System and method for dynamic residual noise shaping

1. A dynamic residual noise shaping method, comprising:

detecting an amount and a type of hiss noise in an audio signal;

calculating noise suppression gains by applying a noise reduction filter to the audio signal; and

modifying the calculated noise suppression gains by applying a dynamic attenuation floor responsive to the detected amount and type of hiss noise.

2. The method of claim 1 , where the act of modifying the calculated noise suppression gains responsive to the detected amount and type of hiss noise further comprises applying the dynamic attenuation floor above a hiss cutoff frequency.

3. The method of claim 1 , where the act of modifying the calculated noise suppression gains responsive to the detected amount and type of hiss noise further comprises applying a constant attenuation floor below a hiss cutoff frequency.

4. The method of claim 2 , where the hiss cutoff frequency is a fixed frequency.

5. The method of claim 2 , where the hiss cutoff frequency is a function of a bandwidth of the audio signal.

6. The method of claim 2 , where the hiss cutoff frequency is adaptive responsive to a spectral shape of the hiss noise.

7. The method of claim 1 , where detecting an amount and a type of hiss noise in the audio signal further comprises analyzing any of the audio signal, a spectral magnitude of the audio signal and a background noise estimate.

8. The method of claim 1 , where modifying the noise suppression gains responsive to the detected amount and type of hiss noise comprises modifying the noise suppression gains to substantially correlate to a target noise shape for each of a plurality of frequency bins of the audio signal.

9. The method of claim 8 , where the target noise shape comprises one of a white, a pink or a brown noise.

10. The method of claim 8 , where the target noise shape comprises an increasing gain with an increasing frequency.

11. The method of claim 1 , where the dynamic attenuation floor is applied for each of a plurality of frequency bins of the audio signal when a difference between a noise estimate and a target noise exceeds a hiss threshold for each respective frequency bin of the audio signal.

12. The method of claim 11 , further comprising:

applying the modified noise suppression gains to the audio signal to reduce the amount of hiss noise in the audio signal.

13. A system for dynamic residual noise shaping, the system comprising:

a processor;

a memory coupled to the processor containing instructions, executable by the processor, programmed to:

detect an amount and a type of hiss noise in an audio signal;

calculate noise suppression gains by applying a noise reduction filter to the audio signal; and

modify the calculated noise suppression gains by applying a dynamic attenuation floor responsive to the detected amount and type of hiss noise.

14. The system of claim 13 , where the program that modifies the calculated noise suppression gains responsive to the detected amount and type of hiss noise further applies the dynamic attenuation floor above a hiss cutoff frequency.

15. The system of claim 13 , where the program that modifies the calculated noise suppression gains responsive to the detected amount and type of hiss noise further applies a constant attenuation floor below a hiss cutoff frequency.

16. The system of claim 14 , where the hiss cutoff frequency comprises a fixed frequency.

17. The system of claim 14 , where the hiss cutoff frequency comprises a function of a bandwidth of the audio signal.

18. The system of claim 14 , where the hiss cutoff frequency comprises an adaptive response to a spectral shape of the hiss noise.

19. The system of claim 13 , where the program that detects the amount and the type of hiss noise in the audio signal further analyzes any of the audio signal, a spectral magnitude of the audio signal and a background noise estimate.