Method for the directional processing of signals from a microphone arrangement

Description

[0001] The invention relates to a method for the directional signal processing of signals from a microphone arrangement which comprises at least a first microphone for generating a first input signal from ambient sound and a second microphone for generating a second input signal from the ambient sound, wherein the first input signal and the second input signal are each transformed into the frequency domain, and a first frequency-domain directional signal is formed on the basis of the first input signal thus transformed and the second input signal thus transformed in the frequency domain, wherein frequency-dependent first amplification factors are generated in the signal derived in the frequency domain, wherein an output signal is generated on the basis of the first amplification factors and on the basis of the frequency-domain directional signal.

[0002] Frequency-resolved signal processing for a microphone array with one or more microphones is typically performed in the frequency domain by splitting one or more microphone signals from the array into individual frequency bands. Signal components from each frequency band are processed separately, amplified and / or compressed differently, and possibly combined into directional signals. The individual signal components in the frequency bands are then "synthesized" into a single output signal in the time domain.

[0003] The decomposition into individual frequency bands (in combination with the final synthesis) leads to a latency of the signal processing, which depends on the desired frequency resolution in the frequency domain and is usually about 5-8 ms.

[0004] InIn real-time applications such as hearing instruments, especially in hearing aids, such latency often leads to an overlap of the latency-delayed output signal reproduced by the hearing instrument with the direct ambient sound, which can cause comb filter effects. These are usually more pronounced the more similar the direct ambient sound is to the output signal processed and reproduced by the hearing instrument.

[0005] Comb filter effects are often perceived as unpleasant, which is why they should be avoided. One way to achieve this is to significantly reduce latency during signal processing or, if possible, eliminate it entirely. However, this significantly limits the possibilities for frequency-resolved signal processing, as this often requires a secondary signal path ("analysis path") to determine how the signal processing should be applied to the signal components of the main signal path. In particular, direction-dependent signal processing of the individual input signals generated by multiple microphones in the microphone array is difficult to implement in this scenario.

[0006] WO 2021 / 110 924 A1 shows a hearing aid in which two microphone signals from two different microphones are processed to form a preliminary directional signal, a frequency band-wise target gain for the preliminary directional signal is determined, and based on the target gain, a filter with minimum phase is generated, which is applied to the preliminary directional signal to generate an output signal.

[0007] The invention is therefore based on the object of specifying a method for the directional signal processing of signals from a microphone arrangement, which method should have the lowest possible latency of a generated output signal compared to the input signals of the microphone arrangement to be processed.

[0008] The stated object is achieved according to the invention by a method for the directional signal processing of signals from a microphone arrangement which comprises at least a first microphone for generating a first input signal from an ambient sound and a second microphone for generating a second input signal from the ambient sound, wherein a reference signal is formed on the basis of the first input signal, wherein the reference signal is transformed into the frequency domain, and thereby a frequency domain reference signal is generated, wherein the first input signal and the second input signal are each transformed into the frequency domain, and a first frequency domain directional signal is formed on the basis of the first input signal thus transformed and the second input signal thus transformed in the frequency domain.

[0009] This provides that frequency-dependent first amplification factors are generated based on a frequency-resolved comparison of the frequency-domain reference signal with the first frequency-domain directional signal or with a signal derived from the first frequency-domain directional signal in the frequency domain, that a time filter is generated in the time domain based on the first amplification factors, that the reference signal is filtered by the time filter, and that an output signal is generated based on the reference signal filtered by the time filter. Advantageous and inventive embodiments are the subject of the dependent claims and the following description.

[0010] A microphone, as used herein and in the following, is intended to include any electroacoustic input transducer configured to generate a corresponding input signal from the ambient sound, with sound pressure fluctuations of the ambient sound being translated by the input transducer into corresponding voltage or current fluctuations. Accordingly, a microphone arrangement is to be understood as any spatial arrangement of at least two such input transducers, with the spatial spacing of the input transducers from one another enabling direction-dependent signal processing.

[0011] The formation of the reference signal based on the first input signal includes, in particular, the two cases where, on the one hand, the reference signal is formed solely based on the signal components of the first input signal, but other signal components, in particular those of the second input signal, are not included in the reference signal, or, on the other hand, the reference signal can be formed based on the signal components of both input signals, in particular by means of a time-delayed superposition of the two input signals. The reference signal is a signal in the time domain.

