Hearing aids based on models of cochlear compression using adaptive compression thresholds

• US 6,970,570 B2
• Filed: 08/23/2001
• Issued: 11/29/2005
• Est. Priority Date: 09/22/1998
• Status: Expired due to Term

• Current Assignee: Hearing Emulations, LLC
• Original Assignee: Hearing Emulations, LLC
• Inventors: Goldstein, Julius L.
• Primary Examiner(s): Pendleton, Brian T.
• Application Number:
  US09/935,510
• Publication Number:
  US 20020057808A1
• Time in Patent Office: 1,559 Days
• Field of Search: 381/106, 381/321, 381/312, 381/104, 381/107, 381/320
• US Class Current: 381/321
• CPC Class Codes: H04R 2225/67 Implantable hearing aids or parts thereof not covered by H04R25/606
• CPC Class Codes: H04R 25/356 Amplitude, e.g. amplitude shift or compression
• CPC Class Codes: H04R 25/502 using analog signal processing
• CPC Class Codes: H04R 25/70 Adaptation of deaf aid to hearing loss, e.g. initial electronic fitting

Abstract

A hearing aid device providing instantaneous gain compression for sound signals and adaptive control of nonlinear waveform distortion, the device comprising: (a) at least one bandpass nonlinearity (BPNL) amplifier comprising a first bandpass filter, a second bandpass filter, and a memoryless nonlinear (MNL) compressive audio amplifier configured to receive a sound signal from the first bandpass filter and provide an MNL compressive audio amplifier output to the second bandpass filter, wherein the MNL compressive audio amplifier is configured to produce the MNL compressive audio amplifier output by providing memoryless gain compression directly on a sound signal that is (1) received from the first bandpass filter and (2) exhibits instantaneous amplitudes greater than a compression threshold, the BPNL amplifier thereby producing a desired gain compression on the received sound signal at an output of the second bandpass filter, and (b) a controller in communication with the BPNL amplifier, the controller being configured to adjust the compression threshold of the MNL compressive audio amplifier. Adjustment of the compression threshold in each BPNL amplifier may be achieved at least partially in response to a user input and/or to sound signal changes. By adaptively controlling the compression threshold, performance of the device can by optimized to match its environment.

Claims

1. In a hearing amplification device adapted to receive a sound signal, the hearing amplification device having at least one bandpass non-linearity (BPNL) channel configured to receive an input representative of said sound signal, the improvement comprising:

said channel being further configured with a memoryless nonlinear amplifier to provide (1) linear gain for an input representative of a portion of a subband of said sound signal having an instantaneous sound level less than a compression threshold, (2) instantaneous compressive gain for an input representative of a portion of a subband of said sound signal having an instantaneous sound level greater than said compression threshold, wherein said instantaneous compressive gain is less than said linear gain, and (3) control of said compression threshold that is adaptive at least partially in response to changes in said sound signal, and

wherein said at least one channel is configured to have its compression threshold initially set to a predetermined quiescent level, and wherein said at least one channel is further configured to adjust said compression threshold such that said compression threshold is in a range of about said predetermined quiescent level to about 20 decibels below an average sound level of at least a portion of said sound signal.

2. The hearing amplification device of claim 1 wherein said at least one channel is further configured to provide a smooth transition between said linear gain and said instantaneous compressive gain.

3. The hearing amplification device of claim 1 wherein said at least one channel is further configured to provide a sharp transition between said linear gain and said instantaneous compressive gain.

4. The hearing amplification device of claim 1 wherein said at least one channel is further configured to provide (1) decompression for an input representative of a portion of a subband of said sound signal having an instantaneous sound level greater than a decompression threshold, said decompression threshold being greater than said compression threshold.

5. The hearing amplification device of claim 4 wherein said at least one channel is configured to have its compression threshold initially set to a predetermined quiescent level, and wherein said at least one channel is further configured to adjust said compression threshold such that said compression threshold is in a range of about said predetermined quiescent level to about said decompression threshold.

6. The hearing amplification device of claim 4 wherein said at least one channel is further configured to provide attenuation for an input representative of a portion of a subband of said sound signal having an instantaneous sound level greater than an attenuation threshold, said attenuation threshold being greater than said decompression threshold.

