知搜-全球专利检索

  1. Login
  2. Sign Up

Search Results

17 records (took 0.019 seconds)

    Systems, Computer-Implemented Methods, and Tangible Computer-Readable Storage Media For Transcription Alighnment
    1.
    发明申请
    Systems, Computer-Implemented Methods, and Tangible Computer-Readable Storage Media For Transcription Alighnment 有权
    系统，计算机实现的方法和有形计算机可读存储介质的转录缩写

    公开(公告)号：US20150046160A1

    公开(公告)日：2015-02-12

    申请号：US14492616

    申请日：2014-09-22

    Applicant: AT&T Intellectual Property I, L.P.

    Inventor： Yeon-Jun KIM ,  David C. GIBBON ,  Horst J. SCHROETER

    IPC: G10L15/26 ,  G10L21/06

    CPC classification number: G10L15/265 ,  G10L13/08 ,  G10L15/26 ,  G10L21/055 ,  G10L21/06 ,  G10L25/51 ,  G11B27/10 ,  H04N21/44004 ,  H04N21/4884

    Abstract: Disclosed herein are systems, computer-implemented methods, and tangible computer-readable storage media for captioning a media presentation. The method includes receiving automatic speech recognition (ASR) output from a media presentation and a transcription of the media presentation. The method includes selecting via a processor a pair of anchor words in the media presentation based on the ASR output and transcription and generating captions by aligning the transcription with the ASR output between the selected pair of anchor words. The transcription can be human-generated. Selecting pairs of anchor words can be based on a similarity threshold between the ASR output and the transcription. In one variation, commonly used words on a stop list are ineligible as anchor words. The method includes outputting the media presentation with the generated captions. The presentation can be a recording of a live event.

    Abstract translation: 本文公开了系统，计算机实现的方法和用于标题媒体呈现的有形的计算机可读存储介质。 该方法包括从媒体呈现和媒体呈现的转录接收自动语音识别（ASR）输出。 该方法包括：通过处理器选择基于ASR输出和转录的媒体呈现中的一对锚定词，并通过将转录与所选择的一对锚点之间的ASR输出对齐来产生标题。 转录可以是人类产生的。 选择锚点对可以基于ASR输出和转录之间的相似性阈值。 在一个变体中，停止列表上常用的单词不符合锚点词。 该方法包括用生成的标题输出媒体呈现。 演示文稿可以是现场直播的录音。

    System and Method for Adaptive Media Playback Based on Destination
    6.
    发明申请
    System and Method for Adaptive Media Playback Based on Destination 有权
    基于目的地的自适应媒体播放的系统和方法

    公开(公告)号：US20150235671A1

    公开(公告)日：2015-08-20

    申请号：US14703073

    申请日：2015-05-04

    Applicant: AT&T Intellectual Property I, L.P.

    Inventor： Behzad SHAHRARAY ,  Andrea BASSO ,  Lee BEGEJA ,  David C. GIBBON ,  Zhu LIU ,  Bernard S. RENGER

    IPC: G11B27/10 ,  G11B27/034 ,  G11B27/00

    CPC classification number: G06Q30/0261 ,  G01C21/3697 ,  G11B27/005 ,  G11B27/034 ,  G11B27/105 ,  H04L67/18 ,  H04N5/783

    Abstract: Disclosed herein are systems, methods, and computer readable-media for adaptive media playback based on destination. The method for adaptive media playback comprises determining one or more destinations, collecting media content that is relevant to or describes the one or more destinations, assembling the media content into a program, and outputting the program. In various embodiments, media content may be advertising, consumer-generated, based on real-time events, based on a schedule, or assembled to fit within an estimated available time. Media content may be assembled using an adaptation engine that selects a plurality of media segments that fit in the estimated available time, orders the plurality of media segments, alters at least one of the plurality of media segments to fit the estimated available time, if necessary, and creates a playlist of selected media content containing the plurality of media segments.

    Abstract translation: 本文公开了用于基于目的地的自适应媒体回放的系统，方法和计算机可读介质。 用于自适应媒体播放的方法包括确定一个或多个目的地，收集与一个或多个目的地相关或描述一个或多个目的地的媒体内容，将媒体内容组合成节目并输出该节目。 在各种实施例中，媒体内容可以是基于实时事件的广告，消费者生成的，基于时间表，或被组合以适合估计的可用时间。 可以使用适配引擎来组合媒体内容，该适配引擎选择适合估计的可用时间的多个媒体片段，对多个媒体片段进行排序，如果需要，改变多个媒体片段中的至少一个以适合估计的可用时间 并且创建包含多个媒体片段的所选媒体内容的播放列表。

    SYSTEM AND METHOD FOR DYNAMICALLY CONSTRUCTING PERSONALIZED CONTEXTUAL VIDEO PROGRAMS
    7.
    发明申请
    SYSTEM AND METHOD FOR DYNAMICALLY CONSTRUCTING PERSONALIZED CONTEXTUAL VIDEO PROGRAMS 审中-公开
    动态构建个性化视频节目的系统与方法

    公开(公告)号：US20140298363A1

    公开(公告)日：2014-10-02

    申请号：US14299798

    申请日：2014-06-09

    Applicant: AT&T INTELLECTUAL PROPERTY I, L.P.

