Methods and apparatus for audio recognition
View Patent ↗A method, apparatus and computer memory are provided for recognizing an audio fingerprint of an unknown audio recording. A database stores a plurality of audio recording identifiers corresponding to a plurality of known audio recordings, where the audio recording identifiers are organized by variation information about the audio recordings. A processor searches a database and identifies at least one of the audio recording identifiers corresponding to the audio fingerprint, where the audio fingerprint includes variation information of the unknown audio recording.
1. An apparatus for recognizing an audio fingerprint including variation information of an unknown audio recording, comprising:
a database operable to store a plurality of audio recording identifiers and metadata corresponding to a plurality of known audio recordings, wherein each of the plurality of audio recording identifiers are organized by variation information about the known audio recordings;
a processor operable to search the database to identify at least one of the plurality of audio recording identifiers corresponding to the audio fingerprint using the variation information of the unknown audio recording and the variation information of the known audio recordings, wherein the variation information of the unknown audio recording is based at least in part on a time ordered comparison of a plurality of predetermined model frequencies and a portion of audio data from the unknown audio recording, and the variation information of the plurality of known audio recordings is based at least in part on a time ordered comparison of the plurality of predetermined model frequencies and a portion of audio data from the plurality of known audio recordings; and
the processor further operable to cause the database to communicate to a source of the unknown fingerprint metadata corresponding to at least one of the plurality of audio recording identifiers.
2. An apparatus according to claim 1 , wherein the processor is further operable to determine cluster information from the variation information of the unknown audio recording, and identify a subset the audio recording identifiers based on the cluster data.
3. An apparatus according to claim 1 , wherein the variation information of a known audio recording is variation information about a plurality of frequency families of the known recording, and wherein the variation information of the unknown audio recording is variation information about a plurality of frequency families of the unknown recording.
4. An apparatus according to claim 1 , wherein the variation information of a known audio recording is based on at least one of the predominance and presence of a plurality of frequency families of the known recording, and wherein the variation information of the unknown audio recording is based on at least one of the predominance and presence of a plurality of frequency families of the unknown recording.
5. An apparatus according to claim 1 , wherein the variation information of a known audio recording is based on at least one of the predominance and presence of a combined set of the frequency families of the known recording, and wherein the variation information of the unknown audio recording is based on at least one of the predominance and presence of a combined set of the frequency families of the unknown recording.
6. An apparatus according to claim 1 , wherein the variation information of a known recording is based on an average of a summed set of frequency families plus a deviation of the known recording, and wherein the variation information of the unknown recording is based on an average of a summed set of frequency families plus a deviation the unknown recording.
7. An apparatus according to claim 1 , wherein the variation information of a known audio recording is the sum of the total variation of a predetermined number of the highest variation frequency families of the known audio recording, and wherein the variation information of the unknown audio recording is the sum of the total variation of a predetermined number of the highest variation frequency families of the unknown audio recording.
8. An apparatus according to claim 1 , wherein the variation information of a known recording is the sum of a predetermined number of the frequency families of the known audio recording having the highest level of variation, and wherein the variation information of the unknown recording is the sum of a predetermined number of the frequency families of the unknown audio recording having the highest level of variation.
9. An apparatus according to claim 1 , wherein the variation information of a known audio recording is based on a translation of a total running time of the known audio recording, and wherein the variation information of the unknown audio recording is based on a translation of a total running time of the unknown audio recording.
10. An apparatus according to claim 1 , wherein the variation information of the unknown audio recording is within a predetermined tolerance of the variation information of at least one of the known audio recordings.
11. An apparatus according to claim 1 , wherein the database is further operable to communicate to the source of the unknown fingerprint at least one of the audio recording identifiers corresponding to the unknown fingerprint.
12. An apparatus according to claim 1 , wherein the processor is further operable to recognize the variation information of a known audio recording that is substantially similar to the variation information of the unknown audio recording.
13. A network computer system, comprising the apparatus for recognizing an audio fingerprint of claim 1 .
14. A device-embedded system, comprising the apparatus for recognizing an audio fingerprint of claim 1 .
15. A method for recognizing an audio fingerprint including variation information of an unknown audio recording, comprising:
organizing a plurality of audio recording identifiers corresponding to a plurality of known audio recordings by variation information about the plurality of known audio recordings;
identifying at least one of the plurality of audio recording identifiers corresponding to the audio fingerprint using the variation information of the unknown audio recording and the variation information of the plurality of known audio recordings, wherein the variation information of the unknown audio recording is based at least in part on a time ordered comparison of a plurality of predetermined model frequencies and a portion of audio data from the unknown audio recording, and the variation information of the plurality of known audio recordings is based at least in part on a time ordered comparison of the plurality of predetermined model frequencies and a portion of audio data from the plurality of known audio recordings; and
communicating to a source of the unknown fingerprint at least metadata corresponding to the unknown fingerprint based on the at least one audio recording identifier obtained by the identifying.
