-
公开(公告)号：KR1020110071742A
公开(公告)日：2011-06-29
申请号：KR1020090128386
申请日：2009-12-21
Applicant: 한국전자통신연구원
Abstract: PURPOSE: An utterance verification apparatus based on a word reliability threshold and a method thereof are provided to apply different reliability threshold to each word recognized in a word-based utterance verification system with respect to a voice recognition result. CONSTITUTION: A phoneme segment information extractor(130) extracts phoneme segment information with the analysis of a recognized word. Likelihood value calculators(140,150) calculate an likelihood value for the extracted phoneme and half-phoneme. A threshold calculator(170) calculates a threshold value corresponding to the recognized word. A comparator(190) compares the threshold value with an LLR(Log Likelihood Ratio) calculated by the likelihood value calculator. According to a comparison result, the comparator outputs or secludes a voice recognition result.
Abstract translation: 目的：提供一种基于字可靠性阈值的话语验证装置及其方法，以针对语音识别结果对基于词语的话语验证系统中识别的每个单词应用不同的可靠性阈值。 构成：音素段信息提取器（130）通过识别字词的分析来提取音素段信息。 似然值计算器（140,150）计算提取的音素和半音素的似然值。 阈值计算器（170）计算与所识别的字对应的阈值。 比较器（190）将阈值与由似然值计算器计算的LLR（对数似然比）进行比较。 根据比较结果，比较器输出或隐藏语音识别结果。
-
公开(公告)号：KR1020110071738A
公开(公告)日：2011-06-29
申请号：KR1020090128382
申请日：2009-12-21
Applicant: 한국전자통신연구원
Abstract: PURPOSE: A material-distribution search method using a voice recognizing function and a method thereof are provided to search a proper transportation request by searching a distribution in the moving route in real time. CONSTITUTION: The location of a truck is traced and a truck owner selects a freight search mode through a voice(202). If the selected freight search mode is an automatic search mode, the truck owner selects the forward data search or the backward data search through a voice(203,204). If the forward data search is selected, the transportation request information after the current location is provided(205). If the backward data search is selected, the transportation request information prior to the current location is provided(206).
Abstract translation: 目的：提供一种使用语音识别功能的材料分配搜索方法及其方法，用于通过实时搜索移动路线中的分布来搜索适当的运输请求。 宪法：追查卡车的位置，货车所有人通过声音选择货运搜索模式（202）。 如果所选择的货运搜索模式是自动搜索模式，则卡车所有者通过语音选择前向数据搜索或反向数据搜索（203,204）。 如果选择正向数据搜索，则提供当前位置之后的运送请求信息（205）。 如果选择了反向数据搜索，则提供当前位置之前的运输请求信息（206）。
-
公开(公告)号：KR1020110071297A
公开(公告)日：2011-06-29
申请号：KR1020090127819
申请日：2009-12-21
Applicant: 한국전자통신연구원
IPC: G10L15/183 , G10L15/28
Abstract: PURPOSE: A voice recognition apparatus for generating a plurality of recognition results is provided to generate N-best recognition results using a phoneme column-based search unit. CONSTITUTION: A continuous voice recognition unit(101) performs the voice recognition of input voice data. The continuous voice recognition unit outputs a word column which is most similar to the input voice data as a recognition result. A phoneme column converter(102) changes the recognition result into a phoneme column. A phoneme column-based search unit(103) searches a plurality of word columns of which a phoneme column distance is contiguity with the recognition result from a language model(105).
Abstract translation: 目的：提供用于产生多个识别结果的语音识别装置，以使用基于音素列的搜索单元产生N最佳识别结果。 构成：连续语音识别单元（101）执行输入语音数据的语音识别。 连续语音识别单元输出与输入语音数据最相似的字列作为识别结果。 音素列转换器（102）将识别结果改变为音素列。 基于音素列的搜索单元（103）搜索与语言模型（105）的识别结果相邻的音素列距离的多个单词列。
-
44.发明公开
公开(公告)号：KR1020110070688A
公开(公告)日：2011-06-24
申请号：KR1020100033376
申请日：2010-04-12
Applicant: 한국전자통신연구원
Abstract: PURPOSE: A voice recognition apparatus and method having two-step utterance verification structure for reducing the complexity of N-best recognized word calculation are provided to induce the re-utterance of a user or notify the user of an utterance error. CONSTITUTION: Using a first model, a voice recognition module(130) recognizes the voice of input voice data. The voice recognition module outputs a first N-best word list. An utterance verification module(140) creates a second N-best word list. Using a second model, the utterance verification module creates a final N-best word list from the second N-best word list.
