公开(公告)号：US08355919B2

公开(公告)日：2013-01-15

申请号：US12240449

申请日：2008-09-29

申请人： Kim Silverman ,  Devang Naik ,  Jerome Bellegarda ,  Kevin Lenzo

发明人： Kim Silverman ,  Devang Naik ,  Jerome Bellegarda ,  Kevin Lenzo

IPC分类号： G10L13/08

CPC分类号： G10L13/08

摘要： Algorithms for synthesizing speech used to identify media assets are provided. Speech may be selectively synthesized form text strings associated with media assets. A text string may be normalized and its native language determined for obtaining a target phoneme for providing human-sounding speech in a language (e.g., dialect or accent) that is familiar to a user. The algorithms may be implemented on a system including several dedicated render engines. The system may be part of a back end coupled to a front end including storage for media assets and associated synthesized speech, and a request processor for receiving and processing requests that result in providing the synthesized speech. The front end may communicate media assets and associated synthesized speech content over a network to host devices coupled to portable electronic devices on which the media assets and synthesized speech are played back.

摘要翻译： 提供了用于合成用于识别媒体资产的语音的算法。 可以从与媒体资产相关联的文本串选择性地合成语音。 文本字符串可以被归一化，并且其母语被确定用于获得目标音素，以便以用户熟悉的语言（例如，方言或重音）提供人声音语音。 算法可以在包括几个专用渲染引擎的系统上实现。 该系统可以是耦合到前端的后端的一部分，包括用于媒体资产和相关联的合成语音的存储器，以及用于接收和处理导致提供合成语音的请求的请求处理器。 前端可以通过网络将媒体资产和相关联的合成语音内容通信到主机耦合到其上播放媒体资产和合成语音的便携式电子设备的设备。

公开(公告)号：US20100082344A1

公开(公告)日：2010-04-01

申请号：US12240437

申请日：2008-09-29

申请人： Devang Naik ,  Kim Silverman ,  Jerome Bellegarda

发明人： Devang Naik ,  Kim Silverman ,  Jerome Bellegarda

IPC分类号： G10L13/00

CPC分类号： G10L13/033

摘要： Algorithms for synthesizing speech used to identify media assets are provided. Speech may be selectively synthesized form text strings associated with media assets. A text string may be normalized and its native language determined for obtaining a target phoneme for providing human-sounding speech in a language (e.g., dialect or accent) that is familiar to a user. The algorithms may be implemented on a system including several dedicated render engines. The system may be part of a back end coupled to a front end including storage for media assets and associated synthesized speech, and a request processor for receiving and processing requests that result in providing the synthesized speech. The front end may communicate media assets and associated synthesized speech content over a network to host devices coupled to portable electronic devices on which the media assets and synthesized speech are played back.

摘要翻译： 提供了用于合成用于识别媒体资产的语音的算法。 可以从与媒体资产相关联的文本串选择性地合成语音。 文本字符串可以被归一化，并且其母语被确定用于获得目标音素，以便以用户熟悉的语言（例如，方言或重音）提供人声音语音。 算法可以在包括几个专用渲染引擎的系统上实现。 该系统可以是耦合到前端的后端的一部分，包括用于媒体资产和相关联的合成语音的存储器，以及用于接收和处理导致提供合成语音的请求的请求处理器。 前端可以通过网络将媒体资产和相关联的合成语音内容通信到主机耦合到其上播放媒体资产和合成语音的便携式电子设备的设备。

公开(公告)号：US20060168150A1

公开(公告)日：2006-07-27

申请号：US11369480

申请日：2006-03-06

申请人： Devang Naik ,  Kim Silverman ,  Guy Tribble

发明人： Devang Naik ,  Kim Silverman ,  Guy Tribble

IPC分类号： G06F15/16

CPC分类号： H04L65/607 ,  H04L29/06027 ,  H04L67/02

摘要： Improved techniques for providing supplementary media for media items are disclosed. The media items are typically fixed media items. The supplementary media is one or more of audio, video, image, or text that is provided by a user to supplement (e.g., personalize, customize, annotate, etc.) the fixed media items. In one embodiment, the supplementary media can be provided by user interaction with an on-line media store where media items can be browsed, searched, purchased and/or acquired via a computer network. In another embodiment, the supplementary media can be generated on a playback device.