7. The hearing amplification device of claim 1 wherein said at least one channel is further configured to provide said instantaneous compressive gain as at least square root compression.

8. The hearing amplification device of claim 1 wherein said at least one channel is further configured to adjust said compression threshold at least partially in response to a user input.

9. The hearing amplification device of claim 1 having a plurality of said channels, each of said channels being responsive to an input representative of an audio frequency range different from other channels.

10. The hearing amplification device of claim 9 wherein each channel is configured to have its compression threshold initially set independently of each other channel, and wherein each channel is further configured to adjust its compression threshold at least partially in response to changes in said sound signal.

11. The hearing amplification device of claim 10 wherein each channel is configured to independently adjust its compression threshold at least partially in response to changes in said sound signal.

12. The hearing amplification device of claim 10 wherein each channel is further configured to provide (1) decompression for an input representative of a portion of a subband of said sound signal having an instantaneous sound level greater than a decompression threshold, said decompression threshold being greater than said compression threshold.

13. In a hearing amplification device adapted to receive a sound signal, the hearing amplification device having at least one bandpass non-linearity (BPNL) channel configured to receive an input representative of said sound signal, the improvement comprising:

said channel being further configured with a memoryless nonlinear amplifier to provide (1) linear gain for an input representative of a portion of a subband of said sound signal having an instantaneous sound level less than a compression threshold, (2) instantaneous compressive gain for an input representative of a portion of a subband of said sound signal having an instantaneous sound level greater than said compression threshold, wherein said instantaneous compressive gain is less than said linear gain, and (3) control of said compression threshold that is adaptive at least partially in response to changes in said sound signal, and

wherein said at least one channel is configured to adjust said compression threshold such that said compression threshold is within a range of about 5 decibels below an average sound level of at least a portion of said sound signal to about 5 decibels above said average sound level.

14. In a hearing amplification device adapted to receive a sound signal, the hearing amplification device having at least one bandpass non-linearity (BPNL) channel configured to receive an input representative of said sound signal, the improvement comprising:

a plurality of channels, each of said channels being responsive to an input representative of an audio frequency range different from other channels, each channel being further configured with a memoryless nonlinear amplifier to provide (1) linear gain for an input representative of a portion of a subband of said sound signal having an instantaneous sound level less than a compression threshold, (2) instantaneous compressive gain for an input representative of a portion of a subband of said sound signal having an instantaneous sound level greater than said compression threshold, wherein said instantaneous compressive gain is less than said linear gain, and (3) adaptive control of said compression threshold;

wherein each channel is further configured to have its compression threshold initially set independently of each other channel;

wherein each channel is further configured to adjust its compression threshold at least partially in response to changes in said sound signal;

wherein each channel is further configured to provide (1) decompression for an input representative of a portion of a subband of said sound signal having an instantaneous sound level greater than a decompression threshold, said decompression threshold being greater than said compression threshold; and

wherein each channel is further configured to (1) have its compression threshold initially set to a predetermined quiescent level, and (2) adjust its compression threshold such that its compression threshold is in a range of about its predetermined quiescent level to about said decompression threshold.

15. The hearing amplification device of claim 14 wherein each channel is further configured to adjust its compression threshold such that its compression threshold is in a range of about said predetermined quiescent level to about 20 decibels below an average sound level of at least a portion of said sound signal.

16. The hearing amplification device of claim 14 wherein each channel is configured to adjust its compression threshold such that its compression threshold is within a range of about 5 decibels below an average sound level of at least a portion of said sound signal to about 5 decibels above said average sound level.

17. A method of compensating for impaired hearing, said method comprising:

linearly amplifying an input corresponding to a portion of a subband of a sound signal having an instantaneous sound level less than a compression threshold;

instantaneously compressively amplifying an input corresponding to a portion of a subband of a sound signal having an instantaneous sound level greater than said compression threshold;

initially setting said compression threshold at a predetermined quiescent level;

storing a previously determined peak value for a previous portion of said sound signal;

determining a current peak value for a current portion of said sound signal; and

adaptively controlling said compression threshold at least partially in response to changes in said sound signal; and

wherein said adaptively controlling step includes increasing said compression threshold when said determined current peak value is greater than said stored peak value.

