Method for directional signal processing for a hearing aid

Abstract

A method for directional signal processing for a hearing aid. First and second input transducers generate first and second input signals from an ambient acoustic signal. A forward signal and a backward signal are generated from the first and second input signals and a first directional parameter is determined as a linear factor of a linear combination of the forward and backward signals. The first directional signal has a maximum attenuation in a first direction. A correction parameter is ascertained such that a second directional signal has a defined relative attenuation in the first direction. The second directional signal is generated from the forward signal and the backward signal with the first directional parameter and the correction parameter or with the first directional signal and the omnidirectional signal based on the correction parameter. An output signal of the hearing aid is generated based on the second directional signal.

Claims

1. A method of directional signal processing for a hearing aid, the method comprising: generating a first input signal by a first input transducer of the hearing aid from an ambient acoustic signal; generating a second input signal by a second input transducer of the hearing aid from the ambient acoustic signal; generating a forward signal and a backward signal from the first input signal and the second input signal; determining a first directional parameter as a linear factor of a linear combination of the forward signal and the backward signal for forming a first directional signal from the linear combination having a maximum attenuation in a first direction; ascertaining a correction parameter such that a second directional signal, being a linear combination formed from the first directional signal and an omnidirectional signal with the correction parameter, has a defined relative attenuation in the first direction; generating the second directional signal from the forward signal and the backward signal on a basis of the first directional parameter and the correction parameter, the second directional signal being generated by a linear combination of the forward signal and the backward signal, with a second directional parameter as a linear factor; ascertaining the second directional parameter by a specified functional relationship from the first directional parameter and the correction parameter such that the second directional signal has the defined relative attenuation in the first direction; and generating an output signal of the hearing aid based on the second directional signal.

2. The method according to claim 1, wherein the second directional parameter emerges from the first directional parameter by way of a scaling by the correction parameter and by way of a specified offset.

3. The method according to claim 1, wherein the first directional parameter is generated by adaptive directional microphony with regard to the linear combination of the forward signal and the backward signal.

4. The method according to claim 3, wherein the step of generating the first direction parameter comprises minimizing a signal energy.

5. The method according to claim 4, which comprises ascertaining the correction parameter based on at least one variable characterizing the acoustic signal selected from the group consisting of: a noise floor level, a signal-to-noise ratio, a stationarity parameter, and a directional information item.

6. The method according to claim 5, which comprises forming the correction parameter by a monotonic function of the noise floor level which characterizes the acoustic signal, wherein the monotonic function, above an upper threshold, maps the noise floor level to a first end point of the value range of the correction parameter, at which the second directional signal transitions into the first directional signal.

7. The method according to claim 6, which comprises correcting the monotonic function of the noise floor level which characterizes the acoustic signal based on the signal-to-noise ratio and/or based on a stationarity parameter in conjunction with a directional information item.

8. The method according to claim 1, which comprises: within a defined neighborhood of a second end point of a value range of the correction parameter, effecting a superposition of a third directional signal on the second directional signal, the third directional signal being configured to simulate a natural directional effect of a human ear; and transitioning the superposition into the third directional signal when the correction parameter adopts the second end point of the value range of the correction parameter.

9. The method according to claim 1, which comprises: generating the forward signal on a basis of a time delayed superposition, implemented by way of a first delay parameter, of the first input signal with the second input signal; and/or generating the backward signal on a basis of a time delayed superposition, implemented by way of a second delay parameter, of the second input signal with the first input signal.

10. The method according to claim 9, which comprises: generating the forward signal as a forwardly directed cardioid directional signal; and generating the backward signal as a backwardly directed cardioid directional signal.

11. A hearing system, comprising a hearing aid having a first input transducer for generating a first input signal from an ambient acoustic signal and a second input transducer for generating a second input signal from the ambient acoustic signal; and a control unit configured to carry out the method according to claim 1.

