Methods for embedding data in digital audio data

• US 8,332,954 B2
• Filed: 10/01/2001
• Issued: 12/11/2012
• Est. Priority Date: 10/01/2001
• Status: Active Grant

• Current Assignee: Hong Kong University of Science and Technology
• Original Assignee: Hong Kong University of Science and Technology
• Inventors: Tam, Kar Yan, Tsang, Kai Fung, Kwok, Sai Ho, Cheung, Shing Chi
• Primary Examiner(s): LANIER, BENJAMIN E
• Application Number:
  US09/965,831
• Publication Number:
  US 20030068043A1
• Time in Patent Office: 4,089 Days
• Field of Search: 713/176
• US Class Current: 726/27
• CPC Class Codes: G10L 19/018 Audio watermarking, i.e. embedding inaudible data in the audio signal
• CPC Class Codes: G11B 20/00086 Circuits for prevention of unauthorised reproduction or copying, e.g. piracy indicating unauthorised use of record carriers in general G11B23/28; scrambling for television signal recording H04N5/913; network architectures or network protocols for network security H04L63/00; cryptographic mechanisms or cryptographic arrangements for secret or secure communication H04L9/00
• CPC Class Codes: G11B 20/0021 involving encryption or decryption of contents recorded on or reproduced from a record carrier
• CPC Class Codes: G11B 20/00289 wherein the key is stored as a watermark
• CPC Class Codes: G11B 20/00826 wherein a spoiler signal is added to degrade the signal quality
• CPC Class Codes: G11B 20/00891 embedded in audio data
• CPC Class Codes: G11B 20/00978 wherein the record carrier stores a trial version of a content

Abstract

The present invention provides a method of embedding watermarking data in an audio signal which includes the steps of incorporating watermarking information into said audio signal, sectioning said signal into at least two sections, marking at least one said section whereby said sections may be identified, generating distortion in one said section of said signal in manner recoverable by a key obtainable from at least one other said section, and appending said distorted section to said at least one other section to form a composite signal comprising a distorted section to said at least one other section to form a composite signal comprising a distorted section and at least one undistorted section. The present invention is particularly useful for enabling advertisements and or trial listening sections to be added to an audio file in such a manner that they include the key for recovering a distorted section. If an attempt is made to remove the advertisement or trial section, then the key would be lost.

Claims

1. A method of embedding watermarking data in an audio signal, comprising the steps of:

(a) incorporating watermarking information into said audio signal, to form a watermarked audio signal,

(b) sectioning said watermarked audio signal into at least two sections each section having audio content, each of said sections corresponding to a respective time period of said audio signal,

(c) marking at least one of said sections whereby said sections may be identified,

(d) generating distortion in a first one of said sections of said signal in a manner recoverable by a key obtainable from at least one other section having audio content, and

(e) appending said distorted section to said at least one other section to form a composite signal comprising a distorted section and at least one undistorted section.

2. A method according to claim 1 wherein said key is embedded in said audio content of said at least one other section.

3. A method as claimed in claim 2 wherein the first section is distorted by means of a scrambling function.

4. A method as claimed in claim 1 wherein said key is obtained directly from a sequence of bits contained in said audio content of at least one other section.

5. A method as claimed in claim 4 wherein said key is obtained by applying a hashing function to the bit sequence of said audio content of said at least one other section.

6. A method as claimed in claim 5 wherein the output of the hashing function is added to the bitstream of said first section to create said distortion.

7. A method as claimed in claim 4 wherein a bitstream of said first section is subject to a scrambling function to create said distortion.

8. A method as claimed in claim 1 wherein said first section comprises a section to which access is to be restricted.

9. A method as claimed in claim 1 wherein said at least one other section comprises an advertisement.

10. A method as claimed in claim 1 wherein said at least one other section comprises a trial listening section.

11. A method as claimed in claim 1 wherein said at least one other section comprises an advertisement section and a trial listening section.

12. A method as claimed in claim 1 wherein said audio signal is compressed after watermarking.

13. A method as claimed in claim 12 wherein said first section of said compressed signal is distorted by means of a scrambling function that receives as a key the output of a hashing function that acts upon said at least one other section.

14. A method as claimed in claim 13 wherein said audio signal is compressed in MP3 format and said scrambling function acts upon the bits contained within MP3 frames.

