-
公开(公告)号：KR101134467B1
公开(公告)日：2012-04-13
申请号：KR1020090068268
申请日：2009-07-27
Applicant: 한국전자통신연구원
IPC: G06F17/28 , G10L15/18 , G10L15/183 , G10L15/26
Abstract: 본 발명은 자동 통역 시스템에 적용되는 의미 표현 처리 기법에 관한 것으로, 입력되는 음성에 대응하는 음성 인식 결과의 단어열을 의미 표현으로 변환하고, 변환된 의미 표현을 기 구축된 의미 표현 집합 데이터베이스를 참조하여 검색하며, 검색 결과에 따라 최종 의미 표현을 결정하고, 결정된 최종 의미 표현을 최종 문장으로 생성 및 출력함으로써, 사용자 의도가 반영된 의미 표현을 결정하여 자동 통역 시스템의 성능을 향상시킬 수 있는 것이다.
자동 통역 시스템, 음성 인식, 의미 표현-
公开(公告)号：KR1020110064171A
公开(公告)日：2011-06-15
申请号：KR1020090120657
申请日：2009-12-07
Applicant: 한국전자통신연구원
Abstract: PURPOSE: A sentence pattern classifying device and method of a voice recognition result sentence are provided to offer sentence pattern information about an automatically translated input sentence more accurately. CONSTITUTION: A voice recognition module(10) outputs a text sentence with voice recognition. A semantic module(20) analyzes the morpheme of the text sentence. The semantic module classifies the sentence pattern of a corresponding sentence from a morpheme analysis result. According to a sentence pattern classification result, the semantic module adds punctuation marks to the text sentence. A translator(40) translates the text sentence. A voice output module(60) outputs a voice composite tone corresponding to the translated sentence.
Abstract translation: 目的：提供语音识别结果句子的句型分类装置和方法，以更准确地提供关于自动翻译的输入句子的句型信息。 构成：语音识别模块（10）输出具有语音识别的文本句子。 语义模块（20）分析文本句子的语素。 语义模块从语素分析结果对相应句子的句型进行分类。 根据句型分类结果，语义模块将标点符号添加到文本句子中。 翻译者（40）翻译文本句子。 语音输出模块（60）输出与翻译后的句子相对应的语音合成音。
-
公开(公告)号：KR1020110010916A
公开(公告)日：2011-02-08
申请号：KR1020090068268
申请日：2009-07-27
Applicant: 한국전자통신연구원
IPC: G06F17/28 , G10L15/18 , G10L15/183 , G10L15/26
CPC classification number: G10L15/26 , G06F17/2809 , G10L15/1822
Abstract: PURPOSE: A meaning representation processing device and a method thereof are provided to interpret a sentence which is available for being generated in a constructed meaning expression group and train an auto translating unit. CONSTITUTION: A meaning expression converting unit(202a) converts a word column of an input voice recognition result in a meaning expression. By referring a previously constructed meaning expression group database(206a), a meaning expression matching unit(204a) searches the converted meaning expression. According to the search result, a meaning expression matching unit determines a final meaning expression. A sentence generating unit(208a) generates the generated final meaning expression into a final sentence.
Abstract translation: 目的：提供一种意义表示处理设备及其方法来解释可在构造的意义表达组中生成的句子并训练自动翻译单元。 构成：意义表达转换单元（202a）将意义表达式中的输入语音识别结果的单词列转换。 通过引用先前构造的意义表达组数据库（206a），意义表达匹配单元（204a）搜索转换的含义表达式。 根据搜索结果，意义表达匹配单元确定最终含义表达。 句子生成单元（208a）将生成的最终含义表达式生成到最终句子中。
-
公开(公告)号：KR1020100048842A
公开(公告)日：2010-05-11
申请号：KR1020090021205
申请日：2009-03-12
Applicant: 한국전자통신연구원
CPC classification number: G06F17/2715
Abstract: PURPOSE: A method and an apparatus for automatic classification of sentence corpus are provided to facilitate the guarantee of sentence corpus for each area by re-classifying sentence corpuses of various regions into detailed areas. CONSTITUTION: Au auto area classification apparatus includes a sentence learning unit(100) and a sentence classification unit(200). The sentence learning unit learns training set sentence corpus for each detail area to be classified, and extracts N-gram information of detailed sections. The sentence learning unit converts the N-gram information into weight between the detailed sections, and classifies a test set or classified target sentence corpus to be classified according to the weight.