摘要翻译： 公开了用于为媒体项目提供辅助媒体的改进技术。 媒体项目通常是固定的媒体项目。 辅助媒体是由用户提供以补充（例如，个性化，定制，注释等）固定媒体项目的音频，视频，图像或文本中的一个或多个。 在一个实施例中，可以通过与在线媒体商店的用户交互来提供补充媒体，其中可以经由计算机网络浏览，搜索，购买和/或获取媒体项目。 在另一个实施例中，补充媒体可以在播放设备上产生。

公开(公告)号：US5839103A

公开(公告)日：1998-11-17

申请号：US479012

申请日：1995-06-07

申请人： Richard J. Mammone ,  Kevin Farrell ,  Manish Sharma ,  Devang Naik ,  Xiaoyu Zhang ,  Khaled Assaleh ,  Han-Seng Liou

发明人： Richard J. Mammone ,  Kevin Farrell ,  Manish Sharma ,  Devang Naik ,  Xiaoyu Zhang ,  Khaled Assaleh ,  Han-Seng Liou

IPC分类号： G10L15/10 ,  G10L15/02 ,  G10L15/08 ,  G10L15/20 ,  G10L17/00 ,  G10L21/02 ,  G10L5/00

CPC分类号： G10L17/10 ,  G10L17/20 ,  G10L17/04 ,  G10L17/14 ,  G10L17/18 ,  G10L25/03

摘要： The present invention relates to a pattern recognition system which uses data fusion to combine data from a plurality of extracted features and a plurality of classifiers. Speaker patterns can be accurately verified with the combination of discriminant based and distortion based classifiers. A novel approach using a training set of a "leave one out" data can be used for training the system with a reduced data set. Extracted features can be improved with a pole filtered method for reducing channel effects and an affine transformation for improving the correlation between training and testing data.

摘要翻译： 本发明涉及使用数据融合来组合来自多个提取的特征的数据和多个分类器的模式识别系统。 可以通过基于判别式和基于失真的分类器的组合来准确地验证扬声器模式。 使用训练集“离开一个”数据的新颖方法可用于使用减少的数据集训练系统。 提取的特征可以通过用于减少信道效应的极点滤波方法和用于改善训练和测试数据之间的相关性的仿射变换来改进。

公开(公告)号：US08352272B2

公开(公告)日：2013-01-08

申请号：US12240404

申请日：2008-09-29

申请人： Matthew Rogers ,  Kim Silverman ,  Devang Naik ,  Kevin Lenzo ,  Benjamin Rottler

发明人： Matthew Rogers ,  Kim Silverman ,  Devang Naik ,  Kevin Lenzo ,  Benjamin Rottler

IPC分类号： G10L13/08

CPC分类号： G10L13/00

摘要： Algorithms for synthesizing speech used to identify media assets are provided. Speech may be selectively synthesized form text strings associated with media assets. A text string may be normalized and its native language determined for obtaining a target phoneme for providing human-sounding speech in a language (e.g., dialect or accent) that is familiar to a user. The algorithms may be implemented on a system including several dedicated render engines. The system may be part of a back end coupled to a front end including storage for media assets and associated synthesized speech, and a request processor for receiving and processing requests that result in providing the synthesized speech. The front end may communicate media assets and associated synthesized speech content over a network to host devices coupled to portable electronic devices on which the media assets and synthesized speech are played back.

摘要翻译： 提供了用于合成用于识别媒体资产的语音的算法。 可以从与媒体资产相关联的文本串选择性地合成语音。 文本字符串可以被归一化，并且其母语被确定用于获得目标音素，以便以用户熟悉的语言（例如，方言或重音）提供人声音语音。 算法可以在包括几个专用渲染引擎的系统上实现。 该系统可以是耦合到前端的后端的一部分，包括用于媒体资产和相关联的合成语音的存储器，以及用于接收和处理导致提供合成语音的请求的请求处理器。 前端可以通过网络将媒体资产和相关联的合成语音内容通信到主机耦合到其上播放媒体资产和合成语音的便携式电子设备的设备。

