The patent badge is an abbreviated version of the USPTO patent document. The patent badge does contain a link to the full patent document.

The patent badge is an abbreviated version of the USPTO patent document. The patent badge covers the following: Patent number, Date patent was issued, Date patent was filed, Title of the patent, Applicant, Inventor, Assignee, Attorney firm, Primary examiner, Assistant examiner, CPCs, and Abstract. The patent badge does contain a link to the full patent document (in Adobe Acrobat format, aka pdf). To download or print any patent click here.

Date of Patent:
Jan. 27, 2015

Filed:

Oct. 28, 2011
Applicants:

Tomohiro Konuma, Osaka, JP;

Ryouichi Kawanishi, Kyoto, JP;

Tsutomu Uenoyama, Osaka, JP;

Inventors:

Tomohiro Konuma, Osaka, JP;

Ryouichi Kawanishi, Kyoto, JP;

Tsutomu Uenoyama, Osaka, JP;

Attorney:
Primary Examiner:
Assistant Examiner:
Int. Cl.
CPC ...
H04N 5/92 (2006.01); H04N 5/14 (2006.01); G10L 99/00 (2013.01); H04N 21/439 (2011.01); H04N 21/845 (2011.01); G06F 17/30 (2006.01); G11B 27/034 (2006.01); G11B 27/28 (2006.01); G10L 25/81 (2013.01);
U.S. Cl.
CPC ...
H04N 5/147 (2013.01); G10L 99/00 (2013.01); H04N 21/4394 (2013.01); H04N 21/8456 (2013.01); G06F 17/30787 (2013.01); G11B 27/034 (2013.01); G11B 27/28 (2013.01); G10L 25/81 (2013.01);
Abstract

An interesting section extracting device extracts an interesting section of interest to a user from a video file with reference to an audio signal included in the video file such that a specified time is included in the interesting section. The interesting section extracting device includes an interface device that obtains the specified time; and a likelihood vector generating unit that calculates, in one-to-one correspondence with first unit sections of the audio signal, likelihoods for anchor models that respectively represent features of a plurality of types of sound pieces and generates likelihood vectors having the calculated likelihoods as components thereof. An interesting section extracting unit calculates a first feature section as candidate section, which is candidate for the interesting section to be extracted, by using likelihood vectors and extract, as the interesting section, part of the first feature section including the specified time.


Find Patent Forward Citations

Loading…