18. The method of claim 17 wherein said adaptively controlling step includes:

maintaining said compression threshold at its current level when said determined current peak value does not deviate by more than a predetermined triggering amount from said stored peak value; and

decreasing said compression threshold when said determined current peak value deviates by more than said predetermined triggering amount from said stored peak value.

19. The method of claim 18 wherein said decreasing step includes:

decreasing said compression threshold when said determined current peak value has continuously deviated by more than said predetermined triggering amount from said stored peak value for a predetermined amount of time.

20. The method of claim 19 wherein said adaptively controlling step further includes:

setting a maximum value for said compression threshold;

setting a minimum value for said compression threshold;

estimating an average sound level for at least a portion of said sound signal from said determined current peak value;

wherein said increasing step includes increasing said compression threshold to substantially match said estimated average sound level when said estimated average sound level is less than or equal to said maximum value and increasing said compression threshold to equal said maximum value when said estimated average sound level is greater than said maximum value; and

wherein said decreasing step includes decreasing said compression threshold by a fixed amount when said compression threshold minus said fixed amount would not be less than said minimum value and decreasing said compression threshold to equal said minimum value when said compression threshold minus said fixed amount would be less than said minimum value.

21. The method of claim 20 further comprising:

providing decompression for an input corresponding to a portion of a subband of a sound signal having an instantaneous sound level greater than said decompression threshold, wherein said decompression threshold is greater than said compression threshold.

22. The method of claim 21 wherein said step of setting said maximum value includes setting said maximum value as said decompression threshold, and wherein said step of setting said minimum value includes setting said minimum value as said predetermined quiescent level.

23. The method of claim 17 further comprising:

providing decompression for an input corresponding to a portion of a subband of a sound signal having an instantaneous sound level greater than said decompression threshold, wherein said decompression threshold is greater than said compression threshold.

24. The method of claim 23 further comprising:

attenuating an input corresponding to a portion of a subband of a sound signal having an instantaneous sound level greater than an attenuation threshold, wherein said attenuation threshold is greater than said decompression threshold.

25. The method of claim 17 further comprising:

performing each of said steps for a plurality of different audio frequency ranges.

26. A bandpass non-linearity (BPNL) hearing amplification device adapted to receive and amplify a sound signal, said device comprising:

a memoryless transducer for processing a transducer input according to a transfer function to thereby produce a transducer output, said transducer input being representative of a subband of a sound signal, said transducer output being representative of an amplified sound signal, said transfer function being configured to provide (1) linear gain for a transducer input representative of a portion of a subband of said sound signal having an instantaneous sound level less than a compression threshold, and (2) instantaneous compressive gain for a transducer input representative of a portion of a subband of said sound signal having an instantaneous sound level greater than said compression threshold, wherein said instantaneous compressive gain is less than said linear gain, wherein said compression threshold is initially set to a predetermined quiescent level; and

a compression threshold controller coupled to said transducer for adjusting said compression threshold at least partially in response to changes in said sound signal, and wherein said controller is configured to;

estimate an average sound level for at least a portion in time of said sound signal; and

adjust said compression threshold in a range of about said predetermined quiescent level to about 20 decibels below said estimated average sound level.

27. The device of claim 26 wherein said transfer function is configured to obey odd symmetry.

28. The device of claim 26 wherein said transfer function is further configured to provide a smooth transition between said linear gain and said instantaneous compressive gain.

29. The device of claim 26 wherein said transfer function is further configured to provide a sharp transition between said linear gain and said instantaneous compressive gain.

30. The device of claim 26 wherein said transfer function is further configured to provide decompression for a transducer input representative of a portion of a subband of said sound signal having an instantaneous sound level greater than a decompression threshold, wherein said decompression threshold is greater then said compression threshold.

31. The device of claim 30 wherein said transfer function is further configured to provide attenuation for a transducer input representative of a portion of a subband of said sound signal having an instantaneous sound level greater than an attenuation threshold, wherein said attenuation threshold is greater than said decompression threshold.