[0012] The aforementioned transformations of a signal into the frequency domain—in this case, the reference signal and the first and second input signals—can be performed, in particular, using a suitably configured filter bank. In the following, the frequency domain also includes, in particular, the discrete-time time-frequency domain, in which the spectral components of the individual transformed signals are updated according to a time variable.

[0013] The first frequency-domain directional signal can be generated on the basis of the transformed first input signal and the transformed second input signal, in particular in such a way that the resulting directional characteristic for the first frequency-domain directional signal varies across the individual frequency bands (and that, for example, a linear combination of the two transformed input signals or a linear combination of intermediate signals derived from both input signals, such as cardioid / anticardioid signals, has different linear factors across the frequency bands).

[0014] The frequency-domain reference signal is compared in a frequency-resolved manner, i.e., preferably frequency-band-wise, with the first frequency-domain directional signal or a signal derived therefrom in the frequency domain, so that the frequency-dependent first gain factors are generated as the quantitative result of this comparison. This comparison can be performed, for example, by spectral division (the first gain factors can then be determined directly, frequency-band-wise, as the respective quotients formed by dividing the frequency-domain reference signal by the first frequency-domain directional signal). Other comparisons, for example, in the form of a relative deviation and / or in the form of a nonlinear, preferably monotonic function of the frequency-band-wise deviation of the two aforementioned signals from each other, are also conceivable.For a comparison of the frequency-domain reference signal with a signal derived in the frequency domain from the first frequency-domain directional signal, a second frequency-domain directional signal is preferably used, the signal components of which can be derived directly from the first frequency-domain directional signal and preferably without adding further signal components, for example via a frequency-band-wise amplification of the first frequency-domain directional signal.

[0015] The first gain factors, which thus represent a measure of the deviation of the first frequency-domain directional signal from the frequency-domain reference signal in the respective frequency band (possibly implicitly in the case of a derived signal), are then used to generate a time filter in the time domain. The time filter is preferably a filter with a finite impulse response (FIR). In particular, the time filter can be generated by transforming a transfer function from the frequency domain to the time domain, which represents the frequency-band-wise application of the first gain factors in the frequency domain. Particularly preferably, the time filter is generated as a minimum-phase filter, which therefore has the minimum possible latency for a given absolute frequency response.

[0016] Using the resulting time filter, the reference signal is filtered in the time domain, generating an output signal. The output signal can then be transmitted to a receiver, recorded, or reused. If the microphone array is used in a hearing instrument, the output signal is preferably converted into an output sound signal by a loudspeaker (in the broadest sense, an electroacoustic output transducer of the hearing instrument). Before said conversion, the output signal can be subjected to further signal processing, for example, to suppress acoustic feedback that can occur between the loudspeaker and the microphone array.

[0017] The use of a time filter applied to the reference signal in the time domain results in a significant reduction in signal processing latency compared to signal processing in the frequency domain. Due to the mapping of the first gain factors, which are obtained from the first frequency-domain directional signal, to the time filter, the information regarding the direction-dependent sound image in the individual frequency bands, as it results from the generation of the first frequency-domain directional signal, can be implemented in the time filter. However, the differences that the directional signal processing has on the first frequency-domain directional signal compared to the reference signal - transformed into the frequency domain - are taken into account.

[0018] Based on the comparison of the frequency-domain reference signal with the first frequency-domain directional signal, or with a signal preferably derived directly therefrom, it is determined which frequency-band amplification factors are to be applied to the frequency-domain reference signal (which actually only represents the reference signal transformed into the frequency domain) in order to approximate as closely as possible the frequency-band sound properties, particularly with regard to signal volumes and levels, that exist for the first frequency-domain directional signal. Applying the time filter, particularly an FIR filter, in the time domain corresponds to applying frequency- and time-dependent attenuation or amplification factors in each frequency band, followed by signal synthesis.This "encodes" the frequency-band-wise directional dependence encoded in the first frequency-domain directional signal into the first gain factors for the frequency-domain reference signal. By mapping these first gain factors (and thus the described encoded information on the effect of the directivity) to the time filter, this directional information can be made available to the reference signal with very low latency.

