Method and system for analyzing digital audio files

• US 7,277,766 B1
• Filed: 10/24/2000
• Issued: 10/02/2007
• Est. Priority Date: 10/24/2000
• Status: Expired due to Term

• Current Assignee: Rovi Technologies Corporation (Adeia Inc.)
• Original Assignee: Moodlogic (Adeia Inc.)
• Inventors: Khan, Rehan M., Tzanetakis, George
• Primary Examiner(s): Chin; Vivian
• Assistant Examiner(s): LAO, LUNSEE
• Application Number:
  US09/695,457
• Time in Patent Office: 2,534 Days
• Field of Search: 700/94, 381/56, 381/98, 846/09, 84604-605, 846/02, 707/104.1, 707/102, 707/6, 707/6.8, 709/219, 704246-250, 713/170, 713/190, 713/176, 713/716, 380236-238, 380209-210, 380/53, 702 71- 79, 702/189, 340/5.26
• US Class Current: 700/94
• CPC Class Codes: G06F 16/634 Query by example, e.g. query by humming
• CPC Class Codes: G06F 16/635 Filtering based on additional data, e.g. user or group profiles
• CPC Class Codes: G06F 16/639 using playlists
• CPC Class Codes: G06F 16/68 Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually
• CPC Class Codes: G06F 16/683 using metadata automatically derived from the content
• CPC Class Codes: Y10S 707/99945 Object-oriented database structure processing
• CPC Class Codes: Y10S 707/99948 Application of database or data structure, e.g. distributed, multimedia, or image

Abstract

A method and system for analyzing audio files is provided. Plural audio file feature vector values based on an audio file'"'"'s content are determined and the audio file feature vectors are stored in a database that also stores other pre-computed audio file features. The process determines if the audio files feature vectors match the stored audio file vectors. The process also associates a plurality of known attributes to the audio file.

Claims

1. An apparatus for fingerprinting an audio waveform, comprising:

a memory that stores a codebook comprising one or more multivariate vectors, each multivariate vector comprising one or more spectral features and represents a corresponding one of a plurality of codes; and

a processor configured to divide the audio waveform into a plurality of bins, compute one or more spectral features for each bin, and generate a fingerprint representing the audio waveform with a string of codes from the codebook based on the computed one or more spectral features for each bin.

2. The apparatus of claim 1, wherein the string of codes are aligned in a time series.

3. The apparatus of claim 1 wherein the processor compresses the string of codes to form a compressed string of codes.

4. The apparatus of claim 1 wherein each code in the string of codes is temporally aligned with the audio waveform such that the position of a code within the string corresponds to a time period of the audio waveform.

5. The apparatus of claim 1, wherein the processor compresses the string of codes such that temporal alignment between the string of codes and the audio waveform is maintained.

6. The apparatus of claim 1 wherein each code is a hash code.

7. The apparatus of claim 1, wherein the fingerprint is a unique identifier.

8. The apparatus of claim 1, wherein the fingerprint is a unique audio signature.

9. The apparatus of claim 1, wherein the processor computes the one or more spectral features for a first group of data points within each bin, shifts one or more data points within each bin, and computes the one or more spectral features for a second group of data points within each bin.

10. The apparatus of claim 1, wherein the codebook is predefined prior to representing the audio waveform with the string of codes from the codebook.

11. The apparatus of claim 1, wherein the processor generates the fingerprint at a given time within the audio waveform.

12. The apparatus of claim 1, wherein the processor queries a database of audio files using the fingerprint.

13. The apparatus of claim 1, wherein the processor queries a database of pre-computed audio signatures.

14. The apparatus of claim 13, wherein the pre-computed audio signatures correspond respectively to a plurality of audio files.

15. The apparatus of claim 9, wherein the processor receives at least one attribute associated with the fingerprint.

16. The apparatus of claim 9, wherein the processor receives at least one attribute, wherein the at least one attribute is tagged to an audio file.

17. The apparatus of claim 9, wherein the processor receives a customized playlist based on the fingerprint.

18. The apparatus of claim 17, wherein the customized playlist is further based on a predetermined set of user preferences.

19. The apparatus of claim 9, wherein the processor receives at least one feature value based on the fingerprint.

20. The apparatus of claim 19, wherein the at least one feature value is at least one of an emotional quality vector value, a vocal vector value, a sound quality vector values, a situational quality vector value, an ensemble vector value, a genre vector value, and an instrument vector value.