32. The device of claim 26 further comprising a plurality of said transducers and a plurality of channels, each channel being responsive to a different predetermined channel frequency range and comprising one of said transducers, and wherein said controller is coupled to each of said transducers, wherein said transducer input for each transducer is representative of those frequency components of said sound signal that are within its corresponding predetermined channel frequency range.

33. The device of claim 32 wherein each of said channels and said controller are implemented in a digital signal processor, each transducer input being a digital representation of said subband sound signal.

34. The device of claim 33 wherein said digital signal processor is a multirate digital signal processor.

35. The device of claim 32 wherein each of said channels and said controller are implemented in a plurality of analog components.

36. The device of claim 26 wherein said transducer and said controller are implemented in a digital signal processor, said transducer input being a digital representation of said subband sound signal.

37. The device of claim 36 wherein said digital signal processor is a multirate digital signal processor.

38. The device of claim 26 wherein said transducer and said controller are implemented in a plurality of analog components.

39. A bandpass non-linearity (BPNL) hearing amplification device adapted to receive and amplify a sound signal, said device comprising:

a memoryless transducer for processing a transducer input according to a transfer function to thereby produce a transducer output, said transducer input being representative of a subband of a sound signal, said transducer output being representative of an amplified sound signal, said transfer function being configured to provide (1) linear gain for a transducer input representative of a portion of a subband of said sound signal having an instantaneous sound level less than a compression threshold, and (2) instantaneous compressive gain for a transducer input representative of a portion of a subband of said sound signal having an instantaneous sound level greater than said compression threshold, wherein said instantaneous compressive gain is less than said linear gain; and

a compression threshold controller coupled to said transducer for adjusting said compression threshold at least partially in response to changes in said sound signal, and wherein said controller is configured to;

estimate an average sound level for at least a portion in time of said sound signal; and

adjust said compression threshold in a range of about 5 decibels below said estimated average sound level to about 5 decibels above said estimated average sound level.

40. A bandpass non-linearity (BPNL) hearing amplification device adapted to receive and amplify a sound signal, said device comprising:

a memoryless transducer for processing a transducer input according to a transfer function to thereby produce a transducer output, said transducer input being representative of a subband of a sound signal, said transducer output being representative of an amplified sound signal, said transfer function being configured to provide (1) linear gain for a transducer input representative of a portion of a subband of said sound signal having an instantaneous sound level less than a compression threshold, and (2) instantaneous compressive gain for a transducer input representative of a portion of a subband of said sound signal having an instantaneous sound level greater than said compression threshold, wherein said instantaneous compressive gain is less than said linear gain; and

a compression threshold controller coupled to said transducer for adjusting said compression threshold at least partially in response to changes in said sound signal; and

wherein an asymptotic representation of said transfer function TA1 is defined by the general formula;

TA=TA(u,A,U,p),

wherein for |u|<

U;

TA(u,A,U,p)=Au

wherein for |u|>

U $\mathrm{TA}<br><br>\left(u,A,U,p\right)=\mathrm{sgn}<br><br>\left(u\right)<br><br>\mathrm{AU}|\frac{u}{U}<br><br>{|}^p$

wherein;

TA1=TA1(u,U_c)=TA(u,A(U_c),U_c(Y),p);

wherein U_c(Y)=U₁ for Y less than U₁ and U_c(Y)=Y for Y greater than or equal to U₁, wherein U₁ represents a quiescent level for said compression threshold, wherein U_c represents an adjusted compression threshold, wherein Y represents a control signal from said controller for controlling said compression threshold, wherein u represents said transducer input, wherein p represents a compression power, and wherein A represents a magnitude of gain, wherein for Y less than U₁;

A=G₁

and wherein for Y greater than or equal to U₁;

$A=G_1<br><br>{<br><br>\frac{U_1}{U_C}<br><br>}^{1-p}$

wherein G₁ represents the magnitude of said linear gain.