16. A method according to claim 15 , further comprising:
determining cluster information from the variation information of the unknown audio recording; and identifying a subset the audio recording identifiers based on the cluster data.
17. A method according to claim 15 , wherein the variation information of a known audio recording is variation information about a plurality of frequency families of the known recording, and wherein the variation information of the unknown audio recording is variation information about a plurality of frequency families of the unknown recording.
18. A method according to claim 15 , wherein the variation information of a known audio recording is based on at least one of the predominance and presence of a plurality of frequency families of the known recording, and wherein the variation information of the unknown audio recording is based on at least one of the predominance and presence of a plurality of frequency families of the unknown recording.
19. A method according to claim 15 , wherein the variation information of a known audio recording is based on at least one of the predominance and presence of a combined set of the frequency families of the known recording, and wherein the variation information of the unknown audio recording is based on at least one of the predominance and presence of a combined set of the frequency families of the unknown recording.
20. A method according to claim 15 , wherein the variation information of a known recording is based on an average of a summed set of frequency families plus a deviation of the known recording, and wherein the variation information of the unknown recording is based on an average of a summed set of frequency families plus a deviation the unknown recording.
21. A method according to claim 15 , wherein the variation information of a known audio recording is the sum of the total variation of a predetermined number of the highest variation frequency families of the known audio recording, and wherein the variation information of the unknown audio recording is the sum of the total variation of a predetermined number of the highest variation frequency families of the unknown audio recording.
22. A method according to claim 15 , wherein the variation information of a known recording is the sum of a predetermined number of the frequency families of the known audio recording having the highest level of variation, and wherein the variation information of the unknown recording is the sum of a predetermined number of the frequency families of the unknown audio recording having the highest level of variation.
23. A method according to claim 15 , wherein the variation information of a known audio recording is based on a translation of a total running time of the known audio recording, and wherein the variation information of the unknown audio recording is based on a translation of a total running time of the unknown audio recording.
24. A method according to claim 15 , wherein the variation information of the unknown audio recording is within a predetermined tolerance of the variation information of at least one of the known audio recordings.
25. A method according to claim 15 , further comprising:
communicating to the source of the unknown fingerprint the audio recording identifier corresponding to the unknown fingerprint.
26. A method according to claim 15 , further comprising:
recognizing the variation information of a known audio recording that is substantially similar to the variation information of the unknown audio recording.
27. An apparatus for recognizing an audio fingerprint including variation information of an unknown audio recording, comprising:
means for organizing a plurality of audio recording identifiers corresponding to a plurality of known audio recordings by variation information about the plurality of known audio recordings;
means for identifying at least one of the audio recording identifiers corresponding to the audio fingerprint using the variation information of the unknown audio recording and the variation information of the plurality of known audio recordings, wherein the variation information of the unknown audio recording is based at least in part on a time ordered comparison of a plurality of predetermined model frequencies and a portion of audio data from the unknown audio recording, and the variation information of the plurality of known audio recordings is based at least in part on a time ordered comparison of the plurality of predetermined model frequencies and a portion of audio data from the plurality of known audio recordings; and
means for communicating to a source of the unknown fingerprint at least metadata corresponding to the unknown fingerprint based on the at least one audio recording identifier obtained by the means for identifying.
28. An apparatus according to claim 27 , further comprising:
means for determining cluster information from the variation information of the unknown audio recording; and identifying a subset the audio recording identifiers based on the cluster data.
29. An apparatus according to claim 27 , wherein the variation information of a known audio recording is variation information about a plurality of frequency families of the known recording, and wherein the variation information of the unknown audio recording is variation information about a plurality of frequency families of the unknown recording.
30. An apparatus according to claim 27 , wherein the variation information of a known audio recording is based on at least one of the predominance and presence of a plurality of frequency families of the known recording, and wherein the variation information of the unknown audio recording is based on at least one of the predominance and presence of a plurality of frequency families of the unknown recording.
31. An apparatus according to claim 27 , wherein the variation information of a known audio recording is based on at least one of the predominance and presence of a combined set of the frequency families of the known recording, and wherein the variation information of the unknown audio recording is based on at least one of the predominance and presence of a combined set of the frequency families of the unknown recording.
32. An apparatus according to claim 27 , wherein the variation information of a known recording is based on an average of a summed set of frequency families plus a deviation of the known recording, and wherein the variation information of the unknown recording is based on an average of a summed set of frequency families plus a deviation the unknown recording.
33. An apparatus according to claim 27 , wherein the variation information of a known audio recording is the sum of the total variation of a predetermined number of the highest variation frequency families of the known audio recording, and wherein the variation information of the unknown audio recording is the sum of the total variation of a predetermined number of the highest variation frequency families of the unknown audio recording.
