Methods for collecting media segments in a media signal via comparing segments of the signal to later segments

• US 7,593,850 B2
• Filed: 08/22/2003
• Issued: 09/22/2009
• Est. Priority Date: 08/23/2002
• Status: Expired due to Fees

• Current Assignee: Popcatcher AB
• Original Assignee: Popcatcher AB
• Inventors: Berg, Rickard, Ahrne, Tomas, Berg, Jakob
• Primary Examiner(s): Edouard, Patrick N
• Assistant Examiner(s): GODBOLD, DOUGLAS
• Application Number:
  US10/525,370
• Publication Number:
  US 20060104437A1
• Time in Patent Office: 2,223 Days
• Field of Search: 704/270, 704/278, 704226-228, 386/46, 386/52, 380/200
• US Class Current: 704/226
• CPC Class Codes: H03M 7/30 Compression speech analysis-synthesis for redundancy reduction G10L19/00; for image communication H04N; Expansion; Suppression of unnecessary data, e.g. redundancy reduction

Abstract

A method to achieve signals which are essentially devoid of unwanted signal components uses a search and comparison process. Media signals are received through receiving means, the media signals containing unwanted signal components, a representation for the media signals is chosen, and the media signals are processed in such a way that the unwanted signal components are essentially removed and the remaining signal components are saved. The method includes the steps of: from the media signals, choosing a search key representation; in a search track, conducting a first search after a signal representation that contains a section which is essentially identical with the search key; comparing a first segment, which lies before and after the search key, with a second segment which lies before and after the section, which is essentially identical with the first search key; from said first segment and the second segment, finding a common segment; and storing the common segment in the memory domain as a signal representation without unwanted signal components.

Claims

1. A method for receiving media signals (1) through receiving means, said media signals (1) containing unwanted signal components;

to choose a representation for said media signals and to process these media signals (1) in such a way that said unwanted signal components are essentially removed and the remaining signal components are saved, said method comprising the steps of;

from the media signals (1) choosing a first search key representation (20);

in a search track (30), conducting a first search to find a signal representation (10) that contains a section (11) which is essentially identical with said first search key representation (20);

comparing a first segment (40), which lies before and after said search key (20), with a second segment (41) which lies before and after said section (11) which is essentially identical with the first search key representation (20);

from said first segment (40) and said second segment (41) finding a first common segment (44);

loading said common segment (44) into a memory domain; and

storing said common segment (44) in said memory domain as a signal representation (70) without unwanted signal components.

2. Method according to claim 1, wherein said first search (20) is conducted among media signal representations (70) stored in the memory domain.

3. Method according to claim 1, wherein if no essentially identical copy of the search key representation was found, carrying out the further step of conducting further searches in the search track to locate essentially identical copies of said search key representations and, when such a copy is found, conducting a comparison process to find common segments, and continuing this process until a final common segment is achieved or until the process is terminated, and then loading said common segment into the memory domain as a signal representation.

4. Method according to claim 1, including the step of removing all redundant signal representations from the search track if the search track contains a multiple of essentially identical signal representations, to thereby achieve a better use of the memory capacity.

5. Method according to claim 1, wherein a signal representation(70) that lies between two signal representations (71, 73) contained in the memory domain is removed if said signal representation(70) has a time duration that is shorter than a predetermined threshold value.

6. Method according to claim 1, wherein the section of a signal representation (70) that lies between two signal representations contained in the memory domain is saved if the setting of the search key was activated during this section.

7. Method according to claim 1, wherein said search track consist of every N:

th sample of a signal representation (10).

8. Method according to claim 1, wherein the search tracks, when recorded, are normalized to have a common amplitude and sound level.

9. Method according to claim 1, wherein the signal representations (70) are selected from one or more of the group consisting of representations of music, talk, noise, jingles and logotypes.

10. Method according to claim 1, wherein the signal representations are one or more representations selected from the group consisting of music and movies.