41. A bandpass non-linearity (BPNL) hearing amplification device adapted to receive and amplify a sound signal, said device comprising:

a memoryless transducer for processing a transducer input according to a transfer function to thereby produce a transducer output, said transducer input being representative of a subband of a sound signal, said transducer output being representative of an amplified sound signal, said transfer function being configured to provide (1) linear gain for a transducer input representative of a portion of a subband of said sound signal having an instantaneous sound level less than a compression threshold, and (2) instantaneous compressive gain for a transducer input representative of a portion of a subband of said sound signal having an instantaneous sound level greater than said compression threshold, wherein said instantaneous compressive gain is less than said linear gain; and

a compression threshold controller coupled to said transducer for adjusting said compression threshold at least partially in response to changes in said sound signal;

wherein said transfer function is further configured to provide decompression for a transducer input representative of a portion of a subband of said sound signal having an instantaneous sound level greater than a decompression threshold, wherein said decompression threshold is greater than said compression threshold; and

wherein an asymptotic representation of said transfer function is defined as a cascade of two functions TA1 and TA2, wherein both TA1 and TA2 are defined the general formula;

TA=TA(u,A,U,p),

wherein for |u|<

U;

TA(u,A,U,p)=Au

wherein for |u|>

U $\mathrm{TA}<br><br>\left(u,A,U,p\right)=\mathrm{sgn}<br><br>\left(u\right)<br><br>\mathrm{AU}|\frac{u}{U}<br><br>{|}^p$

wherein;

TA1=TA1(u,U_c)=TA(u,A(U_c),U_c(Y),p);

wherein U_c(Y)=U₁ for Y less than U₁, U_c(Y)=Y for Y greater than or equal to U₁ and less than or equal to U₂, and U_c(Y)=U₂ for Y greater than U₂, wherein U₁ represents a quiescent level for said compression threshold, wherein U₂ represents said decompression threshold, wherein U_c represents an adjusted compression threshold, wherein Y represents a control signal from said controller for controlling said compression threshold, wherein u represents said transducer input, wherein p represents a compression power, and wherein A represents a magnitude of gain, wherein for Y less than U₁;

A=G₁

and wherein for Y greater than or equal to U₁;

$A=G_1<br><br>{<br><br>\frac{U_1}{U_c}<br><br>}^{1-p}$

wherein G₁ represents the magnitude of said linear gain; and

wherein for TA2=TA2(u)=TA(u,1,U₂,p₂), wherein u represents TA1, wherein U₂ represents said decompression threshold, and wherein 1≦

p₂≦

1/p.

42. A bandpass non-linearity (BPNL) hearing amplification device adapted to receive and amplify a sound signal, said device comprising:

a memoryless transducer for processing a transducer input according to a transfer function to thereby produce a transducer output, said transducer input being representative of a subband of a sound signal, said transducer output being representative of an amplified sound signal, said transfer function being configured to provide (1) linear gain for a transducer input representative of a portion of a subband of said sound signal having an instantaneous sound level less than a compression threshold, and (2) instantaneous compressive gain for a transducer input representative of a portion of a subband of said sound signal having an instantaneous sound level greater than said compression threshold, wherein said instantaneous compressive gain is less than said linear gain; and

a compression threshold controller coupled to said transducer for adjusting said compression threshold at least partially in response to changes in said sound signal;

wherein said transfer function is further configured to provide decompression for a transducer input representative of a portion of a subband of said sound signal having an instantaneous sound level greater than a decompression threshold, wherein said decompression threshold is greater than said compression threshold;

wherein said transfer function is further configured to provide attenuation for a transducer input representative of a portion of a subband of said sound signal having an instantaneous sound level greater than an attenuation threshold, wherein said attenuation threshold is greater than said decompression threshold; and

wherein an asymptotic representation of said transfer function is defined as a cascade of three functions TA1, TA2, and TA3, wherein TA1, TA2, and TA3 are each defined the general formula;

TA=TA(u,A,U,p),

wherein for |u|<

U;

TA(u,A,U,p)=Au

wherein for |u|>

U $\mathrm{TA}<br><br>\left(u,A,U,p\right)=\mathrm{sgn}<br><br>\left(u\right)<br><br>\mathrm{AU}<br><br>{<br><br>\frac{u}{U}<br><br>}^p$

wherein;

