Method and apparatus for improved estimation of non-stationary noise for speech enhancement

• US 20070055508A1
• Filed: 08/23/2006
• Published: 03/08/2007
• Est. Priority Date: 09/03/2005
• Status: Active Grant

• Current Assignee: GN Resound A/S (GN Store Nord A/S)
• Original Assignee: GN Resound A/S (GN Store Nord A/S)
• Inventors: Zhao, David, Ypma, Alexander, Devries, Bert, Kleijn, Willem
• Granted Patent: US 7,590,530 B2
• Time in Patent Office: Days
• US Class Current: 704/226
• CPC Class Codes: G10L 21/0216 characterised by the method used for estimating noise
• CPC Class Codes: H04R 25/55 using an external connection, either wireless or wired

Abstract

A central aspect of the invention relates to a method of enhancing speech, the method comprising the steps of, receiving noisy speech comprising a clean speech component and a non-stationary noise component, providing a speech model, providing a noise model having at least one shape and a gain, dynamically modifying the noise model based on the speech model and the received noisy speech, enhancing the noisy speech at least based on the modified noise model. Hereby is achieved a method of speech enhancement that is able to suppress highly non-stationary noise. Another aspect of the invention relates to a speech enhancement system that may be adapted to be used in a hearing system, such as a hearing aid or a headset.

Claims

1. A method of enhancing speech, comprising:

receiving noisy speech comprising a clean speech component and a non-stationary noise component;

providing a speech model;

providing a noise model having at least one shape and a gain;

dynamically modifying the noise model based on the speech model and the received noisy speech;

enhancing the noisy speech at least based on the modified noise model.

2. The method of claim 1, wherein the at least one shape and gain of the noise model are respectively modified separately.

3. The method of claim 1, wherein the gain of the noise model dynamically modified at a higher rate than the at least one shape of the noise model.

4. The method of claim 1, wherein the noisy speech enhancement is further based on the speech model.

5. The method of claim 1, further comprising estimating the noise component based on the modified noise model, wherein the noisy speech is enhanced based on the estimated noise component.

6. The method of claim 1, further comprising estimating the speech component based on the speech model, wherein the noisy speech is enhanced based on the estimated speech component.

7. The method of claim 5, further comprising estimating the speech component based on the speech model, wherein the noisy speech is enhanced further based on the estimated speech component.

8. The method of claim 1, further comprising the steps of dynamically modifying the speech model based on the noise model and the received noisy speech and enhancing the noisy speech based on the modified speech model.

9. The method of claim 1, wherein the noise model is a hidden Markov model (HMM).

10. The method of claim 9, wherein the HMM is a Gaussian mixture model.

11. The method of claim 1, wherein the noise model is derived from at least one code book.

12. The method of claim 1, wherein providing the noise model comprises selecting one of a plurality of noise models based on the non-stationary noise component.

13. The method of claim 1, wherein the dynamic modification of the noise model, the noise component estimation, and the noisy speech enhancement are repeatedly performed.

15. The speech enhancement system of claim 13, further being adapted to be used in a hearing system.

14. A speech enhancement system comprising:

a speech model;

a noise model having at least one shape and a gain;

a microphone for the provision of an input signal based on the reception of noisy speech, which noisy speech comprises a clean speech component and a non-stationary noise component;

a signal processor adapted to modify the noise model based on the speech model and the input signal, and enhancing the noisy speech on the basis of the modified noise model in order to provide a speech enhanced output signal, wherein the signal processor is further adapted to perform the modification of the noise model dynamically.

16. A method of enhancing speech, comprising:

receiving noisy speech comprising a clean speech component and a noise component;

providing a cost function equal to a function of a difference between a candidate for an estimated enhanced speech component and a function of the clean speech component and the noise component;

enhancing the noisy speech based on estimated speech and noise components; and

minimizing the Bayes risk for the cost function to obtain the enhanced speech component.

17. The method of claim 16, where the enhancement is based on a speech model and a noise model.

18. The method of claim 16, where the cost function is the squared error function for a candidate for the estimated speech compared to clean speech plus a function of the residual noise.

19. The method of claim 16, where the function of the residual noise component is multiplying the residual noise component by an epsilon parameter chosen in dependence of the received noisy signal.

20. The method of claim 19, where the epsilon parameter is chosen in dependence of a human perception of the noisy signal.

21. A speech enhancement system comprising:

a microphone for the provision of an input signal based on the reception of noisy speech, which noisy speech comprises a clean speech component and a non-stationary noise component;

a signal processor adapted to process the input signal in order to provide a speech enhanced output signal based on estimated speech and noise components, by minimizing the Bayes risk for a cost function in order to obtain the enhanced speech component, wherein the cost function is equal to a function of a difference between a candidate for an estimated enhanced speech component and a function of the clean speech component and the noise component.

22. The speech enhancement system of claim 21, further being adapted to be used in a hearing system.

23. A method of enhancing speech further comprising the steps of maintaining a list of noise models, said steps comprising:

receiving noisy speech;

dynamically modifying one of the noise models based on the received noisy speech;

comparing the modified noise model to the list of noise models; and

adding the modified noise model to the noise model list based on the comparison.

24. The method of claim 23, wherein the modified noise model is added to the noise model list if a difference between the modified noise model and at least one of the noise models in the list is greater than a threshold.

25. The method of claim 23, wherein the noise models are hidden Markov models (HMM).

26. The method of claim 25, wherein the HMMs are Gaussian mixture models.

27. The method of claim 23, wherein the noise models are initially derived from at least one code book

28. The method of claim 23, further being adapted to be used in a hearing system.