Abstract translation: 目的：提供一种自动分类句子语料库的方法和装置，通过将各区域的句子语料库重新分类为详细区域，便于保证每个区域的句子语料库。 构成：Au自动区域分类装置包括句子学习单元（100）和句子分类单元（200）。 句子学习单元对每个要分类的细节区域学习训练集句子语料库，并提取详细章节的N-gram信息。 句子学习单元将N-gram信息转换成详细部分之间的权重，并根据权重将测试集或分类目标句子语料进行分类。
-
公开(公告)号：KR101295642B1
公开(公告)日：2013-08-13
申请号：KR1020090120657
申请日：2009-12-07
Applicant: 한국전자통신연구원
Abstract: 본 발명은 음성인식결과 문장에 대한 문형분류장치 및 방법에 관한 것으로, 화자의 발성에 의해 입력된 음성을 인식하고, 음성 인식 결과에 의한 텍스트 문장을 대상으로 형태소를 분석하여, 형태소 분석 결과로부터 해당 문장의 문형을 분류하고, 문형 분류 결과에 따라 해당 문장의 말미에 문장 부호를 추가하는 의미분석모듈을 포함하며, 문장 부호가 추가된 문장을 타겟 언어로 번역하여 그에 대응하는 음성합성음을 출력한다. 본 발명에 따르면, 전체 자동통역 과정 중에서, 음성인식결과 및 이를 입력으로 하는 자동번역에 이르는 과정에서 음성인식결과에 의한 텍스트 문장의 문형을 분류하기 위한 것으로, 음성인식결과의 문형을 식별해 줌으로써, 자동번역 입력문의 문형정보를 보다 정확히 제공하게 되며, 궁극적으로 소스언어를 타겟언어로 자동통역하는 것이 원활히 이루어지게 되는 효과가 있다.
-
Patent Agency Ranking
公开(公告)号：KR1020120045582A
公开(公告)日：2012-05-09
申请号：KR1020100107205
申请日：2010-10-29
Applicant: 한국전자통신연구원
IPC: G10L15/14
CPC classification number: G10L15/144 , G10L15/285
Abstract: PURPOSE: A sound model generating apparatus and a method thereof are provided to automatically search for a penalty value about complexity of a sound model of an MDL(Minimum Description Length) standard. CONSTITUTION: A binary tree generating unit(101) generates a binary tree by repetition of Gaussian components in an HMM(Hidden Markov Model) state based on distance standards. An information generating unit(102) generates the maximum scale information of the sound model according to a platform(111) including a sound recognition unit(112). A binary tree reduction unit(103) reduces the binary tree according to the maximum scale information of the sound model.
Abstract translation: 目的：提供一种声音模型生成装置及其方法，以自动搜索关于MDL（最小描述长度）标准的声音模型的复杂度的惩罚值。 构成：二叉树生成单元（101）通过基于距离标准在HMM（隐马尔可夫模型）状态中重复高斯分量来生成二叉树。 信息生成单元（102）根据包括声音识别单元（112）的平台（111）生成声音模型的最大比例信息。 二叉树缩小单元（103）根据声音模型的最大比例信息来减少二叉树。
-
公开(公告)号：KR1020110024075A
公开(公告)日：2011-03-09
申请号：KR1020090081936
申请日：2009-09-01
Applicant: 한국전자통신연구원
CPC classification number: G06F17/2755 , G06F17/2785 , G10L15/1815
Abstract: PURPOSE: A pattern database apparatus and method thereof and a voice recognition apparatus are provided to output a corrected voice recognition result by using a pattern-based semantic representation. CONSTITUTION: A pattern database apparatus(100) analyzes syntax after analyzing a morpheme according to a voice recognition result. The pattern database apparatus recognizes and extracts additional information. A volume expression, a meaningless expression, and additional information are changed after performing a class change. A voice recognition result is outputted after producing a sentence.