TA1=TA1(u,U_c)=TA(u,A(U_c),U_c(Y),p);

wherein U_c(Y)=U₁ for Y less than U₁, U_c(Y)=Y for Y greater than or equal to U₁ and less than or equal to U₂, and U_c(Y)=U₂ for Y greater than U₂, wherein U₁ represents a quiescent level for said compression threshold, wherein U₂ represents said decompression threshold, wherein U_c represents an adjusted compression threshold, wherein Y represents a control signal from said controller for controlling said compression threshold, wherein u represents said transducer input, wherein p represents a first compression power, and wherein A represents a magnitude of gain, wherein for Y less than U₁;

A=G₁

and wherein for Y greater than or equal to U₁;

$A=G_1<br><br>{<br><br>\frac{U_1}{U_c}<br><br>}^{1-p}$

wherein G₁ represents the magnitude of said linear gain;

wherein for TA2=TA2(u)=TA(u,1,U₂,p₂), wherein u represents TA1 or TA3, wherein U₂ represents said decompression threshold, and wherein 1≦

p₂≦

1/p; and

wherein for TA3=TA3(u)=TA(u,1,U₃,p₃), u represents TA1 or TA2, wherein U₃ represents said attenuation threshold, and wherein p₃ represents a second compression power.

43. A bandpass non-linearity (BPNL) hearing amplification device adapted to receive and amplify a sound signal, said device comprising:

a memoryless transducer for processing a transducer input according to a transfer function to thereby produce a transducer output, said transducer input being representative of a subband of a sound signal, said transducer output being representative of an amplified sound signal, said transfer function being configured to provide (1) linear gain for a transducer input representative of a portion of a subband of said sound signal having an instantaneous sound level less than a compression threshold, and (2) instantaneous compressive gain for a transducer input representative of a portion of a subband of said sound signal having an instantaneous sound level greater than said compression threshold, wherein said instantaneous compressive gain is less than said linear gain; and

a compression threshold controller coupled to said transducer for adjusting said compression threshold at least partially in response to changes in said sound signal, wherein said controller is configured with;

a first operating mode in which said controller is configured to not adjust said compression threshold; and

a second operating mode in which said controller is configured to adjust said compression threshold at least partially in response to changes in said sound signal; and

wherein said controller is switchable between said first operating mode and said second operating mode.

44. The device of claim 43 wherein said controller is further configured with:

a third operating mode in which said controller is configured to fix said compression threshold at a current level;

wherein said controller is switchable between said first operating mode, said second operating mode, and said third operating mode.

45. The device of claim 44 wherein said controller is configured to switch between said first operating mode, said second operating mode, and said third operating mode at least partially in response to a user input.

46. A method of diagnosing an extent and form of hearing impairment, said method comprising:

determining an amount of low level gain G₁ needed by a patient in a plurality of different audio frequency ranges for sound signals having a low sound level;

selecting a compression power p;

adjusting a hearing amplification device having a plurality of channels corresponding said audio frequency ranges to provide the determined low level gain G₁ for each channel and selected compression power p, said hearing amplification device being configured to process an input signal representative of a sound signal according to a merging family of channel transducer characteristics to create an amplified signal, said characteristics defined by (1) linear gain for input signals representative of a sound signal having a sound level less than a compression threshold, (2) rapid compressive gain for input signals representative of a sound signal having a sound level greater than a compression threshold;

presenting a sound signal at an input of the hearing amplification device to generate an amplified signal therefrom;

providing to the patient the amplified signal generated from said presented sound signal; and

adjusting the values of said compression threshold for low level gain G₁ for each channel and said compression power p until the patient communicates that he/she has perceived satisfactory results.

47. The method of claim 46 wherein said rapid compressive gain is instantaneous compressive gain.

48. A hearing aid device providing instantaneous gain compression for sound signals and adaptive control of nonlinear waveform distortion, the device comprising:

at least one bandpass nonlinearity (BPNL) amplifier comprising a first bandpass filter, a second bandpass filter, and a memoryless nonlinear (MNL) compressive audio amplifier configured to receive a sound signal from the first bandpass filter and provide an MNL compressive audio amplifier output to the second bandpass filter, wherein the MNL compressive audio amplifier is configured to produce the MNL compressive audio amplifier output by providing memoryless gain compression directly on a sound signal that is (1) received from the first bandpass filter and (2) exhibits instantaneous amplitudes greater than a compression threshold, the BPNL amplifier thereby producing a desired gain compression on the received sound signal at an output of the second bandpass filter; and

a controller in communication with the at least one BPNL amplifier, the controller being configured to adjust the compression threshold of the MNL compressive audio amplifier.

