Systems, methods, and apparatus for equalization preference learning

Abstract

Systems, methods, and apparatus are provided for equalization preference learning for digital audio modification. A method for listener calibration of an audio signal includes modifying a reference sound using at least one equalization curve; playing the modified reference sound for a listener; accepting listener feedback regarding the modified reference sound; and generating a weighting function based on listener feedback. A listener audio configuration system includes an output providing a sound for listener review; an interface accepting listener feedback regarding the sound; and a processor programming an audio device based on listener feedback.

The invention claimed is: 
     
       1. A method for listener calibration of an audio signal, said method comprising:
 generating a plurality of sound/descriptor pairs, each sound/descriptor pair including a sound and an audio descriptor for the sound;   determining a weighting function based on the plurality of sound/descriptor pairs;   scaling the weighting function according to listener feedback;   applying the scaled weighting function as an audio device setting for an audio signal; and   mapping one or more language based descriptors associated with the plurality of sound/descriptor pairs to one or more frequency gain curves,   wherein the one or more frequency gain curves are mapped to the weighting function, wherein a slope of a line fitted between the values of the plurality of sound/descriptor pairs is computed for each sound/descriptor pair and wherein a combination of the plurality of slopes represents the weighting function.   
     
     
       2. The method of  claim 1 , further comprising receiving listener feedback regarding the weighting function via a controller. 
     
     
       3. The method of  claim 1 , further comprising modifying one or more sounds associated with the plurality of sound/descriptor pairs using at least one equalization curve. 
     
     
       4. The method of  claim 1 , further comprising receiving listener feedback regarding a plurality of high and low frequency gain comparisons to converge to a desired scaled weighting function. 
     
     
       5. A tangible computer readable medium comprising computer program code which, when executed by a processor, implements an audio configuration system for individual learning and customization of a listener's subjective audio preference, said system comprising:
 an output to provide a sound for listener review;   an interface to accept listener feedback regarding the sound, the feedback including a descriptor to associate with the sound; and   a processor to program an audio device based on listener feedback, wherein the processor is to analyze the sound and associated listener feedback to generate a weighting function, the weighting function further scalable based on listener feedback, wherein the processor is to apply the weighting function as an audio device setting for the audio device to apply to an audio signal and the processor is to predict listener ratings based on a similarity between a probe curve and the weighting function.   
     
     
       6. The computer readable medium of  claim 5 , wherein the interface further comprises a slider controller to receive listener feedback regarding the weighting function. 
     
     
       7. The computer readable medium of  claim 5 , wherein the processor is to modify the sound using at least one equalization curve. 
     
     
       8. The computer readable medium of  claim 5 , wherein the processor is to map a language-based descriptor associated with the sound to a frequency gain curve. 
     
     
       9. The computer readable medium of  claim 8 , wherein the processor is to map the frequency gain curve to the weighting function. 
     
     
       10. The computer readable medium of  claim 5 , wherein the interface is to receive listener feedback regarding a plurality of high and low frequency gain comparisons to converge to a desired scaled weighting function. 
     
     
       11. The computer readable medium of  claim 5 , wherein the audio device comprises at least one of a hearing aid and an audio equalizer. 
     
     
       12. The computer readable medium of  claim 5 , further comprising one or more bandpass filters to modify the sound for output. 
     
     
       13. The computer readable medium of  claim 5 , further comprising an automatic synonym map relating language-based descriptors representing listener feedback. 
     
     
       14. A tangible computer readable medium comprising computer program code which, when executed by a processor, implements an audio configuration system for individual learning and customization of a listener's subjective audio preference, said system comprising:
 an output to provide a sound for listener review;   an interface to accept listener feedback regarding the sound, the feedback including a descriptor to associate with the sound;   an automatic synonym map relating language-based descriptors representing listener feedback; and   a processor to program an audio device based on listener feedback, wherein the processor is to analyze the sound and associated listener feedback to generate a weighting function, the weighting function further scalable based on listener feedback, wherein the processor is to apply the weighting function as an audio device setting for the audio device to apply to an audio signal.   
     
     
       15. The computer readable medium of  claim 14 , wherein the interface further comprises a slider controller to receive listener feedback regarding the weighting function. 
     
     
       16. The computer readable medium of  claim 14 , wherein the processor is to modify the sound using at least one equalization curve. 
     
     
       17. The computer readable medium of  claim 14 , wherein the processor is to map a language-based descriptor associated with the sound to a frequency gain curve. 
     
     
       18. The computer readable medium of  claim 17 , wherein the processor is to map the frequency gain curve to the weighting function. 
     
     
       19. The computer readable medium of  claim 14 , wherein the interface is to receive listener feedback regarding a plurality of high and low frequency gain comparisons to converge to a desired scaled weighting function. 
     
     
       20. The computer readable medium of  claim 14 , wherein the audio device comprises at least one of a hearing aid and an audio equalizer. 
     
     
       21. The computer readable medium of  claim 14 , further comprising one or more bandpass filters to modify the sound for output. 
     
     
       22. The computer readable medium of  claim 14 , wherein the processor is to predict listener ratings based on a similarity between a probe curve and the weighting function. 
     