Abstract translation: 目的：提供具有用于降低最佳识别字计算的复杂度的两步话语验证结构的语音识别装置和方法，以引起用户的重新发音或通知用户话语错误。 构成：使用第一模型，语音识别模块（130）识别输入语音数据的语音。 语音识别模块输出第一个N最好的单词列表。 话音验证模块（140）创建第二N个最佳词列表。 使用第二个模型，话语验证模块从第二个N最佳词列表中创建一个最后的N个最佳词列表。
-
公开(公告)号：KR1020100073165A
公开(公告)日：2010-07-01
申请号：KR1020080131759
申请日：2008-12-22
Applicant: 한국전자통신연구원
Abstract: PURPOSE: An apparatus and a method for discriminating a speech from music are provided to accurately discriminate a voice from a music for each section by configuring a complex vector value through a similarity test value. CONSTITUTION: A similarity test unit(104) tests voice or music signal similarity for an audio signal by comparing each feature vector value with a voice and music decision model value by feature vector. A complex feature vector extractor(106) constitutes the complex feature vector value based on similarity test value. A voice/music classification unit(108) discriminate a voice or music signal by section from the audio signal by comparing the corresponding voice and music discrimination model with the complex feature vector.
Abstract translation: 目的：提供一种用于区分语音与音乐的装置和方法，以便通过相似性测试值配置复数向量值来准确地区分每个部分的音乐的声音。 构成：相似性测试单元（104）通过将每个特征向量值与特征向量的语音和音乐决策模型值进行比较来测试音频信号的声音或音乐信号相似度。 复杂特征向量提取器（106）基于相似性测试值构成复特征向量值。 语音/音乐分类单元（108）通过将对应的语音和音乐辨别模型与复杂特征向量进行比较，从音频信号逐个区分语音或音乐信号。
-
Patent Agency Ranking
公开(公告)号：KR1020100067174A
公开(公告)日：2010-06-21
申请号：KR1020080125621
申请日：2008-12-11
Applicant: 한국전자통신연구원
CPC classification number: H04N21/6125 , G10L15/07 , H04N21/42203 , H04N21/42204 , H04N21/4334 , H04N21/4394 , H04N21/472 , H04N21/8405
Abstract: PURPOSE: A meta data searcher, a search method and an IPTV receiving apparatus for using the voice recognition of user uttered data by previously extracting an allomorph about a contents title are provided to search the contents which user desires by using information search functions and voice recognition. CONSTITUTION: A metadata processor(250) processes content meta data to inform data. A metadata storage unit stores content meta data. A voice recognizer(230) performs the voice recognition about firing data. A query processor(240) extracts the search word from the voice-recognized vocabulary. A search processor extracts content meta data corresponding to the search word.
Abstract translation: 目的：提供一种元数据搜索器，搜索方法和IPTV接收装置，用于通过预先提取关于内容标题的变形来使用用户发出的数据的语音识别，以通过使用信息搜索功能和语音识别来搜索用户期望的内容 。 构成：元数据处理器（250）处理内容元数据以通知数据。 元数据存储单元存储内容元数据。 语音识别器（230）执行关于触发数据的语音识别。 查询处理器（240）从语音识别的词汇表中提取搜索词。 搜索处理器提取与搜索词相对应的内容元数据。
-
公开(公告)号：KR1020100062824A
公开(公告)日：2010-06-10
申请号：KR1020090026297
申请日：2009-03-27
Applicant: 한국전자통신연구원
Abstract: PURPOSE: An utterance verification apparatus and a method thereof are provided, which can improve the reliability about the voice recognition result by performing the whole utterance verification about the word after the performance of the partial utterance verification about each sub word. CONSTITUTION: A sub word likelihood ratio calculator(204) calculates the sub word likelihood ratio about the divided sub word. A likelihood ratio weight applying unit(206) applies the weighted value of likelihood ratio according to the recognition reliability to the calculated sub word likelihood ratio. A word likelihood ratio calculator(208) calculates the word likelihood ratio by using the sub word likelihood ratio. A speech verification decision unit(210) decides the output of the word after the utterance verification with the calculated word likelihood ratio.