[0019] Preferably, the frequency-resolved comparison of the frequency-domain reference signal with the first frequency-domain directional signal or the signal derived from the first frequency-domain directional signal in the frequency domain is carried out using a spectral division, which is used to generate the frequency-dependent first amplification factors. The comparison using a spectral division can be implemented particularly efficiently, and also takes into account the fact that the first amplification factors thus represent a transfer function between the two aforementioned signals in the frequency domain. The frequency-dependent first amplification factors can be generated, in particular, by dividing the magnitudes of the frequency-domain reference signal and the first frequency-domain directional signal or the signal derived from the first frequency-domain directional signal in the frequency domain.

[0020] Advantageously, a second frequency-domain directional signal is generated as a signal derived from the first frequency-domain directional signal in the frequency domain for the frequency-resolved comparison with the frequency-domain reference signal. This second frequency-domain directional signal is generated by applying frequency-dependent second gain factors to the first frequency-domain directional signal. The first frequency-domain directional signal can additionally be subjected to noise suppression, for example, using a Wiener filter, and, if necessary, dynamic compression, so that the frequency-domain reference signal is compared with the resulting second frequency-domain directional signal for the first gain factors.

[0021] The time filter is expediently formed based on a mapping of the frequency-dependent first gain factors into the time domain. The term "mapping" can in particular mean that a transfer function corresponding to the first gain factors or a transfer function corresponding to a product of the first gain factors with further frequency-band-specific gain factors is transformed from the frequency domain to the time domain. In this way, it can be efficiently ensured that a filter with the correct properties is applied to the reference signal in the time domain, i.e., with the properties that are given in the frequency domain by the first gain factors. The time filter is in particular generated as an FIR filter.

[0022] Advantageously, frequency-dependent second gain factors are determined for the first frequency-domain directional signal, wherein the time filter is formed based on a joint mapping of the first gain factors and the second gain factors into the time domain. In particular, the frequency-dependent second gain factors for the first frequency-domain directional signal are determined based on noise suppression and / or dynamic compression and / or a hearing impairment to be corrected of a receiver of the output signal. The receiver of the output signal is, in particular, a user of a hearing instrument that includes the microphone arrangement.

[0023] In other words, this means: Noise suppression and / or dynamic compression, which are to be applied to the first frequency-domain directional signal depending on the frequency band, determines instantaneous second amplification factors for the first frequency-domain directional signal according to the respective specifications for a signal-to-noise ratio ("signal-to-noise ratio", SNR) or the specifications for maximum signal levels in the frequency bands (which can also be individually tailored to a hearing impairment of the user of the hearing instrument that includes the microphone arrangement). However, these are not applied to the first frequency-domain directional signal. Instead, the first amplification factors, which result from the comparison of the first frequency-domain directional signal with the transformed reference signal, are mapped together with the second amplification factors into the time domain to form the time filter. The properties of the noise suppression orThe dynamic compression therefore does not enter the time filter via the comparison of the transformed reference signal with a second frequency domain directional signal, which results from the application of the noise suppression or the dynamic compression to the first frequency domain directional signal, but directly via a mapping of the second gain factors determined for the noise suppression or the dynamic compression into the time domain.

[0024] It proves advantageous if the reference signal is formed only from signal components of the first input signal. This means, in particular, that no signal components of further signals beyond the signal components of the first input signal are included in the reference signal, and that the first input signal is preferably used as the reference signal, either directly or after single-channel signal processing. The effect of directional microphony thus results in the formation of the first frequency-domain directional signal in the frequency domain (which contains the frequency-domain reference signal, i.e. the transformed first input signal in the frequency domain), which can be compared with the transformed reference signal. The resulting first amplification factors then carry the full spectral information as to how the directional microphony affects the transformed reference signal.A time filter, which is created by mapping these first gain factors into the time domain, then transfers this spectral information to the reference signal.

[0025] In an advantageous embodiment, the reference signal is formed from the first input signal and the second input signal using directional microphony in the time domain as a time-directional signal. The directional microphony in the time domain can be implemented in particular as a time-delayed and possibly differently weighted superposition of the two input signals (where the time delay is implemented in the time domain and is in particular the same for all spectral components of the two input signals). By using such a time-directional signal as a reference signal, the overall directivity can be further enhanced, whereby, for example, broadband and / or dominant interference noises that are highly localized can be removed in the time-directional signal.A frequency band-wise fine tuning of the directional microphone is then carried out by comparing the transformed time-directional signal (i.e. the frequency-domain reference signal) with the first frequency-domain directional signal via the resulting time filter.