12. A method of directional signal processing for a hearing aid, the method comprising: generating a first input signal by a first input transducer of the hearing aid from an ambient acoustic signal; generating a second input signal by a second input transducer of the hearing aid from the ambient acoustic signal; generating a forward signal and a backward signal from the first input signal and the second input signal; determining a first directional parameter as a linear factor of a linear combination of the forward signal and the backward signal for forming a first directional signal from the linear combination having a maximum attenuation in a first direction; ascertaining a correction parameter such that a second directional signal, being a linear combination formed from the first directional signal and an omnidirectional signal with the correction parameter, has a defined relative attenuation in the first direction; generating the second directional signal from the forward signal and the backward signal from the first directional signal and the omnidirectional signal on a basis of the correction parameter, the second directional signal being generated by a convex superposition of the first directional signal and the omnidirectional signal, with the correction parameter as a convexity parameter; and generating an output signal of the hearing aid based on the second directional signal.

13. The method according to claim 12, wherein the first directional parameter is generated by adaptive directional microphony with regard to the linear combination of the forward signal and the backward signal.

14. The method according to claim 13, wherein the step of generating the first direction parameter comprises minimizing a signal energy.

15. The method according to claim 14, which comprises ascertaining the correction parameter based on at least one variable characterizing the acoustic signal selected from the group consisting of: a noise floor level, a signal-to-noise ratio, a stationarity parameter, and a directional information item.

16. The method according to claim 15, which comprises forming the correction parameter by a monotonic function of the noise floor level which characterizes the acoustic signal, wherein the monotonic function, above an upper threshold, maps the noise floor level to a first end point of the value range of the correction parameter, at which the second directional signal transitions into the first directional signal.

17. The method according to claim 16, which comprises correcting the monotonic function of the noise floor level which characterizes the acoustic signal based on the signal-to-noise ratio and/or based on a stationarity parameter in conjunction with a directional information item.

18. A method of directional signal processing for a hearing aid, the method comprising: generating a first input signal by a first input transducer of the hearing aid from an ambient acoustic signal; generating a second input signal by a second input transducer of the hearing aid from the ambient acoustic signal; generating a forward signal and a backward signal from the first input signal and the second input signal; determining a first directional parameter as a linear factor of a linear combination of the forward signal and the backward signal for forming a first directional signal from the linear combination having a maximum attenuation in a first direction; ascertaining a correction parameter such that a second directional signal, being a linear combination formed from the first directional signal and an omnidirectional signal with the correction parameter, has a defined relative attenuation in the first direction; generating a second direction by swiveling the first direction about an angle tabulated on a basis of the correction parameter; generating the second directional signal from the forward signal and the backward signal on a basis of the first directional parameter and the correction parameter, the second directional signal being generated by a linear combination of the forward signal and the backward signal with a second directional parameter as a linear factor; and ascertaining the second directional parameter to form the second directional signal with a maximum attenuation in the second direction; and generating an output signal of the hearing aid based on the second directional signal.

19. The method according to claim 18, wherein the first directional parameter is generated by adaptive directional microphony with regard to the linear combination of the forward signal and the backward signal.

20. The method according to claim 19, wherein the step of generating the first direction parameter comprises minimizing a signal energy.

21. The method according to claim 20, which comprises ascertaining the correction parameter based on at least one variable characterizing the acoustic signal selected from the group consisting of: a noise floor level, a signal-to-noise ratio, a stationarity parameter, and a directional information item.

22. The method according to claim 21, which comprises forming the correction parameter by a monotonic function of the noise floor level which characterizes the acoustic signal, wherein the monotonic function, above an upper threshold, maps the noise floor level to a first end point of the value range of the correction parameter, at which the second directional signal transitions into the first directional signal.

23. The method according to claim 22, which comprises correcting the monotonic function of the noise floor level which characterizes the acoustic signal based on the signal-to-noise ratio and/or based on a stationarity parameter in conjunction with a directional information item.

