Method and apparatus for encoding and decoding multiple audio channels at low bit rates using adaptive selection of encoding method

• US 5,890,125 A
• Filed: 07/16/1997
• Issued: 03/30/1999
• Est. Priority Date: 07/16/1997
• Status: Expired due to Term

• Current Assignee: Dolby Laboratories Licensing Corporation (Dolby Laboratories Incorporated)
• Original Assignee: Dolby Laboratories Licensing Corporation (Dolby Laboratories Incorporated)
• Inventors: Fellers, Matthew Conrad, Davis, Mark Franklin
• Primary Examiner(s): Hudspeth, David R.
• Assistant Examiner(s): ZINTEL, HAROLD ALBERT
• Application Number:
  US08/895,496
• Time in Patent Office: 622 Days
• Field of Search: 711/159, 704/231, 704/500, 704/216, 704/501, 381/1, 381/2
• US Class Current: 704/501
• CPC Class Codes: G10L 19/008 Multichannel audio signal coding or decoding using interchannel correlation to reduce redundancy, e.g. joint-stereo, intensity-coding or matrixing
• CPC Class Codes: G10L 19/0208 Subband vocoders
• CPC Class Codes: H04B 1/665 using psychoacoustic properties of the ear, e.g. masking effect
• CPC Class Codes: H04H 20/88 Stereophonic broadcast systems multiplex system in general H04J
• CPC Class Codes: H04R 5/04 Circuit arrangements, e.g. for selective connection of amplifier inputs/outputs to loudspeakers, for loudspeaker detection, or for adaptation of settings to personal preferences or hearing impairments combinations of amplifiers H03F3/68; stereophonic systems H04S
• CPC Class Codes: H04S 2420/03 Application of parametric coding in stereophonic audio systems

Abstract

A split-band coding system combines multiple channels of input signals into various forms of composite signals and generates spatial-characteristic signals representing soundfield spatial characteristics in a plurality of frequency subbands. The spatial-characteristics signals may be generated in either or both of two forms. In a first form, the signal represents measures of signal levels for subband signals from the input channels. In a second form, the signal represents one or more apparent directions for the soundfield. The type of the spatial-characteristics signal may be adapted dynamically in response to a variety of criteria including input signal characteristics. Temporal smoothing and spectral smoothing of the spatial-characteristics signals may be applied in an encoder. Temporal smoothing and spectral smoothing may be applied to gain factors derived from the spatial-characteristics signals in a decoder.

Claims

1. A method for generating an encoded signal by encoding a plurality of input signals each representing a respective channel of audio information, said method comprising:

generating a plurality of channel subband signals representing said input signals in a plurality of frequency subbands,

generating a composite signal representing at least a portion of the bandwidth of said input signals,

generating a spatial-characteristic signal representing spatial characteristics of a soundfield associated with said composite signal in response to respective channel subband signals in a frequency subband, wherein said spatial-characteristic signal is generated in either or both of a first form conveying information representing signal levels of said respective channel subband signals and a second form representing one or more intended directions, wherein the forms in which to generate said spatial-characteristics signal are determined adaptively, and

assembling said composite signal, said spatial-characteristics signal and an indication of the form of said spatial-characteristics signal into said encoded signal.

2. A method according to claim 1 where, in response to said respective channel subband signals in a respective frequency subband, said spatial-characteristics signal is generated in said first form if said soundfield is deemed to have a number of apparent directions greater than a threshold number.

3. A method according to claim 2 wherein said threshold number is one and said soundfield is deemed to have one apparent direction when only one of said respective channel subband signals has significant spectral energy or when all of said respective channel subband signals having significant spectral energy also have correlated amplitudes and correlated phases.

4. A method according to claim 1 wherein said respective spatial-characteristics signal is generated in said second form such that it also represents a measure of soundfield dispersion about an intended direction.

5. A method according to claim 1 wherein said spatial-characteristics signal is generated in said first form such that rates of decrease in values of said information representing said signal levels are limited to be commensurate with temporal post-masking characteristics of a human auditory system.

