公开(公告)号：US10282469B2

公开(公告)日：2019-05-07

申请号：US14224511

申请日：2014-03-25

Applicant: OATH INC.

Inventor： Inderjeet Mani

IPC: G10L25/48 ,  G06F17/30 ,  G10L15/18 ,  G10L15/26

Abstract: A multimedia content item is summarized based on its audio track and a desired compression budget. The audio track is extracted and processed by an automatic speech recognizer to obtain a time-aligned text transcript. The text-transcript is partitioned into a plurality of segment sequences. An informativeness score based on a salience score and a diversity score is computed for each of the segments. A coherence score is also computed for the segments in the plurality of sequences. A subsequence of one of the segment sequences that optimizes for informativeness and coherence is selected for generating a new content item summarizing the multimedia content item.

公开(公告)号：US10599721B2

公开(公告)日：2020-03-24

申请号：US15915656

申请日：2018-03-08

Applicant: OATH INC.

Inventor： Inderjeet Mani ,  Eugenio Ciurana ,  Nicholas D'Aloisio-Montilla ,  Bart K. Swanson

IPC: G06F17/30 ,  G06F17/27 ,  G06F16/93 ,  G06N20/00 ,  G06F16/34 ,  G06F17/22

Abstract: One embodiment of a method for summarizing an electronic document includes splitting the electronic document into a plurality of terms, wherein each of the plurality of terms is associated with a respective length, a respective informativeness score, and a respective coherence score, automatically selecting a subset of the plurality of terms, such that an aggregate informativeness score of the subset is maximized while an aggregate length of the subset is less than or equal to a maximum length, and arranging the subset as a summary of the electronic document.