49. The device of claim 48 wherein the controller is further configured to adjust the compression threshold from a quiescent compression threshold at least partially in response to changes in the received sound signal.

50. The device of claim 49 wherein the controller is further configured to adjust the compression threshold from the quiescent compression threshold at least partially in response to changes in an estimate of a RMS value for the received sound signal.

51. The device of claim 49 wherein the controller is further configured to control waveform distortion caused by the MNL compressive audio amplifier by adjusting the compression threshold from the quiescent compression threshold at least partially in response to changes in the received sound signal.

52. The device of claim 49 wherein the MNL compressive audio amplifier is further configured to provide BPNL gain compression for a tone within its passband via an MNL transducer input/output (I/O) function that is in close approximation to a desired RMS I/O function.

53. The device of claim 52 wherein the transducer function is configured as an odd symmetric function f(u)=−

f(−

u), wherein u is a sample value of the sound signal processed by the MNL compressive audio amplifier.

54. The device of claim 52 wherein the transducer function is further configured to provide a constant gain for received sound amplitudes below the compression threshold, wherein the compressive gain is less than the constant gain.

55. The device of claim 54 wherein the transducer function is further configured to provide decompression for received sound amplitudes greater than a decompression threshold, the decompression threshold being greater than the compression threshold.

56. The device of claim 54 wherein the controller is further configured to adjust the compression threshold to a value within a range between the quiescent compression threshold and the decompression threshold.

57. The device of claim 56 further comprising a plurality of the BPNL amplifiers, each BPNL amplifier being configured to process a different frequency subband of a broadband sound signal.

58. The device of claim 57 wherein the controller is further configured to adjust the compression threshold for each subband to a value within the range between the quiescent compression threshold and the decompression threshold at least partially in response to an estimate of a short-term RMS amplitude for the broadband sound signal.

59. The device of claim 56 wherein the transducer function is further configured to provide protective attenuation for received sound amplitudes greater than an attenuation threshold, the attenuation threshold being greater than the decompression threshold.

60. The device of claim 49 wherein the controller is further configured to upwardly adjust the compression threshold if the received sound signal exhibits an estimated RMS amplitude greater than a specified amount.

61. The device of claim 60 wherein the controller is further configured to downwardly adjust the compression threshold if the received sound signal exhibits an estimated RMS amplitude below a triggering amount for a duration of time greater than a predetermined release time.

62. The device of claim 61 wherein the triggering amount is derived from a fraction of an estimated RMS amplitude for a most recently received sound signal that caused an upward adjustment of the compression threshold.

63. The device of claim 61 wherein the controller is further configured to maintain the compression threshold and the triggering amount at its current levels if the received sound signal does not maintain an estimated short-term RMS amplitude below the triggering amount throughout the release time duration.

64. The device of claim 48 further comprising a plurality of the BPNL amplifiers, each BPNL amplifier being configured to process a different frequency subband of a broadband sound signal.

65. The device of claim 64 wherein the controller is further configured to coordinate adjustment of the compression thresholds for each BPNL amplifier to produce a desired nonlinear waveform quality.

66. The device of claim 48 wherein the controller is further configured to fix the compression threshold at a current level in response to user input.

67. The device of claim 48 wherein the controller is further configured to adjust the compression threshold between at least (1) a first level associated with a quiet environment and (2) a second level associated with a noisy environment.

68. The device of claim 67 wherein the controller is further configured to adjust the compression threshold between the at least first and second levels at least partially in response to changes in the received sound signal.

69. The device of claim 67 wherein the controller is further configured to adjust the compression threshold between the at least first and second levels at least partially in response to user input.

70. The device of claim 48 wherein the at least one BPNL amplifier is a digital BPNL amplifier.

71. The device of claim 48 wherein the at least one BPNL amplifier is an analog BPNL amplifier.