7. A method according to claim 1 or 6 wherein said composite signal is generated by combining two or more channel subband signals in a respective frequency subband.

8. A method according to claim 1 or 6 wherein said composite signal is a subband signal that is generated by applying a filter bank or a transform to a wideband signal, wherein said wideband signal is formed by combining two or more of said input signals.

9. A method according to claim 1 or 6 wherein said composite signal is generated by combining subband signals that are obtained by applying a filter bank or a transform to two or more of said input signals.

10. A method according to claim 1 or 6 wherein said composite signal is a parametric signal.

13. A method according to claim 1 or 6 wherein each of said respective channel subband signals is generated in response to a common time interval of said input signals, and wherein said method further comprises generating a delay signal indicating where in said common time interval an abrupt change in amplitude or direction occurs in said respective soundfield, and assembling said delay signal into said encoded signal.

14. A method according to claim 1 or 6 wherein said composite signal is a subband signal that corresponds to a respective frequency subband, and wherein said method further comprises normalizing information conveyed by said composite signal with respect to a measure of signal level for a channel subband signal in that respective frequency subband having the largest measure.

15. A method according to claim 1 or 6 that further comprises generating a differential-encoded representation of said spatial-characteristics signals corresponding to a plurality of adjacent frequency subbands, wherein said differential-encoded representation comprises one or more codes having a dynamic range that is limited according to spectral leakage characteristics between said channel subband signals in adjacent frequency subbands.

6. A method for generating an encoded signal by encoding a plurality of input signals each representing a respective channel of audio information, said method comprising:

generating a plurality of channel subband signals representing said input signals in a plurality of frequency subbands,

generating a composite signal representing at least a portion of the bandwidth of said input signals,

generating a spatial-characteristic signal representing spatial characteristics of a soundfield associated with said composite signal in response to respective channel subband signals in a frequency subband, wherein rates of change in said spatial-characteristic signal are limited to be commensurate with temporal post-masking characteristics of a human auditory system, and

assembling said composite signal and said spatial-characteristics signal into said encoded signal.

11. A method according to claim 6 wherein said spatial-characteristic signal represents one or more intended directions and rates of change in the spatial-characteristic signal are limited to control a rate at which said intended directions can change orientation.

12. A method according to claim 6 wherein said spatial-characteristic signal conveys information representing signal levels of said respective channel subband signals and said rates of change in the spatial-characteristic signal correspond to rates of decrease in values of said information representing the signal levels.

16. A method for decoding an encoded signal to generate one or more output signals for presentation via one or more output transducers, said method comprising:

obtaining from said encoded signal one or more composite signals and a plurality of spatial-characteristics signals, and deriving a plurality of composite subband signals from said one or more composite signals, wherein each spatial-characteristics signal is associated with a respective composite subband signal and represents spatial characteristics of a respective soundfield corresponding to said respective composite subband signal,

deriving from said spatial-characteristics signals a plurality of gain factors, wherein rates of change in values of said gain factors are limited to be commensurate with temporal post-masking characteristics of a human auditory system, and mapping a respective composite subband signal into one or more interim subband signals according to a respective gain factor, and

generating said plurality of output signals by applying one or more inverse filter banks to said interim subband signals.

19. A method according to claim 16 or 17 wherein said composite subband signals are derived by applying a filter bank or a transform to said one or more composite signals.

20. A method according to claim 16 or 17 wherein said composite signal is a parametric signal and said deriving comprises generating a spectral or a temporal signal in response thereto.

21. A method according to claim 16 or 17 that further comprises obtaining from said encoded signal a delay signal and delaying the mapping of said respective composite signal in response to said delay signal.

22. A method according to claim 16 or 17 wherein said interim subband signals have bandwidths that are commensurate with the critical-band bandwidths of a human auditory system.

23. A method according to claim 16 or 17 wherein said output signals are generated by applying an inverse filter bank having aliasing cancellation properties or by applying an inverse transform having aliasing cancellation properties, and wherein said mapping limits differences between levels of said interim subband signals in adjacent frequency subbands such that noise resulting from incomplete aliasing cancellation is rendered substantially inaudible.