34. An apparatus according to claim 27 , wherein the variation information of a known recording is the sum of a predetermined number of the frequency families of the known audio recording having the highest level of variation, and wherein the variation information of the unknown recording is the sum of a predetermined number of the frequency families of the unknown audio recording having the highest level of variation.
35. An apparatus according to claim 27 , wherein the variation information of a known audio recording is based on a translation of a total running time of the known audio recording, and wherein the variation information of the unknown audio recording is based on a translation of a total running time of the unknown audio recording.
36. An apparatus according to claim 27 , wherein the variation information of the unknown audio recording is within a predetermined tolerance of the variation information of at least one of the known audio recordings.
37. An apparatus according to claim 27 , further comprising:
means for communicating to the source of the unknown fingerprint the audio recording identifier corresponding to the unknown fingerprint.
38. An apparatus according to claim 27 , further comprising:
means for recognizing the variation information of a known audio recording that is substantially similar to the variation information of the unknown audio recording.
39. Computer-readable medium storing computer executable code including instructions which when executed by a computer processor causes the computer processor to perform:
organizing a plurality of audio recording identifiers corresponding to a plurality of known audio recordings by variation information about the plurality of known audio recordings;
identifying at least one of the audio recording identifiers corresponding to the audio fingerprint using the variation information of the unknown audio recording and the variation information of the plurality of known audio recordings, wherein the variation information of the unknown audio recording is based at least in part on a time ordered comparison of a plurality of predetermined model frequencies and a portion of audio data from the unknown audio recording, and the variation information of the plurality of known audio recordings is based at least in part on a time ordered comparison of the plurality of predetermined model frequencies and a portion of audio data from the plurality of known audio recordings; and
communicating to a source of the unknown fingerprint at least metadata corresponding to the unknown fingerprint based on the at least one audio recording identifier obtained by the identifying.
40. Computer-readable medium storing computer executable code according to claim 39 , further including code for causing the computer processor to perform:
determining cluster information from the variation information of the unknown audio recording; and identifying a subset the audio recording identifiers based on the cluster data.
41. Computer-readable medium storing computer executable code according to claim 39 , wherein the variation information of a known audio recording is variation information about a plurality of frequency families of the known recording, and wherein the variation information of the unknown audio recording is variation information about a plurality of frequency families of the unknown recording.
42. Computer-readable medium storing computer executable code according to claim 39 , wherein the variation information of a known audio recording is based on at least one of the predominance and presence of a plurality of frequency families of the known recording, and wherein the variation information of the unknown audio recording is based on at least one of the predominance and presence of a plurality of frequency families of the unknown recording.
43. Computer-readable medium storing computer executable code according to claim 39 , wherein the variation information of a known audio recording is based on at least one of the predominance and presence of a combined set of the frequency families of the known recording, and wherein the variation information of the unknown audio recording is based on at least one of the predominance and presence of a combined set of the frequency families of the unknown recording.
44. Computer-readable medium storing computer executable code according to claim 39 , wherein the variation information of a known recording is based on an average of a summed set of frequency families plus a deviation of the known recording, and wherein the variation information of the unknown recording is based on an average of a summed set of frequency families plus a deviation the unknown recording.
45. Computer-readable medium storing computer executable code according to claim 39 , wherein the variation information of a known audio recording is the sum of the total variation of a predetermined number of the highest variation frequency families of the known audio recording, and wherein the variation information of the unknown audio recording is the sum of the total variation of a predetermined number of the highest variation frequency families of the unknown audio recording.
46. Computer-readable medium storing computer executable code according to claim 39 , wherein the variation information of a known recording is the sum of a predetermined number of the frequency families of the known audio recording having the highest level of variation, and wherein the variation information of the unknown recording is the sum of a predetermined number of the frequency families of the unknown audio recording having the highest level of variation.
47. Computer-readable medium storing computer executable code according to claim 39 , wherein the variation information of a known audio recording is based on a translation of a total running time of the known audio recording, and wherein the variation information of the unknown audio recording is based on a translation of a total running time of the unknown audio recording.
48. Computer-readable medium storing computer executable code according to claim 39 , wherein the variation information of the unknown audio recording is within a predetermined tolerance of the variation information of at least one of the known audio recordings.
49. Computer-readable medium storing computer executable code according to claim 39 , further including code for causing the computer processor to perform:
communicating to the source of the unknown fingerprint the audio recording identifier corresponding to the unknown fingerprint.
50. Computer-readable medium storing computer executable code according to claim 39 , further including code for causing the computer processor to perform:
recognizing the variation information of a known audio recording that is substantially similar to the variation information of the unknown audio recording.
51. A computer processor implemented to execute the instructions stored on the computer-readable medium of claim 39 .
52. A device-embedded system implemented to execute the instructions stored on the computer-readable medium of claim 39 .