公开(公告)号：US20070106742A1

公开(公告)日：2007-05-10

申请号：US11643304

申请日：2006-12-20

申请人： Jerome Bellegarda ,  Devang Naik ,  Kim Silverman

发明人： Jerome Bellegarda ,  Devang Naik ,  Kim Silverman

IPC分类号： G06F15/16

CPC分类号： G06Q10/107 ,  H04L51/12

摘要： A method and apparatus for filtering messages comprising determining a first semantic anchor corresponding to a first group of messages, for example, legitimate messages and a second semantic anchor corresponding to a second group of messages, for example, unsolicited messages. Determining a vector corresponding to an incoming message; comparing the vector corresponding to the incoming message with at least one of the first semantic anchor and the second semantic anchor to obtain a first comparison value and a second comparison value; and filtering the incoming message based on the first comparison value and the second comparison value.

摘要翻译： 一种用于过滤消息的方法和装置，包括确定对应于第一组消息的第一语义锚，例如合法消息，以及对应于第二组消息的第二语义锚，例如非请求消息。 确定对应于传入消息的向量; 将与所述输入消息对应的向量与所述第一语义锚和所述第二语义锚中的至少一个进行比较，以获得第一比较值和第二比较值; 以及基于所述第一比较值和所述第二比较值对所述传入消息进行过滤。

公开(公告)号：US08583418B2

公开(公告)日：2013-11-12

申请号：US12240420

申请日：2008-09-29

申请人： Kim Silverman ,  Devang Naik ,  Kevin Lenzo ,  Caroline Henton

发明人： Kim Silverman ,  Devang Naik ,  Kevin Lenzo ,  Caroline Henton

IPC分类号： G06F17/27 ,  G06F17/20

CPC分类号： G10L15/005 ,  G10L13/08

摘要： Algorithms for synthesizing speech used to identify media assets are provided. Speech may be selectively synthesized form text strings associated with media assets. A text string may be normalized and its native language determined for obtaining a target phoneme for providing human-sounding speech in a language (e.g., dialect or accent) that is familiar to a user. The algorithms may be implemented on a system including several dedicated render engines. The system may be part of a back end coupled to a front end including storage for media assets and associated synthesized speech, and a request processor for receiving and processing requests that result in providing the synthesized speech. The front end may communicate media assets and associated synthesized speech content over a network to host devices coupled to portable electronic devices on which the media assets and synthesized speech are played back.

摘要翻译： 提供了用于合成用于识别媒体资产的语音的算法。 可以从与媒体资产相关联的文本串选择性地合成语音。 文本字符串可以被归一化，并且其母语被确定用于获得目标音素，以便以用户熟悉的语言（例如，方言或重音）提供人声音语音。 算法可以在包括几个专用渲染引擎的系统上实现。 该系统可以是耦合到前端的后端的一部分，包括用于媒体资产和相关联的合成语音的存储器，以及用于接收和处理导致提供合成语音的请求的请求处理器。 前端可以通过网络将媒体资产和相关联的合成语音内容通信到主机耦合到其上播放媒体资产和合成语音的便携式电子设备的设备。

公开(公告)号：US20100082348A1

公开(公告)日：2010-04-01

申请号：US12240449

申请日：2008-09-29

申请人： Kim Silverman ,  Devang Naik ,  Jerome Bellegarda ,  Kevin Lenzo

发明人： Kim Silverman ,  Devang Naik ,  Jerome Bellegarda ,  Kevin Lenzo

IPC分类号： G10L13/08

CPC分类号： G10L13/08

摘要： Algorithms for synthesizing speech used to identify media assets are provided. Speech may be selectively synthesized form text strings associated with media assets. A text string may be normalized and its native language determined for obtaining a target phoneme for providing human-sounding speech in a language (e.g., dialect or accent) that is familiar to a user. The algorithms may be implemented on a system including several dedicated render engines. The system may be part of a back end coupled to a front end including storage for media assets and associated synthesized speech, and a request processor for receiving and processing requests that result in providing the synthesized speech. The front end may communicate media assets and associated synthesized speech content over a network to host devices coupled to portable electronic devices on which the media assets and synthesized speech are played back.

