公开(公告)号：US20070239634A1

公开(公告)日：2007-10-11

申请号：US11400629

申请日：2006-04-07

申请人： Jilei Tian ,  Jani Nurminen ,  Victor Popa

发明人： Jilei Tian ,  Jani Nurminen ,  Victor Popa

IPC分类号： G06N3/02

CPC分类号： G10L21/00 ,  G10L13/033 ,  G10L2021/0135

摘要： An apparatus for providing efficient evaluation of feature transformation includes a training module and a transformation module. The training module is configured to train a Gaussian mixture model (GMM) using training source data and training target data. The transformation module is in communication with the training module. The transformation module is configured to produce a conversion function in response to the training of the GMM. The training module is further configured to determine a quality of the conversion function prior to use of the conversion function by calculating a trace measurement of the GMM.

摘要翻译： 用于提供特征变换的有效评估的装置包括训练模块和变换模块。 训练模块被配置为使用训练源数据和训练目标数据训练高斯混合模型（GMM）。 变换模块与训练模块通信。 转换模块被配置为响应于GMM的训练而产生转换功能。 训练模块还被配置为通过计算GMM的跟踪测量来确定在使用转换功能之前的转换功能的质量。

公开(公告)号：US20070073541A1

公开(公告)日：2007-03-29

申请号：US11605655

申请日：2006-11-29

申请人： Jilei Tian

发明人： Jilei Tian

IPC分类号： G10L15/04

CPC分类号： G10L15/12 ,  G10L2015/025 ,  H03M7/30

摘要： The invention relates to pre-processing of a pronunciation dictionary for compression in a data processing device, the pronunciation dictionary comprising at least one entry, the entry comprising a sequence of character units and a sequence of phoneme units. According to one aspect of the invention the sequence of character units and the sequence of phoneme units are aligned using a statistical algorithm. The aligned sequence of character units and aligned sequence of phoneme units are interleaved by inserting each phoneme unit at a predetermined location relative to the corresponding character unit.

公开(公告)号：US20070016421A1

公开(公告)日：2007-01-18

申请号：US11180316

申请日：2005-07-12

申请人： Jani Nurminen ,  Hannu Mikkola ,  Jilei Tian

发明人： Jani Nurminen ,  Hannu Mikkola ,  Jilei Tian

IPC分类号： G10L13/08

CPC分类号： G10L13/08

摘要： This invention relates to a method, a device and a software application product for correcting a pronunciation of a speech object. The speech object is synthetically generated from a text object in dependence on a segmented representation of the text object. It is determined if an initial pronunciation of the speech object, which initial pronunciation is associated with an initial segmented representation of the text object, is incorrect. Furthermore, in case it is determined that the initial pronunciation of the speech object is incorrect, a new segmented representation of the text object is determined, which new segmented representation of the text object is associated with a new pronunciation of the speech object.

摘要翻译： 本发明涉及一种用于校正语音对象的发音的方法，装置和软件应用产品。 语音对象根据文本对象的分段表示从文本对象合成生成。 确定初始发音是否与文本对象的初始分段表示相关联的语音对象的初始发音是不正确的。 此外，在确定语音对象的初始发音不正确的情况下，确定文本对象的新的分段表示，文本对象的哪个新的分段表示与语音对象的新发音相关联。

公开(公告)号：US20060229864A1

公开(公告)日：2006-10-12

申请号：US11100991

申请日：2005-04-07

申请人： Janne Suontausta ,  Jilei Tian

发明人： Janne Suontausta ,  Jilei Tian

IPC分类号： G06F17/20

CPC分类号： G06F17/2217 ,  G06F17/275 ,  G10L13/08 ,  G10L15/005 ,  G10L15/187

摘要： A method of multi-lingual speech recognition can include determining whether characters in a word are in a source list of a language-specific alphabet mapping table for a language, converting each character not in the source list according to a general alphabet mapping table, converting each converted character according to the language-specific alphabet mapping table, verifying that each character in the word is in a character set of the language, removing characters not in the character set of the language, and identifying a pronunciation of the word.