17. A method for decoding an encoded signal to generate one or more output signals for presentation via one or more output transducers, said method comprising:

obtaining from said encoded signal one or more composite signals, a plurality of spatial-characteristics signals and an indication of the form of said spatial-characteristics signals, and deriving a plurality of composite subband signals from said one or more composite signals, wherein each spatial-characteristics signal is associated with a respective composite subband signal and represents spatial characteristics of a respective soundfield corresponding to said respective composite subband signal, wherein said indication indicates whether said spatial-characteristics signals are in a first form representing a plurality of signal levels and/or are in a second form representing one or more directions,

deriving from said spatial-characteristics signals a plurality of gain factors and mapping a respective composite subband signal into one or more interim subband signals according to a respective gain factor wherein, for said associated spatial-characteristic signal having said first form, said mapping is also according to a spatial configuration signal associated with said signal levels and a transducer configuration signal representing a spatial configuration of said output transducers, and wherein, for said associated spatial-characteristic signal having said second form, said mapping is also according to said one or more directions and said transducer configuration signal, and

generating said plurality of output signals by applying one or more inverse filter banks to said interim subband signals.

18. A method according to claim 17 wherein rates of decrease in values of said gain factors are limited to be commensurate with temporal post-masking characteristics of a human auditory system.

24. A medium readable by a machine, embodying a program of instructions executable by said machine to perform a method for generating an encoded signal by encoding a plurality of input signals each representing a respective channel of audio information, said method comprising:

generating a plurality of channel subband signals representing said input signals in a plurality of frequency subbands,

generating a composite signal representing at least a portion of the bandwidth of said input signals,

generating a spatial-characteristic signal representing spatial characteristics of a soundfield associated with said composite signal in response to respective channel subband signals in a frequency subband, wherein said spatial-characteristic signal is generated in either or both of a first form conveying information representing signal levels of said respective channel subband signals and in a second form representing one or more intended directions, wherein the forms in which to generate said spatial-characteristics signal are determined adaptively, and

assembling said composite signal, said spatial-characteristics signal and an indication of the form of said spatial-characteristics signal into said encoded signal.

25. A medium according to claim 24 wherein said method is such that, in response to said respective channel subband signals in a respective frequency subband, said spatial-characteristics signal is generated in said first form if said soundfield is deemed to have a number of apparent directions greater than a threshold number.

26. A medium according to claim 25 wherein said method is such that said threshold number is one and said soundfield is deemed to have one apparent direction when only one of said respective channel subband signals has significant spectral energy or when all of said respective channel subband signals having significant spectral energy also have correlated amplitudes and correlated phases.

27. A medium according to claim 24 wherein said method is such that said respective spatial-characteristics signal is generated in said second form such that it also represents a measure of soundfield dispersion about an intended direction.

28. A medium according to claim 24 wherein said method is such that said spatial-characteristics signal is generated in said first form such that rates of decrease in values of said information representing said signal levels are limited to be commensurate with temporal post-masking characteristics of a human auditory system.

30. A medium according to claim 24 or 29 wherein said method is such that said composite signal is generated by combining two or more channel subband signals in a respective frequency subband.

31. A medium according to claim 24 or 29 wherein said method is such that said composite signal is a subband signal that is generated by applying a filter bank or a transform to a wideband signal, wherein said wideband signal is formed by combining two or more of said input signals.

32. A medium according to claim 24 or 29 wherein said method is such that said composite signal is generated by combining subband signals that are obtained by applying a filter bank or a transform to two or more of said input signals.

33. A medium according to claim 24 or 29 wherein said method is such that said composite signal is a parametric signal.

36. A medium according to claim 24 or 29 wherein said method is such that each of said respective channel subband signals is generated in response to a common time interval of said input signals, and wherein said method further comprises generating a delay signal indicating where in said common time interval an abrupt change in amplitude or direction occurs in said respective soundfield, and assembling said delay signal into said encoded signal.