Abstract translation: 目的：提供一种话语验证装置及其方法，通过在关于每个子字的部分话语验证之后执行关于该单词的整个话语验证，可提高语音识别结果的可靠性。 构成：子字似然比计算器（204）计算关于分割子字的子字似然比。 似然比权重赋予单元（206）将识别可靠性的似然比的加权值应用于计算出的子字似然比。 字似然比计算器（208）通过使用子字似然比来计算字似然比。 语音验证判定单元（210）用所计算的字似然比来确定在话语验证之后的单词的输出。
-
公开(公告)号：KR1020070061288A
公开(公告)日：2007-06-13
申请号：KR1020060084813
申请日：2006-09-04
Applicant: 한국전자통신연구원
IPC: A63F13/833 , A63F13/45 , G06Q50/10
CPC classification number: A63F13/833 , A63F13/45 , G06Q50/10
Abstract: A method for creating and controlling AI characters in fighting action games are provided to enable the AI character to autonomously learn a game rule through fighting with a randomly operated character in an initial stage. An input layer inputs action of an opponent character, action steps, a distance between the opponent character and the AI character, and past action of both characters by using an integer. A hidden layer finds an optimal hidden node by using a non-linear model for estimating the action of the opponent character, the action steps, the distance between the opponent character and the AI character, and the past action of both characters. An output layer selects output nodes as many as the number of actions available to the AI. The AI selects the action outputting the biggest value among the available actions.
Abstract translation: 提供了一种在打击动作游戏中创建和控制人物角色的方法，以使得AI角色能够在初始阶段与随机操作角色的战斗自主地学习游戏规则。 输入层输入对手角色的动作，动作步骤，对方角色和AI角色之间的距离，以及通过使用整数的两个角色的过去动作。 隐藏层通过使用非线性模型来估计对手角色的动作，动作步骤，对方角色与AI角色之间的距离以及两个角色的过去动作，找到最佳隐藏节点。 输出层选择与AI可用动作数相同的输出节点。 AI选择可用动作中输出最大值的动作。
-
公开(公告)号：KR101905827B1
公开(公告)日：2018-10-08
申请号：KR1020130073990
申请日：2013-06-26
Applicant: 한국전자통신연구원
Abstract: 본발명은대용량어휘가포함된연속어에대하여음성인식을수행하는장치및 방법에관한것이다. 본발명은동일종류의어휘의수가매우많은대어휘연속어에서대어휘를적당한개수의클러스터로나누고, 해당클러스터의대표어휘를선정하여대표어휘만포함된 1단계인식을수행하며, 1단계수행한결과를이용하여대표어휘가인식되었을경우인식된대표어휘가속한클러스터내의모든단어에대하여재인식을수행한다.
-
公开(公告)号：KR1020150001189A
公开(公告)日：2015-01-06
申请号：KR1020130073987
申请日：2013-06-26
Applicant: 한국전자통신연구원
Abstract: 본 발명은 음성인식을 이용한 외국어 말하기 능력 훈련 및 평가 방법에 관한 것으로, 보다 상세하게는 컴퓨터 또는 기타 측정장치를 이용하여 외국어 말하기 능력을 사용자 스스로 훈련하고 평가하는 방법에 관한 것이다.
본 발명의 음성인식을 이용한 외국어 말하기 능력 훈련 및 평가 방법은 제1사용자로부터 발성된 외국어 음성신호를 입력 받는 단계, 입력 받은 음성신호를 레코딩하는 단계, 레코딩된 음성신호를 재생하여 제2사용자에게 제공하는 단계, 제공된 음성신호에 대응되는 텍스트 데이터를 제2사용자로부터 입력 받는 단계, 입력 받은 텍스트 데이터와 상기 레코딩된 음성신호를 비교하여 정확도를 측정하는 단계, 및 측정된 정확도에 따른 평가결과를 상기 제1사용자에게 제공하는 단계를 포함한다.Abstract translation: 本发明涉及使用语音识别的外语能力的训练和评估方法，更具体地，涉及一种允许用户通过使用计算机亲自训练和评估外语能力的方法， 其他测量设备。 根据本发明，使用语音识别的使用外语的能力的训练和评估方法包括以下步骤：接收从第一用户发音的外语的语音信号; 记录已经接收到的语音信号; 再生已经记录的语音信号并向第二用户提供语音信号; 接收与从第二用户提供的语音信号对应的文本数据; 比较已经接收的文本数据和已记录的语音信号来测量精度; 并根据测量的精度向第一用户提供评估结果。
-
-
-
-
-
-
-
-
-
Search Results
95
records
(took 0.025 seconds)
