整合并应用于基于质量的概率模型来确定域。 对话专家包括存储先前用户语音和对话/主流分析结果的对话示例会话信息的对话框（150），以及对话例示选择器（160）索引诸如先前用户语音的语义信息的信息，以及 从基于场景的对话框语料库添加的场景会话信息。
-
公开(公告)号：KR1020070102267A
公开(公告)日：2007-10-18
申请号：KR1020060034187
申请日：2006-04-14
Applicant: 학교법인 포항공과대학교 , 포항공과대학교 산학협력단
Abstract: A system and a method for dialog management are provided to improve the responding accuracy of the system by using a dialog example extracted from dialog corpus so as to predict the response of the system. An input unit(113) receives voice or text from user speech. A language comprehension unit(115) extracts a meaning frame of the user speech by using character string information and morpheme analysis result from a character string signal inputted from the input unit(113). An agent determination unit(117) analyzes the meaning frame of the user speech to determine whether the user speech is for simple chatting or purposeful dialog based on a trained possibility model using keyword feature extracted from dialog corpus, language analysis feature, and meaning analysis feature. A domain determination unit(118) determines the domain type of the user speech. A dialog example selection unit(120) selects the most appropriate dialog example by constructing dialog example database from the dialog corpus and searching for dialog examples using the meaning frame and dialog history. A response generation unit(122) generates device speech by using the selected dialog example.
Abstract translation: 提供了一种用于对话管理的系统和方法，通过使用从对话语料库提取的对话示例来提高系统的响应精度，以预测系统的响应。 输入单元（113）从用户语音接收语音或文本。 语言理解单元（115）通过使用从输入单元（113）输入的字符串信号的字符串信息和语素分析结果来提取用户语音的含义帧。 代理确定单元（117）分析用户语音的含义帧，以基于使用从对话语料库，语言分析特征和意义分析特征提取的关键词特征的训练可能性模型来确定用户语音是用于简单聊天还是有目的的对话 。 域确定单元（118）确定用户语音的域类型。 对话示例选择单元（120）通过从对话语料库构建对话示例数据库并使用含义框和对话历史搜索对话示例来选择最合适的对话示例。 响应生成单元（122）通过使用所选择的对话示例来生成设备语音。
-
Patent Agency Ranking
公开(公告)号：KR101751113B1
公开(公告)日：2017-06-27
申请号：KR1020150083855
申请日：2015-06-15
Applicant: 포항공과대학교 산학협력단
Abstract: 다중사용자기반의대화관리방법및 장치가개시된다. 본발명의일 실시예에따른다중사용자기반의대화관리방법은다중사용자로부터입력문장을수신하는단계, 입력문장을분석하여사용자별로저장함으로써대화이력데이터베이스를구축하는단계, 입력문장으로부터단위지식을추출하고, 다중사용자를식별할수 있는사용자정보의유무를기반으로단위지식을선별하여온톨로지를구축하는단계및 대화이력데이터베이스및 온톨로지를기반으로미리추출된시스템응답후보를수정하여입력문장에적합한시스템응답을생성하는단계를포함한다. 따라서, 사용자각각에게맞춤형대화서비스를제공할수 있다.
Abstract translation: 公开了一种用于基于多用户的会话管理的方法和设备。 一个实施方案中，根据的方法基于本发明的对话框的用户的管理包括：分析从多用户，输入句子通过存储为每个用户，从一个单元知识提取输入句子建立对话历史数据库接收该输入句子的步骤的步骤 并确定和通过修改阶段和通话历史数据库和本体先前提取基于所述系统响应候选建立在系统反应适合于多个用户的输入句子的本体被选择用于单元知识，无论是基于所述用户信息 和生成的步骤。 因此，可以为每个用户提供定制的对话服务。
-
公开(公告)号：KR101661669B1
公开(公告)日：2016-09-30
申请号：KR1020150001398
申请日：2015-01-06
Applicant: 포항공과대학교 산학협력단
Abstract: 대화시스템및 그방법에관한것이다. 대화시스템은, 음성신호를수신하며, 상기음성신호로부터제1자연어를인식하는음성인식기, 상기제1자연어로부터사용자발화의도를획득하는발화의도검출모듈, 카메라를통해획득되는이미지로부터시각정보를추출하는시각정보추출모듈, 상기사용자발화의도에따른외부응용데이터베이스선택정보와, 상기사용자발화의도및 상기시각정보에따라서응답정보를추출하기위한쿼리문을포함하는스마트미러모델을저장하는제1데이터베이스, 상기스마트미러모델을토대로, 상기사용자발화의도에대응하는외부응용데이터베이스를선택하고, 상기사용자발화의도에따라선택된외부응용데이터베이스로부터상기사용자발화의도및 상기시각정보에대응하는상기응답정보를추출하는스마트미러제어기, 그리고상기이미지에상기응답정보에대응하는응답이미지를중첩하여출력하는디스플레이를포함할수 있다.