37. A medium according to claim 24 or 29 wherein said method is such that said composite signal is a subband signal that corresponds to a respective frequency subband, and wherein said method further comprises normalizing information conveyed by said composite signal with respect to a measure of signal level for a channel subband signal in that respective frequency subband having the largest measure.

38. A medium according to claim 24 or 29 wherein said method further comprises generating a differential-encoded representation of said spatial-characteristics signals corresponding to a plurality of adjacent frequency subbands, wherein said differential-encoded representation comprises one or more codes having a dynamic range that is limited according to spectral leakage characteristics between said channel subband signals in adjacent frequency subbands.

29. A medium readable by a machine, embodying a program of instructions executable by said machine to perform a method for generating an encoded signal by encoding a plurality of input signals each representing a respective channel of audio information, said method comprising:

generating a plurality of channel subband signals representing said input signals in a plurality of frequency subbands,

generating a composite signal representing at least a portion of the bandwidth of said input signals,

generating a spatial-characteristic signal representing spatial characteristics of a soundfield associated with said composite signal in response to respective channel subband signals in a frequency subband, wherein rates of change in said spatial-characteristic signal are limited to be commensurate with temporal post-masking characteristics of a human auditory system, and

assembling said composite signal and said spatial-characteristics signal into said encoded signal.

34. A medium according to claim 29 wherein said method is such that said spatial-characteristic signal represents one or more intended directions and rates of change in the spatial-characteristic signal are limited to control a rate at which said intended directions can change orientation.

35. A medium according to claim 29 wherein said method is such that said spatial-characteristic signal conveys information representing signal levels of said respective channel subband signals and said rates of change in the spatial-characteristic signal correspond to rates of decrease in values of said information representing the signal levels.

39. A medium readable by a machine, embodying a program of instructions executable by said machine to perform a method for decoding an encoded signal to generate one or more output signals for presentation via one or more output transducers, said method comprising:

obtaining from said encoded signal one or more composite signals and a plurality of spatial-characteristics signals, and deriving a plurality of composite subband signals from said one or more composite signals, wherein each spatial-characteristics signal is associated with a respective composite subband signal and represents spatial characteristics of a respective soundfield corresponding to said respective composite subband signal,

deriving from said spatial-characteristics signals a plurality of gain factors, wherein rates of change in values of said gain factors are limited to be commensurate with temporal post-masking characteristics of a human auditory system, and mapping a respective composite subband signal into one or more interim subband signals according to a respective gain factor, and

generating said plurality of output signals by applying one or more inverse filter banks to said interim subband signals.

42. A medium according to claim 39 or 40 wherein said method is such that said composite subband signals are derived by applying a filter bank or a transform to said one or more composite signals.

43. A medium according to claim 39 or 40 wherein said method is such that said composite signal is a parametric signal and said deriving comprises generating a spectral or a temporal signal in response thereto.

44. A medium according to claim 39 or 40 wherein said method further comprises obtaining from said encoded signal a delay signal and delaying the mapping of said respective composite signal in response to said delay signal.

45. A medium according to claim 39 or 40 wherein said method is such that said interim subband signals have bandwidths that are commensurate with the critical-band bandwidths of a human auditory system.

46. A medium according to claim 39 or 40 wherein said method is such that said output signals are generated by applying an inverse filter bank having aliasing cancellation properties or by applying an inverse transform having aliasing cancellation properties, and wherein said mapping limits differences between levels of said interim subband signals in adjacent frequency subbands such that noise resulting from incomplete aliasing cancellation is rendered substantially inaudible.

