Contents analysis method and device

Patent metadata

Field	Value
Publication number	US-10014008-B2
Application number	US-201515122344-A
Country	US
Kind code	B2
Filing date	Mar 3, 2015
Priority date	Mar 3, 2014
Publication date	Jul 3, 2018
Grant date	Jul 3, 2018

Abstract

Provided is a method of analyzing audio content. The method includes: extracting a feature value of the audio content; classifying the audio content on a section basis of the audio content based on the extracted feature value; and selecting at least one section for analyzing the audio content, and performing analysis on audio content of the at least one selected section based on a class to which the audio content of each section belongs.

First claim

The invention claimed is:

1. A method of analyzing an audio signal in a device, the method comprising:
extracting a feature value of the audio signal;
determining a first classification of the audio signal from among at least one of a voice, background music and environment noise based on the extracted feature value;
determining a second classification of the audio signal based on the first classification of the audio signal; and
displaying information related to the audio signal, based on the first and second classification of the audio signal,
wherein the determining the second classification comprises:
performing a voice recognition for the audio signal if the first classification of the audio signal is the voice;
extracting a keyword from the audio signal, based on a result of the voice recognition; and
determining a topic of the audio signal, based on a frequency of at least one of the extracted keyword and reserved word related to the extracted keyword.

2. The method of claim 1 , wherein the determining the first classification comprises determining the first classification of the audio signal by comparing a feature value of the audio signal with a feature value of a database by using the database comprising information regarding at least one feature value of the audio signal.

3. The method of claim 1 , wherein the extracting of the feature value comprises:
decomposing the audio signal into at least one elementary function;
selecting at least one of the at least one elementary function as a dominant elementary function with respect to the decomposed audio signal; and
extracting a basis function as the feature value of the audio signal by using the at least one selected dominant elementary function.

4. The method of claim 1 , wherein the extracting of the feature value comprises:
extracting at least one instantaneous feature value from the audio signal; and
extracting a statistical feature value from the at least one extracted instantaneous feature value.

5. The method of claim 1 , wherein the determining the second classification comprises:
if the first classification of the audio signal is the environment noise
detecting an acoustic event included in the audio signal.

6. The method of claim 1 , wherein the determining the second classification comprises:
performing analysis on video signal corresponding to the audio signal; and
determining the second classification of the audio signal by using analysis results for the video signal.

7. A device comprising:
a receiver for receiving audio signal;
a controller for extracting a feature value of the audio signal, determining a first classification of the audio signal from among at least one of a voice, background music and environment noise based on the extracted feature value, and determining a second classification of the audio signal based on the first classification of the audio signal;
a display for displaying information related to the audio signal, based on the first and second classification of the audio signal,
wherein the controller:
performs a voice recognition for the audio signal if the first classification of the audio signal is the voice;
extracts a keyword from the audio signal, based on a result of the voice recognition; and
determines a topic of the audio signal, based on a frequency of at least one of the extracted keyword and reserved word related to the extracted keyword.

8. The device of claim 7 , wherein the controller determines the first classification of the audio signal by comparing a feature value of the audio signal with a feature value of a database by using the database comprising information regarding at least one feature value of the audio signal.

9. The device of claim 7 , wherein the controller decomposes the audio signal into at least one elementary function, selects at least one of the at least one elementary function as a dominant elementary function with respect to the decomposed audio signal, and extracts a basis function as the feature value of the audio signal by using the at least one selected dominant elementary function.

10. The device of claim 7 , wherein the controller extracts at least one instantaneous feature value from the audio signal, and extracts a statistical feature value from the at least one extracted instantaneous feature value.

11. The device of claim 7 , wherein if the first classification of the audio signal is the environment noise, the controller detects an acoustic event included in the audio signal.

12. The method of claim 2 , wherein the feature value of the database is updated based on the extracted feature value and the determined first classification.

13. The device of claim 8 , wherein the feature value of the database is updated based on the extracted feature value and the determined first classification.

14. The method of claim 1 , wherein the reserved word is a word registered in advance.

15. The device of claim 7 , wherein the reserved word is a word registered in advance.

Assignees

• Samsung Electronics Co Ltd

Inventors

• Cho Nam-Gook
• Lee Woo-Jung
• Kim Sang-Yoon
• Ryu Young-Jun
• Ko Byeong-Seob

Classifications

• H04N21/434

  Disassembling of a multiplex stream, e.g. demultiplexing audio and video streams, extraction of additional data from a video stream; Remultiplexing of multiplex streams; Extraction or processing of SI; Disassembling of packetised elementary stream {(demultiplexing of data packets for data networks, e.g. RTP/UDP H04L65/00)} · CPC title

• H04N21/4394

  involving operations for analysing the audio stream, e.g. detecting features or characteristics in audio streams (arrangements characterised by components specially adapted for monitoring, identification or recognition of audio in broadcast systems H04H60/58) · CPC title

• G10L25/93 Primary

  Discriminating between voiced and unvoiced parts of speech signals (G10L25/90 takes precedence) · CPC title

• G10L25/57

  for processing of video signals · CPC title

• G06F17/30

  Physics · mapped topic