Description

BRIEF DESCRIPTION OF THE SEVERAL VIEWS OF THE DRAWING

(1) FIG. 1 shows a block diagram of a hearing aid according to the prior art, in which a directional signal with a maximum attenuation in a first direction is generated by means of adaptive directional microphony;

(2) FIG. 2 shows a block diagram of a development according to the invention of the hearing aid of FIG. 1, wherein the attenuation is reduced in defined fashion in the first direction;

(3) FIG. 3 shows a functional diagram of a correction parameter for reducing the attenuation as per FIG. 2 on the basis of a noise floor level;

(4) FIG. 4 shows a block diagram of an alternative configuration of the hearing aid according to FIG. 2; and

(5) FIG. 5 shows a diagram of the direction of maximum attenuation for a first directional signal and a directional signal developed as per FIG. 2 or FIG. 4, as a function of the directional parameter.

(6) Mutually corresponding parts and variables are respectively provided with identical reference signs and numerals throughout the figures.

DETAILED DESCRIPTION OF THE INVENTION

(7) Referring now to the figures of the drawing in detail and first, in particular, to FIG. 1 thereof, there is shown a schematic block diagram of a method for directional signal processing in a hearing aid 1 according to the prior art. The hearing aid 1 has a first input transducer 2 and a second input transducer 4, which generate a first input signal E1 and a second input signal E2, respectively, from an acoustic signal 6 that is injected from the surroundings, i.e., an ambient acoustic signal 6. Each of the input transducers 2, 4 may be a microphone, for example. Here, in respect of a frontal direction 7 of the hearing aid 1 (which is defined by the intended wear during operation), the first input transducer 2 is disposed further forward than the second input transducer 4.

(8) The second input signal E2 is now delayed by a first delay parameter T1 and the second input signal, thus delayed, is subtracted from the first input signal E1 in order to generate a forward signal Z1. In a similar fashion, the first input signal E1 is delayed by a second delay parameter T2 and the second input signal E2 is subtracted from the first input signal, thus delayed, in order to generate a backward (i.e., rearward) signal Z2. Here, apart from possible quantification errors during the digitization, the first delay parameter T1 and the second delay parameter T2 are given by the time-of-flight T, which precisely corresponds to the spatial acoustic path d between the first input transducer 2 and the second input transducer 4. Consequently, the forward signal Z1 is given by a forwardly directed cardioid signal 16 and the backward signal Z2 is given by a rearwardly directed cardioid signal 18 (i.e., an anti-cardioid).

(9) A first directional signal R1 is obtained by way of adaptive directional microphony 20 from the forward signal Z1 and the backward signal Z2 by way of minimizing the signal energy of the signal Z1+a1.Math.Z2 over a first directional parameter a1. Here, the first directional signal R1 has a directional characteristic 22 with a maximum attenuation in a first direction 24. As a consequence of choosing the first directional parameter a1 by means of the adaptive directional microphony 20, the first direction 24 coincides with the direction of a dominant, localized sound source 25 in the rear half space 26. In the example illustrated in FIG. 1, the first direction is twisted through about 120° with respect to the frontal direction 7, which coincides with a frontal direction of the wearer of the hearing aid 1 (not illustrated) when the hearing aid 1 is worn as intended. Here, a maximum attenuation means that the sound coming from the first direction 24 is completely canceled (i.e., “infinitely” attenuated) in the ideal case. In other words, the first directional signal 1 has a so-called “notch” in the first direction 24.

(10) An output signal out, which is converted into an acoustic output signal 34 by an output transducer 32 of the hearing aid 1, is now generated from the signal contributions of the first directional signal R1, and possibly by way of even further non-directional signal processing 29. In the present case, the output transducer 32 may be a loudspeaker or else a bone conduction receiver.

(11) If the dominant sound source 25 in the rear half space 26 (i.e., the rear hemisphere) originates from a speaker, for example, the presently implemented, maximum attenuation of their speech contributions may often not be desirable for the wearer of the hearing aid 1. In this case, it would be advantageous to use an output signal out with a directional characteristic that has no maximum attenuation in the first direction 24.