     
       23. A system comprising:
 a hearing device to generate audio processed according to a frequency gain curve; and   a computing device that is separate from the hearing device and that comprises one or more processing devices to execute instructions to generate a user interface for setting the frequency gain curve,   wherein the user interface comprises a control element to enable the user to set the frequency gain curve, and   wherein the one or more processing devices are configured to generate the frequency gain curve from a weighted combination of a set of component weighting functions, the component weighting functions comprising at least one of: (a) a component weighting function corresponding to a modification of a spectral tilt of the frequency gain curve, and (b) a component weighting function corresponding to a modification of a middle frequency balance of the frequency gain curve, the component weighting functions   (i) generated based on a set of user-specific weighting functions, wherein each user-specific weighting function corresponds to an individual's audio preferences, and   (ii) generated to account for a threshold level of variance among the set of user-specific weighting functions.   
     
     
       24. The system of claim 23 wherein the set of user-specific weighting functions is representative of a population of user-specific weighting functions across various individuals and audio descriptors. 
     
     
       25. The system of claim 23, wherein the weighted combination of the set of component weighting functions is determined in accordance with input received through the user interface. 
     
     
       26. The system of claim 23, wherein the set of component weighting functions comprise a set of principal components derived using principal component analysis. 
     
     
       27. The system of claim 23, wherein the hearing device comprises a medical hearing aid device. 
     
     
       28. The system of claim 23, wherein the computing device comprises a handheld computing device. 
     
     
       29. The system of claim 23, wherein the computing device comprises a mobile computing device. 
     
     
       30. The system of claim 23, wherein the computing device is wirelessly connected to the hearing device. 
     
     
       31. The system of claim 23, wherein the control element of the user interface comprises a two-dimensional graphical control. 
     
     
       32. The system of claim 31, wherein the two-dimensional graphical control comprises a graphical slider control. 
     
     
       33. The system of claim 31, wherein the two-dimensional graphical control comprises a movable graphic presented in a box. 
     
     
       34. The system of claim 24, wherein the set of user-specific weighting functions is representative of audio preferences of the entire human population. 
     
     
       35. The system of claim 24, wherein the set of user-specific weighting functions is representative of audio preferences of a population of people with hearing loss. 
     
     
       36. The system of claim 23, wherein the hearing device is configured to process the audio according to the frequency gain curve. 
     
     
       37. The system of claim 23, wherein each user-specific weighting function is derived from user ratings of audio processed by a set of probe equalization curves. 
     
     
       38. One or more non-transitory computer-readable storage media storing instructions that are executable by one or more processing devices to perform operations comprising:
 presenting a user interface on a display of a computing device,
 wherein the user interface comprises a control element configured to receive user input for setting a frequency gain curve used in processing audio for reproduction by a hearing device in communication with the computing device, and 
 wherein the frequency gain curve is derived from a weighted combination of a set of component weighting functions, the component weighting functions comprising at least one of: (a) a component weighting function corresponding to a modification of a spectral tilt of the frequency gain curve, and (b) a component weighting function corresponding to a modification of a middle frequency balance of the frequency gain curve, the component weighting functions
 (i) generated based on a set of user-specific weighting functions, wherein each user-specific weighting function corresponds to an individual's audio preferences, and 
 (ii) generated to account for a threshold level of variance among the set of user-specific weighting functions; and 
 
 causing the frequency gain curve set via the control element to be used to process the audio for reproduction by the hearing device. 
   
     
     
       39. The one or more non-transitory computer-readable storage media of claim 38, wherein the set of user-specific weighting functions is representative of a population of user-specific weighting functions across various individuals and audio descriptors. 
     
     
       40. The one or more non-transitory computer-readable storage media of claim 38, wherein the weighted combination of the set of component weighting functions is determined in accordance with input received through the user interface. 
     
     
       41. The one or more non-transitory computer-readable storage media of claim 38, wherein the set of component weighting functions comprise a set of principal components derived using principal component analysis. 
     
     
       42. The one or more non-transitory computer-readable storage media of claim 38, wherein the operations further comprise processing the audio using the frequency gain curve set via the control element. 
     
     
       43. The one or more non-transitory computer-readable storage media of claim 38, wherein the hearing device is configured to process the audio using the frequency gain curve set via the control element. 
     
     
       44. The one or more non-transitory computer-readable storage media of claim 38, wherein causing the frequency gain curve to be used to process the audio comprises: causing the frequency gain curve set via the control element to be wirelessly transmitted to the hearing device. 
     
     
       45. The one or more non-transitory computer-readable storage media of claim 38, wherein the control element comprises a two-dimensional graphical control. 
     
     
       46. The one or more non-transitory computer-readable storage media of claim 45, wherein the two-dimensional graphical control comprises a graphical slider control. 
     
     
       47. A method for operating an audio device, the method comprising:
 receiving user input comprising a first user-adjustable control and a second user-adjustable control; and   applying a frequency gain curve to an audio signal for reproduction by the audio device, the frequency gain curve being derived from a weighted combination of a set of component weighting functions, the component weighting functions comprising at least one of: (a) a component weighting function corresponding to a modification of a spectral tilt of the frequency gain curve, and (b) a component weighting function corresponding to a modification of a middle frequency balance of the frequency gain curve, the component weighting functions
 (i) generated based on a set of user-specific weighting functions, wherein each user-specific weighting function corresponds to an individual's audio preferences, and 
 (ii) generated to account for a threshold level of variance among the set of user-specific weighting functions, 
 wherein the first user-adjustable control corresponds to a weight of a first component weighting function from the set of component weighting functions, and 
 wherein the second user-adjustable control corresponds to a weight of a second component weighting function from the set of component weighting functions. 
   
     
     
       48. The method of claim 47, wherein the first component weighting function corresponds to a modification of a spectral tilt of the frequency gain curve and the second component weighting function corresponds to a modification of a middle frequency balance of the frequency gain curve.