International Business Machines Corporation

Jeff Amsterdam

Aaron K Baughman

Craig M Trim

Stephen Carl Hammer

Christopher Eythan Holladay

David Alexander Provan

Systems and methods for dynamically providing customized versions of video content are disclosed. In embodiments, a method comprises: analyzing a video to determine content and context of portions of the video; assigning one or more content editing categories to the portions of the video based on the analyzing; determining an unwanted scene of the video based on the one or more content editing categories and user profile data of a viewer; determining a style component of the unwanted scene based on context of the unwanted scene and the user profile data; generating custom content to replace the unwanted scene of the video based on an acceptable portion of content corresponding to the unwanted scene and the style component; editing the video to replace the unwanted scene of the video with the custom content to produce an edited video including the custom content; and providing the edited video to the viewer.

Dynamically providing customized versions of video content

Aspects of the present invention provide devices that subtitle streaming video with audio and identify a speaker in a streaming video with audio according to words spoken by the speaker matched to a cognitive print. The cognitive print includes traits classified according a hierarchical long short term model (LSTM). The hierarchical LSTM includes layers of LSTMs and each layer corresponds to the classification of one trait. A processor annotates a subtitle of the words spoken by the speaker, which decorates the subtitle with a label representative of the identified speaker, and streams the decorated subtitle with the streaming video with audio.

Growing community of inventors

Roswell, GA, United States of America

Jeff Amsterdam