40. A medium readable by a machine, embodying a program of instructions executable by said machine to perform a method for decoding an encoded signal to generate one or more output signals for presentation via one or more output transducers, said method comprising:

obtaining from said encoded signal one or more composite signals, a plurality of spatial-characteristics signals and an indication of the form of said spatial-characteristics signals, and deriving a plurality of composite subband signals from said one or more composite signals, wherein each spatial-characteristics signal is associated with a respective composite subband signal and represents spatial characteristics of a respective soundfield corresponding to said respective composite subband signal, wherein said indication indicates whether said spatial-characteristics signals are in a first form representing a plurality of signal levels and/or are in a second form representing one or more directions,

deriving from said spatial-characteristics signals a plurality of gain factors and mapping a respective composite subband signal into one or more interim subband signals according to a respective gain factor wherein, for said associated spatial-characteristic signal having said first form, said mapping is also according to a spatial configuration signal associated with said signal levels and a transducer configuration signal representing a spatial configuration of said output transducers, and wherein, for said associated spatial-characteristic signal having said second form, said mapping is also according to said one or more directions and said transducer configuration signal, and

generating said plurality of output signals by applying one or more inverse filter banks to said interim subband signals.

41. A medium according to claim 40 wherein said method is such that rates of decrease in values of said gain factors are limited to be commensurate with temporal post-masking characteristics of a human auditory system.

47. An encoder for generating an encoded signal by encoding a plurality of input signals each representing a respective channel of audio information, said encoder comprising:

means for generating a plurality of channel subband signals representing said input signals in a plurality of frequency subbands,

means for generating a composite signal representing at least a portion of the bandwidth of said input signals,

means for generating a spatial-characteristic signal representing spatial characteristics of a soundfield associated with said composite signal in response to respective channel subband signals in a frequency subband, wherein said spatial-characteristic signal is generated in either or both of a first form conveying information representing signal levels of said respective channel subband signals and in a second form representing one or more intended directions, wherein the forms in which to generate said spatial-characteristics signal are determined adaptively, and

means for assembling said composite signal, said spatial-characteristics signal and an indication of the form of said spatial-characteristics signal into said encoded signal.

48. An encoder according to claim 47 where, in response to said respective channel subband signals in a respective frequency subband, said spatial-characteristics signal is generated in said first form if said soundfield is deemed to have a number of apparent directions greater than a threshold number.

49. An encoder according to claim 48 wherein said threshold number is one and said soundfield is deemed to have one apparent direction when only one of said respective channel subband signals has significant spectral energy or when all of said respective channel subband signals having significant spectral energy also have correlated amplitudes and correlated phases.

50. An encoder according to claim 47 wherein said respective spatial-characteristics signal is generated in said second form such that it also represents a measure of soundfield dispersion about an intended direction.

51. An encoder according to claim 47 wherein said spatial-characteristics signal is generated in said first form such that rates of decrease in values of said information representing said signal levels are limited to be commensurate with temporal post-masking characteristics of a human auditory system.

53. An encoder according to claim 47 or 52 further comprising means for combining two or more channel subband signals in a respective frequency subband to generate said composite signal.

54. An encoder according to claim 47 or 52 further comprising means for generating a wideband signal by combining two or more of said input signals, and means for generating said composite signal by applying a filter bank or a transform to said wideband signal.

55. An encoder according to claim 47 or 52 further comprising means for applying a filter bank or a transform to two or more of said input signals and means for generating said composite signal by combining subband signals that are obtained by applying said filter bank or said transform.

56. An encoder according to claim 47 or 52 wherein said composite signal is a parametric signal.

59. An encoder according to claim 47 or 52 wherein each of said respective channel subband signals is generated in response to a common time interval of said input signals, and wherein said encoder further comprises means for generating a delay signal indicating where in said common time interval an abrupt change in amplitude or direction occurs in said respective soundfield, and means for assembling said delay signal into said encoded signal.

60. An encoder according to claim 47 or 52 wherein said composite signal is a subband signal that corresponds to a respective frequency subband, and wherein said encoder further comprises means for normalizing information conveyed by said composite signal with respect to a measure of signal level for a channel subband signal in that respective frequency subband having the largest measure.

61. An encoder according to claim 47 or 52 further comprising means for generating a differential-encoded representation of said spatial-characteristics signals corresponding to a plurality of adjacent frequency subbands, wherein said differential-encoded representation comprises one or more codes having a dynamic range that is limited according to spectral leakage characteristics between said channel subband signals in adjacent frequency subbands.