Abstract translation: 目的：提供一种图案数据库装置及其方法和语音识别装置，以通过使用基于模式的语义表示来输出校正的语音识别结果。 构成：模式数据库装置（100）根据语音识别结果分析语素后分析语法。 模式数据库装置识别并提取附加信息。 执行类更改后，更改卷表达式，无意义表达式和附加信息。 语音识别结果在产生句子后输出。
-
公开(公告)号：KR1020100067727A
公开(公告)日：2010-06-22
申请号：KR1020080126244
申请日：2008-12-12
Applicant: 한국전자통신연구원
Abstract: PURPOSE: A voice recognition unit and a method thereof of a multiple search base for performing a multi-search about the input speech signal of the multiple search base are provided to improve voice recognition performance about the voice signal by using FSN(Finite State Network) mode and N-gram mode. CONSTITUTION: A speech feature extracting block(102) extracts feature data about the inputted voice signal. A language model database(108) stores the FSN language model and N-gram language model. A multi-search block(104) is parallel performed the first voice search and the second voice search. The multiple search block is created in the integration search network. The multiple search block outputs the voice recognition result according to the third voice search.
Abstract translation: 目的：提供一种用于执行关于多个搜索库的输入语音信号的多次搜索的多重搜索基的语音识别单元及其方法，以通过使用FSN（有限状态网络）来改善关于语音信号的语音识别性能， 模式和N-gram模式。 构成：语音特征提取块（102）提取关于输入的语音信号的特征数据。 语言模型数据库（108）存储FSN语言模型和N-gram语言模型。 多搜索块（104）并行执行第一语音搜索和第二语音搜索。 多个搜索块在集成搜索网络中创建。 多重搜索块根据第三语音搜索输出语音识别结果。
-
公开(公告)号：KR1020100066978A
公开(公告)日：2010-06-18
申请号：KR1020080125524
申请日：2008-12-10
Applicant: 한국전자통신연구원
IPC: G10L15/183 , G06F17/28
CPC classification number: G10L15/1815 , G10L15/183
Abstract: PURPOSE: A method and a device for voice recognition using domain ontology are provided to build a domain ontology of voice recognition target and generate voice recognition grammar applying the built domain ontology and recognize voice through the voice recognition grammar, thereby improving performance of voice recognition device. CONSTITUTION: If a voice signal is inputted through a mike, a feature extraction unit extracts specific vector of a frame unit from the voice signal(S401). A sound model unit provides voice model to a voice recognition unit through modeling the signal characteristic of the voice signal(S403). The voice recognition unit performs voice recognition through a voice model, a voice recognition dictionary(S405), and voice recognition grammar(S407)(S409).
Abstract translation: 目的：提供一种使用域本体进行语音识别的方法和装置，构建语音识别目标的领域本体，并通过语音识别语法生成应用建立的域本体并识别语音的语音识别语法，从而提高语音识别设备的性能 。 构成：如果通过麦克风输入语音信号，则特征提取单元从语音信号中提取帧单位的特定向量（S401）。 声音模型单元通过对语音信号的信号特征进行建模来向语音识别单元提供语音模型（S403）。 语音识别单元通过语音模型，语音识别字典（S405）和语音识别语法（S407）进行语音识别（S409）。
-
公开(公告)号：KR101092356B1
公开(公告)日：2011-12-09
申请号：KR1020080131220
申请日：2008-12-22
Applicant: 한국전자통신연구원
Abstract: 본발명은상호정보를이용한형태소품사태깅기술에관한것으로, 텍스트코퍼스와그 텍스트에대한형태소품사태깅된 DB를이용하여형태소간에순서정보를가진상호정보를구축하고, 입력문장에대해형태소해석을한 후, 그결과인형태소리스트들에대해형태소간의상호정보를계산하고, 그결과를기존의형태소품사태깅방법, 즉통계기반태깅방법과조합하여최적의품사태깅결과를출력할수 있도록하는것을특징으로한다. 본발명에의하면, 형태소품사태깅된 DB를이용하여형태소간에순서정보를가진상호정보를구축하고, 입력문장에대해형태소해석리스트에기 구축된상호정보를계산하고, 기존의형태소품사태깅방법과조합함으로써, 형태소품사태깅의정확도를높일수 있다.
-
-
-
-
-
-
-
-
-
Search Results
28
records
(took 0.023 seconds)