-
公开(公告)号：KR1020160084748A
公开(公告)日：2016-07-14
申请号：KR1020150001398
申请日：2015-01-06
Applicant: 포항공과대학교 산학협력단
Abstract: 대화시스템및 그방법에관한것이다. 대화시스템은, 음성신호를수신하며, 상기음성신호로부터제1자연어를인식하는음성인식기, 상기제1자연어로부터사용자발화의도를획득하는발화의도검출모듈, 카메라를통해획득되는이미지로부터시각정보를추출하는시각정보추출모듈, 상기사용자발화의도에따른외부응용데이터베이스선택정보와, 상기사용자발화의도및 상기시각정보에따라서응답정보를추출하기위한쿼리문을포함하는스마트미러모델을저장하는제1데이터베이스, 상기스마트미러모델을토대로, 상기사용자발화의도에대응하는외부응용데이터베이스를선택하고, 상기사용자발화의도에따라선택된외부응용데이터베이스로부터상기사용자발화의도및 상기시각정보에대응하는상기응답정보를추출하는스마트미러제어기, 그리고상기이미지에상기응답정보에대응하는응답이미지를중첩하여출력하는디스플레이를포함할수 있다.
Abstract translation: 本发明涉及一种能够使用户以固定姿势进行操作的对话系统及其方法。 对话系统可以包括：语音识别器，其接收语音信号，并从语音信号识别第一自然语言; 语音意图检测模块，其从第一自然语言获取用户的语音意图; 视觉信息提取模块，其从由相机获取的图像中提取视觉信息; 存储智能镜模型的第一数据库，其包括基于用户的语音意图提取外部应用数据库选择信息的查询语句和基于用户语音意图和视觉信息的响应信息; 智能镜控制器，其基于智能镜模型，选择与用户的语音意图对应的外部应用数据库，并从所选择的外部应用数据库中提取与用户的语音意图相对应的响应信息和可视信息 基于用户的言语意图; 以及显示器，其输出图像和与响应图像相对应的响应图像，其中响应图像与图像重叠。
-
公开(公告)号：KR101633556B1
公开(公告)日：2016-06-24
申请号：KR1020140125944
申请日：2014-09-22
Applicant: 포항공과대학교 산학협력단
IPC: G06F17/27
Abstract: 문법오류수정장치및 이를이용한문법오류수정방법이개시된다. 입력텍스트에대해문법오류말뭉치에기반하여오류후보단어와오류후보군을생성하는오류후보생성부; 상기오류후보단어와상기오류후보군을바탕으로문맥프레임집합을추출하는라우터; 상기문맥프레임집합에대해 n-gram 점수를산출하는점수산출부; 및상기 n-gram 점수를이용하여상기입력텍스트를수정하는단어수정부를구성한다. 따라서, 사용자로부터제공받은입력텍스트에서문법오류에대한문맥프레임집합별로 n-gram 점수를비교하므로외국어학습자의문법오류를효율적으로수정할수 있고, 입력텍스트에포함된문법오류를정확하게수정함으로써외국어학습을효과적으로수행할수 있다.
-
公开(公告)号：KR1020160046572A
公开(公告)日：2016-04-29
申请号：KR1020140142664
申请日：2014-10-21
Applicant: 포항공과대학교 산학협력단
IPC: G06F17/30
CPC classification number: G06F17/30946
Abstract: 데이터베이스의데이터확장방법및 장치가개시된다. 본발명에따른데이터확장방법은질의문장을수신하는단계, 질의문장을구성하는복수의단어들을포함하는원시트리플을생성하는단계, 원시트리플에포함된복수의단어들각각이가지는복수의의미들중 하나의의미를가지도록원시트리플을명확화하는단계, 명확화된복수의단어들을포함하는최종트리플을생성하는단계및 생성된최종트리플을데이터베이스에저장하는단계를포함한다. 본발명에의하면데이터베이스에보다많은데이터를저장함으로써, 데이터베이스를사용하는질의응답시스템에방대한양의데이터를제공가능하다.
Abstract translation: 公开了一种用于扩展数据库数据的方法和装置。 根据本发明，一种数据扩展方法包括：接收查询短语的步骤; 生成包括构成查询短语的多个项的原始三元组的步骤; 澄清每个术语的一个步骤，使得包含在原始三元组中的术语具有多个含义中的一个含义; 产生包括消歧词的最终三元组的步骤; 并在数据库中保存最后三元组的步骤。 根据本发明，通过在数据库中保存更多数据，可以向使用数据库的查询响应系统提供大量的数据。
-
-
-
-
-
-
-
-
-
Search Results
147
records
(took 0.025 seconds)