(12) A corresponding method which can achieve this objective is illustrated with reference to FIG. 2. A block diagram shows a hearing aid 1 which is the same as the hearing aid according to FIG. 1 up to the point of generation of the first directional signal R1. Now, in the example according to FIG. 2, an omnidirectional signal om is formed on the basis of the forward signal Z1 and the backward signal Z2. The omnidirectional signal is superposed on the first directional signal R1 according to a specification yet to be described. This superposition is implemented according to the stipulation of a correction parameter e, which can be ascertained on the basis of the noise floor level NP and the SNR of the acoustic signal 6; however, it can moreover also be ascertained on the basis of a stationarity parameter S1 and a direction information item IR for the acoustic signal 6. Here, the variables can be ascertained either from the input signals E1 and E2 or from the forward and the backward signal Z1, Z2.

(13) A second directional signal R2 emerges from the superposition according to
R2=(1−e).Math.om+e.Math.R1  (cf. equation i).

(14) On the basis of the second directional signal R2, possibly also on the basis of further, non-directional signal processing 29 which may comprise, inter alia, a frequency band-dependent amplification and/or compression, the output signal out is generated in a manner analogous to the procedure illustrated in FIG. 1, the output signal being converted by the output transducer 32 into the acoustic output signal 34. Now, the directional characteristic 38 of the second directional signal R2 has its maximum attenuation along a second direction 40, whereas there is a relative attenuation 42 in the first direction 24.

(15) FIG. 3 illustrates a function f which maps the noise floor level NP on the correction parameter e of the method illustrated on the basis of FIG. 2 (solid line). Above an upper threshold Th.sub.Hi, which is chosen as Th.sub.Hi=80 dB in the example as per FIG. 3, any floor noise level is mapped to e=1. This means the following: In the method illustrated in FIG. 2, the first directional signal R1 is completely converted into the second directional signal R2 for a noise floor level NP of 80 dB and more. Below a lower threshold Th.sub.Lo, which is chosen as Th.sub.Lo=40 dB in the example as per FIG. 3, any floor noise level is mapped to e=0. This means the following: In the method illustrated in FIG. 2, the omnidirectional signal om is completely converted into the second directional signal R2 for a noise floor level NP of 40 dB and less. In the range Th.sub.Lo<NP<Th.sub.Hi, the function f has a linear gradient, which can be described by
e=f(NP)=(NP−Th.sub.Lo)/(Th.sub.Hi−Th.sub.Lo).

(16) A different characteristic to the linear relation illustrated here is likewise conceivable, as long as the monotonic gradient for f(NP) is maintained between Th.sub.Lo and Th.sub.Hi.

(17) If the SNR now lies above a specified threshold Th.sub.SNR, i.e., SNR≥Th.sub.SNR, the characteristic provided by the function f(NP) is capped, a new function f′ (dashed line) emerging therefrom. In this case, this means the following: If the SNR is above Th.sub.SNR, the behavior is identical to the original function f for comparatively low values of the noise floor level NP. However, above approximately NP=65 dB, e is always mapped to the value e=0.675. This takes account of the fact that, in the case of a high SNR, the directional noise suppression need not be completely implemented even in the case of a high noise floor level NP, and a greater component of the omnidirectional signal om can remain mixed in for reasons of the improved spatial hearing perception.

(18) Should it moreover be determined that the acoustic signal 6 is firstly sufficiently non-stationary—e.g., on account of dropping below an upper limit Th.sub.S by the stationarity parameter S1—and moreover has a significant component originating from the rear half space (which is identified on the basis of the direction information item IR, which, for example, specifies the half space of the first direction 24 emerging from the adaptive directional microphony 20), the gradient of the function f is reduced in a range above 55 dB for the noise floor level NP (dotted line), as a result of which e=1 is only reached for a noise floor level NP above the threshold Th.sub.Hi (under the assumption SNR<Th.sub.SNR because otherwise the function f′ is immediately applied).