15. A method of playing back an audio signal having data embedded within it by the method of claim 1, comprising;

(a) reading said composite signal,

(b) identifying said sections,

(c) obtaining said key from said at least one undistorted section, and

(d) recovering said distorted section.

16. A method as claimed in claim 15 wherein said distorted section is recovered in real time without being written to memory.

28. A method as claimed in claim 2 wherein said distortion is generated by creating a pseudo-random number sequence for adding as pseudo-random noise to said first section, and wherein said pseudo-random number sequence is embedded in said at least one other section to enable said random noise to be subsequently removed.

29. A method according to claim 28 wherein said key is embedded using a fragile data hiding technique.

17. A watermarked audio signal stored in a non-transitory memory or a non-transitory computer readable medium comprising at least two sections, each section having audio content and corresponding to a respective time period of said audio signal, said sections including a first section which is distorted in a manner recoverable by means of a key obtainable from audio content in at least one other section.

18. A watermarked audio signal as claimed in claim 17 wherein said first section is a section to which access is restricted.

19. A watermarked audio signal as claimed in claim 17 wherein said at least one other section is an advertisement section.

20. A watermarked audio signal as claimed in claim 17 wherein said at least one other section comprises a trial listening section.

21. A watermarked audio signal as claimed in claim 17 wherein said at least one other section comprises an advertisement section and a trial listening section.

22. A non-transitory computer-readable storage medium comprising computer-executable instructions that, in response to execution, cause a system including a processer to perform operations comprising:

incorporating watermarking , information into said audio signal to form a watermarked audio signal,

sectioning said watermarked audio signal into at least two sections each having audio content, each section corresponding to a respective time period of said audio signal,

marking at east one of said sections whereby said sections may be identified,

generating distortion in one of said sections of said signal in a manner recoverable by a key obtainable from at least one other section having audio content, and

appending said distorted section to said at least one other section to form a composite signal comprising a distorted section and at least one undistorted section.

23. A non-transitory computer-readable storage medium comprising computer-executable instructions that, in response to execution, cause a system including a processor to perform operations comprising:

incorporating watermarking information into an audio signal, to form a watermarked audio signal,

sectioning, said watermarked audio signal into at least two sections each section having audio content, each of said sections corresponding to a respective time period of said audio signal,

marking at least one of said sections whereby said sections may be indentified,

generation distortion in a first one of said sections of said signal in a manner recoverable by a key obtainable from at least one other section having audio content,

appending said distorted section to said at least one other section to form a composite signal comprising a distorted section and at least one undistorted section,

reading said composite signal,

identifying said sections,

obtaining said key from said at least one undistorted section, and

recovering said distorted section.

24. A method for including an advertisement with audio data in an audio signal comprising, providing or creating an audio signal comprising a first section having audio content and an advertisement section having audio content, said first section and said advertisement section corresponding to respective time periods of said audio signal, generating distortion of said first section in a manner recoverable by a key obtainable from said advertisement section, and appending said distorted first section to said advertisement section wherein said key is obtainable from said audio content in said advertisement section.

25. A method for including a trial listening section with audio data in an audio signal comprising, sectioning said audio signal into a first section and a trial listening section, said first section and trial listening section corresponding to respective time periods of said audio signal, generating distortion of said first section in a manner recoverable by a key obtainable from said trial listening section, and appending said distorted first section to said trial listening section, wherein the key is obtainable from audio content in said trial listening section.

26. A method for including an advertisement section and a trial listening section with audio data in an audio signal, including sectioning said signal into a first section, an advertisement section, and a trial listening section, said sections corresponding to respective time periods of said audio signal, marking at least one of said sections whereby said sections may be identified, generating distortion in said first section in a manner recoverable by a key obtainable from at least one of said advertisement and trial listening sections, and appending said distorted first section to said advertisement and trial listening sections to form a composite signal, wherein said key is obtainable from audio content in said advertisement section.

27. A method of restricting access to a part of a media signal, comprising the steps of:

(a) sectioning said signal into at least two sections each having media content, each section corresponding to a respective period of time of said signal,

(b) marking at least one of said sections whereby said sections may be identified,

(c) generating distortion in one of said sections of said signal in a manner recoverable by a key obtainable from or more sections having media content, wherein said key is, obtainable from said media content in said one or more other sections, and

(d) appending said distorted section to said one or more other sections to form a composite signal comprising a distorted section and at least one undistorted section.