摘要翻译： 提供了用于合成用于识别媒体资产的语音的算法。 可以从与媒体资产相关联的文本串选择性地合成语音。 文本字符串可以被归一化，并且其母语被确定用于获得目标音素，以便以用户熟悉的语言（例如，方言或重音）提供人声音语音。 算法可以在包括几个专用渲染引擎的系统上实现。 该系统可以是耦合到前端的后端的一部分，包括用于媒体资产和相关联的合成语音的存储器，以及用于接收和处理导致提供合成语音的请求的请求处理器。 前端可以通过网络将媒体资产和相关联的合成语音内容通信到主机耦合到其上播放媒体资产和合成语音的便携式电子设备的设备。

公开(公告)号：US20100082347A1

公开(公告)日：2010-04-01

申请号：US12240433

申请日：2008-09-29

申请人： Matthew Rogers ,  Kim Silverman ,  Devang Naik ,  Benjamin Rottler

发明人： Matthew Rogers ,  Kim Silverman ,  Devang Naik ,  Benjamin Rottler

IPC分类号： G10L13/08

CPC分类号： G10L13/08

摘要： Algorithms for synthesizing speech used to identify media assets are provided. Speech may be selectively synthesized form text strings associated with media assets. A text string may be normalized and its native language determined for obtaining a target phoneme for providing human-sounding speech in a language (e.g., dialect or accent) that is familiar to a user. The algorithms may be implemented on a system including several dedicated render engines. The system may be part of a back end coupled to a front end including storage for media assets and associated synthesized speech, and a request processor for receiving and processing requests that result in providing the synthesized speech. The front end may communicate media assets and associated synthesized speech content over a network to host devices coupled to portable electronic devices on which the media assets and synthesized speech are played back.

摘要翻译： 提供了用于合成用于识别媒体资产的语音的算法。 可以从与媒体资产相关联的文本串选择性地合成语音。 文本字符串可以被归一化，并且其母语被确定用于获得目标音素，以便以用户熟悉的语言（例如，方言或重音）提供人声音语音。 算法可以在包括几个专用渲染引擎的系统上实现。 该系统可以是耦合到前端的后端的一部分，包括用于媒体资产和相关联的合成语音的存储器，以及用于接收和处理导致提供合成语音的请求的请求处理器。 前端可以通过网络将媒体资产和相关联的合成语音内容通信到主机耦合到其上播放媒体资产和合成语音的便携式电子设备的设备。

公开(公告)号：US20100082328A1

公开(公告)日：2010-04-01

申请号：US12240397

申请日：2008-09-29

申请人： Matthew Rogers ,  Kim Silverman ,  Devang Naik ,  Kevin Lenzo ,  Benjamin Rottler

发明人： Matthew Rogers ,  Kim Silverman ,  Devang Naik ,  Kevin Lenzo ,  Benjamin Rottler

IPC分类号： G06F17/20 ,  G10L13/08

CPC分类号： G10L13/08 ,  G06F17/275

摘要： Algorithms for synthesizing speech used to identify media assets are provided. Speech may be selectively synthesized form text strings associated with media assets. A text string may be normalized and its native language determined for obtaining a target phoneme for providing human-sounding speech in a language (e.g., dialect or accent) that is familiar to a user. The algorithms may be implemented on a system including several dedicated render engines. The system may be part of a back end coupled to a front end including storage for media assets and associated synthesized speech, and a request processor for receiving and processing requests that result in providing the synthesized speech. The front end may communicate media assets and associated synthesized speech content over a network to host devices coupled to portable electronic devices on which the media assets and synthesized speech are played back.

摘要翻译： 提供了用于合成用于识别媒体资产的语音的算法。 可以从与媒体资产相关联的文本串选择性地合成语音。 文本字符串可以被归一化，并且其母语被确定用于获得目标音素，以便以用户熟悉的语言（例如，方言或重音）提供人声音语音。 算法可以在包括几个专用渲染引擎的系统上实现。 该系统可以是耦合到前端的后端的一部分，包括用于媒体资产和相关联的合成语音的存储器，以及用于接收和处理导致提供合成语音的请求的请求处理器。 前端可以通过网络将媒体资产和相关联的合成语音内容通信到主机耦合到其上播放媒体资产和合成语音的便携式电子设备的设备。