(19) A procedure analogous to the method explained on the basis of FIG. 2 is illustrated in FIG. 4. In a block diagram, the latter shows a hearing aid 1, which is modeled on the hearing aid 1 illustrated in FIG. 2. However, in this case, the second directional signal R2 is not formed as a superposition of the first directional signal R1 with the omnidirectional signal om according to the correction parameter e as a convexity parameter. Rather, the first directional parameter a1, which emerges from the generation of the first directional signal R1 by the adaptive directional microphony 20, is mapped as per the specification
a2=e+e.Math.a1−1  (cf. equation vi)
on a second directional parameter a2, which is formed by scaling of the first directional parameter a1 by the factor e (the convexity parameter as per FIG. 2) and by shifting by the offset e−1. The second directional signal R2 is formed, in a manner analogous to the first directional signal R1, from the forward signal Z1 and the backward signal Z2 as
R2=Z1+a2.Math.Z2  (cf. equations v and vi).

(20) The directional characteristic 38 is accordingly equal to the directional characteristic of the second directional signal R2 according to FIG. 2 since, under the same conditions, the procedure illustrated in FIG. 4 is analogous to the procedure illustrated in FIG. 2, apart from an expansion for e≤0.1, which is described below. The maximum attenuation is now implemented in a second direction 40, while a defined relative attenuation 42 is present in the first direction 24.

(21) In the case that a value in the vicinity of zero emerges from the calculation of the correction parameter e as per FIG. 3, i.e., e smaller than a specified threshold e.sub.Lo=M with, e.g., M=0.1, the output signal out is generated by virtue of a third directional signal R3 being mixed to the second directional signal R2, for example according to the following formula:
out=(e/M).Math.R2+[(M−e)/M].Math.R3  (cf. equation xi).

(22) Here, the third directional signal R3 is generated with a fixed directional characteristic from the forward signal Z1 and the backward signal Z2. Alternative transitions between R2 and R3, which do not have the aforementioned linear relationship in e, are likewise conceivable.

(23) FIG. 5 schematically shows, in a diagram, the relationship between the first directional parameter a1, which characterizes the first directional signal R1, and the second directional parameter a2 of the second directional signal R2 according to FIG. 4. Here, the functional relationship is a2=0.7.Math.a1−0.3. In the example illustrated in FIG. 5, the lower symbols are formed by the respective first direction 24 with respect to the parameter value of the first directional parameter a1, while the upper symbols are given by the second direction with respect to the given parameter value for a1, i.e., by the angle to which, in the second directional signal R2, the second direction 40, i.e., the direction of maximum attenuation after applying the mapping of the first directional parameter R1 on the second directional parameter a2, adjusts. In respect of a given value of a1, it is possible to determine that the angle increases, wherein, as a consequence of the axial symmetry of the directional characteristics with respect to the frontal direction, there is clipping in the angle direction of 180°, which is counter to the frontal direction. As a result of the shown swiveling of the direction of maximum attenuation during the transition from the first to the second directional signal, a relative attenuation, defined in relation to the maximum sensitivity and controlled by the correction parameter e, emerges in the first direction, which still had the maximum attenuation in the first directional signal.

(24) Even though the invention was illustrated more closely and described in detail by way of the preferred exemplary embodiment, the invention is not restricted by the disclosed examples and other variations can be derived therefrom by a person skilled in the art without departing from the scope of protection of the invention.

(25) The following is a summary list of reference numerals and the corresponding structure used in the above description of the invention: 1 Hearing aid 2 First input transducer 4 Second input transducer 6 Ambient acoustic signal, acoustic signal of the surroundings 7 Frontal direction 16 Forwardly directed cardioid (signal) 18 Backwardly directed cardioid (signal) 20 Adaptive directional microphony 22 Directional characteristic 24 First direction 25 Dominant sound source 26 Rear half space 29 Non-directional signal processing 32 Output transducer 34 Acoustic output signal 38 Directional characteristic 40 Second direction 42 Relative attenuation a1 First directional parameter a2 Second directional parameter e Correction parameter E1 First input signal E2 Second input signal IR Directional information item om Omnidirectional signal out Output signal NP Noise floor level R1 First directional signal R2 Second directional signal R3 Third directional signal S1 Stationarity parameter SNR Signal-to-noise ratio Th.sub.Lo Lower threshold (for the noise floor level NP) Th.sub.Hi Upper threshold (for the noise floor level NP) Th.sub.S Upper threshold (for the SNR) Z1 Forward signal Z2 Backward signal