    Inventor： Lee BEGEJA ,  David C. GIBBON ,  Paul VAN VLECK

    IPC: H04N21/2543 ,  H04N21/81 ,  H04N21/235

    CPC classification number: H04N21/2543 ,  G06Q30/02 ,  H04N7/165 ,  H04N21/234318 ,  H04N21/235 ,  H04N21/44222 ,  H04N21/4524 ,  H04N21/45452 ,  H04N21/812

    Abstract: Systems, methods, and computer readable-media for dynamically constructing personalized contextual video programs include extracting video metadata from a video program, extracting component metadata from video components stored in a media object library, extracting viewer preferences, receiving synchronization information about the video program, identifying a video program segment susceptible to inserting a video component, and transmitting the video component to a playback device and instructions detailing how to insert the video component in the video program segment. Video metadata can be extracted in real time. A viewer profile can be based on demographic information and user behavior. The video program and the video component can be combined before transmitting the video component and instructions to the playback device. A video component can be selected based on which advertiser offers to pay the most. The transmitted video component and set of instructions can be stored as a construction list for future use.

    Abstract translation: 用于动态构建个性化上下文视频节目的系统，方法和计算机可读介质包括从视频节目提取视频元数据，从存储在媒体对象库中的视频组件中提取组件元数据，提取观看者喜好，接收关于视频节目的同步信息， 识别容易插入视频组件的视频节目片段，以及将视频组件发送到播放设备以及详细说明如何将视频组件插入视频节目片段中的指令。 视频元数据可以实时提取。 观众简介可以基于人口统计信息和用户行为。 视频节目和视频分量可以在将视频分量和指令发送到播放设备之前被组合。 可以根据哪个广告客户提供的付款最多可以选择一个视频组件。 传输的视频组件和指令集可以存储为构造列表以备将来使用。

    SYSTEM AND METHOD FOR RICH MEDIA ANNOTATION
    8.
    发明申请
    SYSTEM AND METHOD FOR RICH MEDIA ANNOTATION 审中-公开

    公开(公告)号：US20190079932A1

    公开(公告)日：2019-03-14

    申请号：US16186956

    申请日：2018-11-12

    Applicant: AT&T Intellectual Property I, L.P.

    Inventor： Paul GAUSMAN ,  David C. GIBBON

    IPC: G06F17/30

    Abstract: Disclosed herein are systems, methods, and computer readable-media for rich media annotation, the method comprising receiving a first recorded media content, receiving at least one audio annotation about the first recorded media, extracting metadata from the at least one of audio annotation, and associating all or part of the metadata with the first recorded media content. Additional data elements may also be associated with the first recorded media content. Where the audio annotation is a telephone conversation, the recorded media content may be captured via the telephone. The recorded media content, audio annotations, and/or metadata may be stored in a central repository which may be modifiable. Speech characteristics such as prosody may be analyzed to extract additional metadata. In one aspect, a specially trained grammar identifies and recognizes metadata.

    SYSTEMS, COMPUTER-IMPLEMENTED METHODS, AND TANGIBLE COMPUTER-READABLE STORAGE MEDIA FOR TRANSCRIPTION ALIGNMENT
    9.
    发明申请
    SYSTEMS, COMPUTER-IMPLEMENTED METHODS, AND TANGIBLE COMPUTER-READABLE STORAGE MEDIA FOR TRANSCRIPTION ALIGNMENT 有权
    系统，计算机实现方法和可变数据可读存储介质用于转码对齐

    公开(公告)号：US20170061986A1

    公开(公告)日：2017-03-02

    申请号：US15350339

    申请日：2016-11-14

    Applicant: AT&T Intellectual Property I, L.P.

    Inventor： Yeon-Jun KIM ,  David C. GIBBON ,  Horst J. SCHROETER

    IPC: G10L21/055 ,  G10L25/51 ,  H04N21/44 ,  G10L13/08 ,  H04N21/488 ,  G10L21/06 ,  G10L15/26

    CPC classification number: G10L15/265 ,  G10L13/08 ,  G10L15/26 ,  G10L21/055 ,  G10L21/06 ,  G10L25/51 ,  G11B27/10 ,  H04N21/44004 ,  H04N21/4884

    Abstract: Disclosed herein are systems, computer-implemented methods, and tangible computer-readable storage media for captioning a media presentation. The method includes receiving automatic speech recognition (ASR) output from a media presentation and a transcription of the media presentation. The method includes selecting via a processor a pair of anchor words in the media presentation based on the ASR output and transcription and generating captions by aligning the transcription with the ASR output between the selected pair of anchor words. The transcription can be human-generated. Selecting pairs of anchor words can be based on a similarity threshold between the ASR output and the transcription. In one variation, commonly used words on a stop list are ineligible as anchor words. The method includes outputting the media presentation with the generated captions. The presentation can be a recording of a live event.

    Abstract translation: 本文公开了系统，计算机实现的方法和用于标题媒体呈现的有形的计算机可读存储介质。 该方法包括从媒体呈现和媒体呈现的转录接收自动语音识别（ASR）输出。 该方法包括：通过处理器选择基于ASR输出和转录的媒体呈现中的一对锚定词，并通过将转录与所选择的一对锚点之间的ASR输出对齐来产生标题。 转录可以是人类产生的。 选择锚点对可以基于ASR输出和转录之间的相似性阈值。 在一个变体中，停止列表上常用的单词不符合锚点词。 该方法包括用生成的标题输出媒体呈现。 演示文稿可以是现场直播的录音。

Patent Agency Ranking