利索-全球专利检索

  1. 登录
  2. 注册

搜索结果

176 条记录 (耗时 0.024 秒)

    LANGUAGE IDENTIFICATION
    1.
    发明授权
    LANGUAGE IDENTIFICATION 有权

    公开(公告)号：EP3147796B1

    公开(公告)日：2018-06-06

    申请号：EP16161095.1

    申请日：2016-03-18

    申请人： Facebook, Inc.

    发明人： HUANG, Fei

    IPC分类号： G06F17/27 ,  G06F17/28 ,  G06Q50/00

    CPC分类号： G06F17/275 ,  G06F17/2854 ,  G06Q10/10 ,  G06Q50/01 ,  G06Q50/10

    摘要： A likely source language of a media item can be identified by attempting an initial language identification of the media item based on intrinsic or extrinsic factors, such as words in the media item and languages known by the media item author. This initial identification can generate a list of most likely source languages with corresponding likelihood factors. Translations can then be performed presuming each of the most likely source languages. The translations can be performed for multiple output languages. Each resulting translation can receive a corresponding score based on a number of factors. The scores can be combined where they have a common source language. These combined scores can be used to weight the previously identified likelihood factors for the source languages of the media item.

    Character input method, program and apparatus
    2.
    发明公开
    Character input method, program and apparatus 审中-公开
    字符输入法，程序和设备

    公开(公告)号：EP2717170A3

    公开(公告)日：2017-06-14

    申请号：EP13177739.3

    申请日：2013-07-24

    申请人： FUJITSU LIMITED

    发明人： Horiguchi, Nao ,  Watanabe, Ena ,  Tanimura, Masashi ,  Ariyama, Kota ,  Ueno, Takeshi ,  Akama, Katsuaki

    IPC分类号： G06F17/22 ,  G06F17/27

    CPC分类号： G09G5/24 ,  G06F3/0237 ,  G06F17/2217 ,  G06F17/2223 ,  G06F17/273 ,  G06F17/275 ,  G06F17/276

    摘要： When a French character is selected while an English character string candidate is presented, a mobile terminal presents English and French character string candidates by using an English/French mixed candidate dictionary. Furthermore, when a French character is selected while English and French character string candidates are presented by using the English/French mixed candidate dictionary, the mobile terminal presents a French character string candidate by using a French candidate dictionary. Moreover, when a character other than French is selected while a French character string candidate is presented by using the French candidate dictionary, the mobile terminal presents an English character string candidate by using an English candidate dictionary. The technique can also be applied to one and two byte Katakana characters.

    摘要翻译： 当在提供英文字符串候选时选择法文字符时，移动终端通过使用英文/法文混合候选字典来呈现英文和法文字符串候选。 此外，当通过使用英语/法语混合候选词典呈现英语和法语字符串候选时选择法语字符时，移动终端通过使用法语候选词典呈现法语字符串候选。 此外，当通过使用法语候选词典呈现法语字符串候选时选择法语以外的字符时，移动终端通过使用英语候选词典呈现英语字符串候选。 该技术也可以应用于一个和两个字节的片假名字符。

    PREDICTING PRONUNCIATION IN SPEECH RECOGNITION
    3.
    发明公开
    PREDICTING PRONUNCIATION IN SPEECH RECOGNITION 审中-公开
    AUSSPRACHEVORHERSAGE IN DER SPRACHERKENNUNG

    公开(公告)号：EP3114679A1

    公开(公告)日：2017-01-11

    申请号：EP15757998.8

    申请日：2015-02-27

    申请人： Amazon Technologies, Inc.

    发明人： ADAMS, Jeffrey Penrod ,  PARLIKAR, Alok Ulhas ,  LILLY, Jeffrey Paul ,  RASTROW, Ariya

    IPC分类号： G10L13/02

    CPC分类号： G10L15/08 ,  G06F17/275 ,  G10L15/187

    摘要： An automatic speech recognition (ASR) device may be configured to predict pronunciations of textual identifiers (for example, song names, etc.) based on predicting one or more languages of origin of the textual identifier. The one or more languages of origin may be determined based on the textual identifier. The pronunciations may include a hybrid pronunciation including a pronunciation in one language, a pronunciation in a second language and a hybrid pronunciation that combines multiple languages. The pronunciations may be added to a lexicon and matched to the content item (e.g., song) and/or textual identifier. The ASR device may receive a spoken utterance from a user requesting the ASR device to access the content item. The ASR device determines whether the spoken utterance matches one of the pronunciations of the content item in the lexicon. The ASR device then accesses the content when the spoken utterance matches one of the potential textual identifier pronunciations.

    摘要翻译： 自动语音识别（ASR）设备可以被配置为基于预测文本标识符的一个或多个原始语言来预测文本标识符的发音（例如，歌曲名称等）。 可以基于文本标识符来确定一个或多个原始语言。 发音可以包括包括一种语言的发音，第二种语言的发音和组合多种语言的混合发音的混合发音。 发音可被添加到词典中并与内容项（例如，歌曲）和/或文本标识符匹配。 ASR设备可以从请求ASR设备的用户接收到语音话语来访问内容项。 ASR设备确定口语话语是否匹配词典中的内容项的发音之一。 ASR设备然后在口语发音与潜在文本标识符发音之一匹配时访问该内容。