摘要翻译： 一种多语言语音识别的方法可以包括确定单词中的字符是否在用于语言的语言特定字母映射表的源列表中，根据一般字母映射表转换不在源列表中的每个字符，转换 每个转换的字符根据语言特定的字母映射表，验证该单词中的每个字符是该语言的字符集，去除不在该语言的字符集中的字符，以及识别该单词的发音。

公开(公告)号：US10404764B2

公开(公告)日：2019-09-03

申请号：US14428231

申请日：2012-09-14

申请人： Sailesh Sathish ,  Jilei Tian

发明人： Sailesh Sathish ,  Jilei Tian

IPC分类号： H04L29/06 ,  G06F16/28 ,  G06Q50/00 ,  H04L29/08

摘要： An approach is provided for constructing dynamic latent models that determine consumer/social network intrinsic properties and automatically recommend user interactions with different social networks. A modeling platform determines one or more social networks associated with one or more users, one or more devices associated with the one or more users, or a combination thereof. A modeling platform processes and/or facilitates a processing of data associated with the one or more social networks to generate one or more latent models describing the one or more social networks. A modeling platform causes, at least in part, a presentation of a recommendation to interact with the one or more social networks, one or more other social networks, or a combination thereof based, at least in part, on the one or more latent models.

公开(公告)号：US09106672B2

公开(公告)日：2015-08-11

申请号：US13519649

申请日：2009-12-31

申请人： Guohua Zhang ,  Jilei Tian ,  Xia Wang ,  Wei Wang ,  Feng Ding ,  Yuezhong Tang ,  Xinxing Yang ,  Ari Aarnio

发明人： Guohua Zhang ,  Jilei Tian ,  Xia Wang ,  Wei Wang ,  Feng Ding ,  Yuezhong Tang ,  Xinxing Yang ,  Ari Aarnio

IPC分类号： H04L29/06 ,  H04W4/00 ,  H04W4/02 ,  H04W4/18

CPC分类号： H04L65/1066 ,  G10L13/00 ,  G10L15/26 ,  G10L25/48 ,  H04L65/1089 ,  H04L65/4007 ,  H04W4/02 ,  H04W4/18 ,  H04W4/50

摘要： An approach is provided for performing multiple and hybrid forms of communication in the same communication session. A communication manager receives a to establish a communication session using a first form of communication, wherein the communication session supports multiple and simultaneous forms of communication. Next, the communication manager selects a second form of communication to conduct the communication session. Then, the communication manager transcodes the second form of communication to the first form of communication. The different forms can be converted to facilitate and enrich the communication capability, according to an embodiment of the invention.

摘要翻译： 提供了一种用于在相同通信会话中执行多种和混合形式的通信的方法。 通信管理器接收用于使用第一通信形式建立通信会话，其中通信会话支持多种并且同时形式的通信。 接下来，通信管理器选择第二形式的通信以进行通信会话。 然后，通信管理器将第二形式的通信转码为第一通信形式。 根据本发明的实施例，可以转换不同的形式以促进和丰富通信能力。

公开(公告)号：US20140222622A1

公开(公告)日：2014-08-07

申请号：US14122403

申请日：2011-05-27

申请人： Nan Du ,  Jilei Tian

发明人： Nan Du ,  Jilei Tian

IPC分类号： G06Q30/06

CPC分类号： G06Q30/0627 ,  G06F16/9535

摘要： An approach is provided for generating one or more recommendations to a user based on interactions the user may have with items or topics of interest. The approach involves processing and/or facilitating a processing of one or more interactions of a user with one or more content items. The approach further involves causing, at least in part, an accumulation of the one or more processed interactions of the user. The approach also involves causing, at least in part, a determination of one or more user preferences based, at least in part, on the accumulated one or more processed interactions. The approach additionally involves causing, at least in part, a generation of a rating score of the user for the topic based, at least in part, on the one or more user preferences.