52. An encoder for generating an encoded signal by encoding a plurality of input signals each representing a respective channel of audio information, said encoder comprising:

means for generating a plurality of channel subband signals representing said input signals in a plurality of frequency subbands,

means for generating a composite signal representing at least a portion of the bandwidth of said input signals,

means for generating a spatial-characteristic signal representing spatial characteristics of a soundfield associated with said composite signal in response to respective channel subband signals in a frequency subband, wherein rates of change in said spatial-characteristic signal are limited to be commensurate with temporal post-masking characteristics of a human auditory system, and

means for assembling said composite signal and said spatial-characteristics signal into said encoded signal.

57. An encoder according to claim 52 wherein said spatial-characteristic signal represents one or more intended directions and rates of change in the spatial-characteristic signal are limited to control a rate at which said intended directions can change orientation.

58. An encoder according to claim 52 wherein said spatial-characteristic signal conveys information representing signal levels of said respective channel subband signals and said rates of change in the spatial-characteristic signal correspond to rates of decrease in values of said information representing the signal levels.

62. A decoder for decoding an encoded signal to generate one or more output signals for presentation via one or more output transducers, said decoder comprising:

means for obtaining from said encoded signal one or more composite signals and a plurality of spatial-characteristics signals, and for deriving a plurality of composite subband signals from said one or more composite signals, wherein each spatial-characteristics signal is associated with a respective composite subband signal and represents spatial characteristics of a respective soundfield corresponding to said respective composite subband signal,

means for deriving from said spatial-characteristics signals a plurality of gain factors, wherein rates of change in values of said gain factors are limited to be commensurate with temporal post-masking characteristics of a human auditory system, and for mapping a respective composite subband signal into one or more interim subband signals according to a respective gain factor, and

means for generating said plurality of output signals by applying one or more inverse filter banks to said interim subband signals.

65. A decoder according to claim 62 or 63 further comprising means for applying a filter bank or a transform to said one or more composite signals to derive said composite subband signals.

66. A decoder according to claim 62 or 63 wherein said composite signal is a parametric signal and said means for deriving is also for generating a spectral or a temporal signal in response thereto.

67. A decoder according to claim 62 or 63 further comprising means for obtaining from said encoded signal a delay signal and for delaying the mapping of said respective composite signal in response to said delay signal.

68. A decoder according to claim 62 or 63 wherein said interim subband signals have bandwidths that are commensurate with the critical-band bandwidths of a human auditory system.

69. A decoder according to claim 62 or 63 further comprising means for applying an inverse filter bank having aliasing cancellation properties or an inverse transform having aliasing cancellation properties to generate said output signals, wherein said means for mapping limits differences between levels of said interim subband signals in adjacent frequency subbands such that noise resulting from incomplete aliasing cancellation is rendered substantially inaudible.

63. A decoder for decoding an encoded signal to generate one or more output signals for presentation via one or more output transducers, said decoder comprising:

means for obtaining from said encoded signal one or more composite signals, a plurality of spatial-characteristics signals and an indication of the form of said spatial-characteristics signals, and for deriving a plurality of composite subband signals from said one or more composite signals, wherein each spatial-characteristics signal is associated with a respective composite subband signal and represents spatial characteristics of a respective soundfield corresponding to said respective composite subband signal, wherein said indication indicates whether said spatial-characteristics signals are in a first form representing a plurality of signal levels and/or are in a second form representing one or more directions,

means for deriving from said spatial-characteristics signals a plurality of gain factors and for mapping a respective composite subband signal into one or more interim subband signals according to a respective gain factor wherein, for said associated spatial-characteristic signal having said first form, said mapping is also according to a spatial configuration signal associated with said signal levels and a transducer configuration signal representing a spatial configuration of said output transducers, and wherein, for said associated spatial-characteristic signal having said second form, said mapping is also according to said one or more directions and said transducer configuration signal, and

means for generating said plurality of output signals by applying one or more inverse filter banks to said interim subband signals.

64. A decoder according to claim 63 wherein rates of decrease in values of said gain factors are limited to be commensurate with temporal post-masking characteristics of a human auditory system.