    PROCÉDÉ DE DÉTERMINATION D'UN GAP SÉMANTIQUE, DISPOSITIF ET PROGRAMME CORRESPONDANT
    4.
    发明公开
    PROCÉDÉ DE DÉTERMINATION D'UN GAP SÉMANTIQUE, DISPOSITIF ET PROGRAMME CORRESPONDANT 审中-公开
    方法，设备和程序用于确定SEMANTICHEN GAP

    公开(公告)号：EP3079076A1

    公开(公告)日：2016-10-12

    申请号：EP16164332.5

    申请日：2016-04-07

    申请人： Pixalione

    发明人： ALIZON, Fabrice ,  GIARIKANI, Antonia

    IPC分类号： G06F17/27 ,  G06F17/30 ,  G06Q30/02

    CPC分类号： G06F17/2785 ,  G06F17/275 ,  G06F17/30684 ,  G06F17/30737 ,  G06Q30/02

    摘要： L'invention se rapporte à un procédé, mis en oeuvre par l'intermédiaire d'un dispositif électronique, de détermination d'une différence entre un contenu à optimiser (CntO) et un ensemble de contenus de référence (Cnt1,..., CntN), lesdits contenus de référence étant réunis au sein d'un corpus (COR). Un tel procédé comprend :
    la création (20) de champs contextuels (ChTxt0, ChTxt1,...,ChTxtN, ChtTxtX), répartis en un champ contextuel (ChTxt0) du contenu à optimiser (CntO), au moins un champ contextuel (ChTxt1,...,ChTxtN) correspondant à au moins un contenu de référence (Cnt1,..., CntN) et un champ contextuel global (ChtTxtX) obtenu à partir d'au moins un contenu de référence (Cnt1,..., CntN) réuni au sein d'un contenu global, chaque champ contextuel (ChTxt0, ChTxt1,...,ChTxtN, ChtTxtX) comprenant au moins un ensemble de sommets deux-à-deux adjacents dénommé clique, chaque sommet comprenant un mot et chaque relation entre mot étant représentative d'une signification ;
    le calcul (30) d'un gap sémantique (GapS) sur la base des champs contextuels (ChTxt0, ChTxt1,...,ChTxtN, ChtTxtX) créés.

    LANGUAGE LEARNING SYSTEMS AND METHODS
    5.
    发明公开
    LANGUAGE LEARNING SYSTEMS AND METHODS 审中-公开
    语言学习系统和方法

    公开(公告)号：EP2956919A2

    公开(公告)日：2015-12-23

    申请号：EP14751441.8

    申请日：2014-02-14

    申请人： Voxy, Inc.

    发明人： NIELSON, Katharine ,  TYSON, Na-im ,  BREEN, Andrew ,  KIRKHAM, Kasey

    IPC分类号： G09B19/00

    CPC分类号： G09B19/06 ,  A61B5/0002 ,  A61B5/0476 ,  A61B5/048 ,  A61B5/04842 ,  A61B5/1124 ,  A61B5/162 ,  A61B5/165 ,  A61B5/4064 ,  A61B5/4076 ,  A61B5/4088 ,  A61B5/411 ,  A61B5/4842 ,  A61B5/486 ,  G06F17/2705 ,  G06F17/274 ,  G06F17/275 ,  G09B5/00 ,  G09B5/02 ,  G09B7/08

    摘要： Systems, methods, and products for language learning that may extract text from various resources having text, using various natural-language processing features, which can be combined with custom-designed learning activities to offer a needs-based, adaptive learning methodology. The system may receive a resource, extract keywords pedagogically valuable to non-native language learning and academic exercises. Metadata describing various aspects of resources from which keywords are extracted may be associated with keywords. Metadata describing various aspects of keywords may also be associated with keywords. Extracted keywords may be stored into a keyword store along with any metadata associated with keywords.

    摘要翻译： 一种用于语言学习软件的系统，方法和产品，可以使用各种自然语言处理产品功能从语料库中自动提取文本，这些功能可以与定制设计的学习活动相结合，提供基于需求的自适应学习方法。 该系统可以接收文本，提取在非母语学习中对教学有价值的关键词，使用文本的各种语言属性为文本指定难度分数，为文本中的每个关键词生成潜在的分心词列表以在学习活动中实施 并根据基于内容的分类标题对文本进行标记。 然后将此输出与一系列旨在满足学习者语言技能需求的学习活动类型结合使用，然后用于为他们创建动态的自适应活动。

    MULTILINGUAL KEYBOARD
    7.
    发明公开
    MULTILINGUAL KEYBOARD 审中-公开
    多语言键盘

    公开(公告)号：EP2926221A2

    公开(公告)日：2015-10-07

    申请号：EP13859413.0

    申请日：2013-12-02

    申请人： Semeco, Maria Daniela

    发明人： Semeco, Maria Daniela

    IPC分类号： G06F3/023

    CPC分类号： G06F3/0233 ,  G06F3/018 ,  G06F3/04886 ,  G06F17/275 ,  H04M2250/70

    摘要： A diacritic chording system uses a single standard keyboard layout to generate diacritical characters used, for example, by Latin-based languages or languages based on a Roman character set, providing a universal keyboard. The key combination required to select a diacritic is consistent, independent of the language the user is typing. The diacritic chording system uses chording to obtain characters modified by diacritics, i.e., diacritical characters. The key combinations in the diacritic chord are chosen in such a way to aid memorization by positional association of the keys or by logic association of the keys within the diacritic chord. The use of a combination of positional association and logical association eliminates conflicts in mapping diacritics for different languages. The diacritic chording system also provides a screen overlay as a mnemonic to illustrate which diacritic chords provide the desired diacritical character.