摘要翻译： 提供了一种基于用户可能具有的与感兴趣的项目或主题的交互来向用户生成一个或多个推荐的方法。 该方法涉及处理和/或促进用户与一个或多个内容项目的一个或多个交互的处理。 该方法还包括至少部分地导致用户的一个或多个经处理的交互的累积。 该方法还包括至少部分地至少部分地基于累积的一个或多个处理的相互作用来导致确定一个或多个用户偏好。 该方法还包括至少部分地至少部分地导致基于一个或多个用户偏好来生成用户对该主题的评分分数。

公开(公告)号：US08751239B2

公开(公告)日：2014-06-10

申请号：US11867196

申请日：2007-10-04

申请人： Jilei Tian ,  Victor Popa ,  Jani K. Nurminen

发明人： Jilei Tian ,  Victor Popa ,  Jani K. Nurminen

IPC分类号： G10L13/00

CPC分类号： G10L15/063 ,  G10L15/26 ,  G10L21/00 ,  G10L2021/0135

摘要： An apparatus for providing text independent voice conversion may include a first voice conversion model and a second voice conversion model. The first voice conversion model may be trained with respect to conversion of training source speech to synthetic speech corresponding to the training source speech. The second voice conversion model may be trained with respect to conversion to training target speech from synthetic speech corresponding to the training target speech. An output of the first voice conversion model may be communicated to the second voice conversion model to process source speech input into the first voice conversion model into target speech corresponding to the source speech as the output of the second voice conversion model.

摘要翻译： 用于提供文本独立语音转换的装置可以包括第一语音转换模型和第二语音转换模型。 关于将训练源语音转换成对应于训练源语音的合成语音，可以对第一语音转换模型进行训练。 关于从对应于训练目标语音的合成语音到训练目标语音的转换，可以对第二语音转换模型进行训练。 第一语音转换模型的输出可以被传送到第二语音转换模型，以将输入到第一语音转换模型的源语音转换为与源语音相对应的目标语音作为第二语音转换模型的输出。

公开(公告)号：US20130254900A1

公开(公告)日：2013-09-26

申请号：US13816118

申请日：2010-08-13

申请人： Sailesh Kumar Sathish ,  Jilei Tian

发明人： Sailesh Kumar Sathish ,  Jilei Tian

IPC分类号： G06F21/62 ,  G06F17/30

CPC分类号： G06F21/62 ,  G06F17/30241 ,  H04L63/107 ,  H04W4/02 ,  H04W4/185

摘要： Techniques to provide a secure, shared personal map layer include determining a geographic location. The geographic location is associated with operation of a device. The techniques also include determining indication that describes a relationship between the geographic location and a first user of the device. The techniques also include determining a privacy level for the indication. Then, the first user of the device is associated with the indication and the geographic location and the privacy level. In some embodiments, the techniques also include determining a personal description vocabulary word based, at least in part, on the geographic location and a context for the device. Then it is determined to present on the device a prompt that includes the personal description vocabulary word.

摘要翻译： 提供安全的共享个人地图层的技术包括确定地理位置。 地理位置与设备的操作相关联。 这些技术还包括确定描述地理位置与设备的第一用户之间的关系的指示。 这些技术还包括确定该指示的隐私级别。 然后，设备的第一个用户与指示和地理位置和隐私级别相关联。 在一些实施例中，技术还包括至少部分地基于地理位置和设备的上下文来确定个人描述词汇单词。 然后确定在设备上呈现包括个人描述词汇单词的提示。

公开(公告)号：US20130013644A1

公开(公告)日：2013-01-10

申请号：US13637001

申请日：2010-03-29

申请人： Sailesh Sathish ,  Jilei Tian ,  Rile Hu

发明人： Sailesh Sathish ,  Jilei Tian ,  Rile Hu

IPC分类号： G06F17/30

CPC分类号： G06F17/30867 ,  G06F17/30734

摘要： Methods and apparatuses are provided for user interest modeling. A method may include receiving an input from a user for specifying one or more topics from among a predetermined hierarchy of topics and subtopics. The method may additionally include retrieving one or more documents associated with the user and extracting language tokens from the documents based, at least in part, on the specified topics. Corresponding apparatuses are also provided.

摘要翻译： 为用户兴趣建模提供了方法和设备。 一种方法可以包括从用户接收用于从主题和子主题的预定层级中指定一个或多个主题的输入。 该方法可以另外包括：至少部分地基于指定的主题来检索与用户相关联的一个或多个文档和从文档中提取语言令牌。 还提供了相应的装置。