[0026] Preferably, the microphone arrangement for implementing the method further comprises a third microphone for generating a third input signal from the ambient sound, wherein the third input signal is transformed into the frequency domain, and the first frequency-domain directional signal is also formed based on the thus transformed third input signal in the frequency domain. In particular, the microphone arrangement can accordingly comprise a fourth or further microphones. The described method can be readily applied to such a microphone arrangement with three or more microphones.

[0027] The invention further mentions a method for directional signal processing in a hearing instrument, wherein the hearing instrument comprises a microphone arrangement with at least one first microphone for generating a first input signal from ambient sound and a second microphone for generating a second input signal from the ambient sound, as well as a control unit, and wherein, based on the first input signal and the second input signal, an output signal of the hearing instrument intended for reproduction is generated according to the method described above. The advantages specified for the method for directional signal processing of signals from a microphone arrangement and for its further developments can be transferred analogously to the method for directional signal processing in a hearing instrument.The method for directional signal processing of signals from a microphone arrangement is therefore applied in particular to the input signals of a microphone arrangement which is part of a hearing instrument.

[0028] The invention further relates to a hearing instrument with a microphone arrangement comprising at least a first microphone for generating a first input signal from ambient sound and a second microphone for generating a second input signal from the ambient sound, and further comprising a control unit, wherein the control unit is configured to carry out the aforementioned method based on the first and second input signals. The hearing instrument shares the advantages of the method for directional signal processing of signals from a microphone arrangement. The advantages specified for said method and for its further developments can be analogously transferred to the hearing instrument. In particular, the hearing instrument can be designed as a hearing aid intended and configured to treat a hearing impairment.

[0029] An exemplary embodiment of the invention is explained in more detail below with reference to a drawing. The drawings schematically show: Fig. 1 shows a block diagram of a hearing instrument with a microphone arrangement and a directional signal processing according to the prior art, Fig. 2 shows an alternative embodiment of the signal processing of the hearing instrument according to Figure 1 , according to the state of the art, Fig. 3 in a block diagram an embodiment of the directional signal processing of the hearing instrument according to Figure 1 with reduced latency, and Fig. 4 in a block diagram a Figure 3 alternative design of directional signal processing with reduced latency.

[0030] Corresponding parts and sizes are provided with the same reference numerals in all figures.

[0031] In Figure 1A block diagram of a hearing instrument 1 is shown schematically, which comprises a microphone arrangement 2 with a first microphone 4 and a second microphone 6. The signal processing of the device described below Figure 1The hearing instrument 1 shown is designed according to the prior art. The first microphone 4 is configured to generate a first input signal 8 from ambient sound 7 impinging on the microphone arrangement 2. Correspondingly, the second microphone 6 is configured to generate a second input signal 10 from the ambient sound 7. A possible pre-amplification and / or digitization of the first or second input signal 8, 10 is already incorporated in the corresponding first or second microphone 4, 6. The first and second input signals 8, 10 are each fed to a first filter bank 12, where they are each transformed into the frequency domain, so that a transformed first input signal 14 and a transformed second input signal 16 are generated.

[0032] InIn a directional microphone module 18, a first frequency-domain directional signal 20 is formed from the transformed first input signal 14 and the transformed second input signal 16 in the frequency domain. To form the first frequency-domain directional signal 20, any algorithm suitable for forming a frequency-band-wise directional signal can be used in the directional microphone module 18, in particular delay-and-sum beamforming, delay-and-subtract beamforming, adaptive differential directional microphones or the like. The first frequency-domain directional signal 20 is subjected to noise suppression 22, in which a useful signal component and an interference signal component are estimated, in particular for the individual frequency bands, and depending on the said useful signal orAn amplification factor is determined for each frequency band so that frequency bands with a high useful signal component are relatively boosted and frequency bands with a high interference signal component are relatively lowered.

[0033] After the noise suppression 22, the resulting signal is fed to an amplification module 24, which can in particular comprise an AGC for dynamic compression of the frequency band-wise signal components, and can compensate for an individual hearing impairment of a user of the hearing instrument 1 by means of a corresponding adaptation of frequency band-wise amplification factors.

[0034] The amplification module 24 produces a processed second frequency-domain directional signal 26. This second frequency-domain directional signal 26 is fed to a synthesis filter bank 28, which combines the frequency-band signal components of the second frequency-domain directional signal 26 and transforms them into the time domain. The output signal 30 resulting from said transformation is converted by a loudspeaker 32 of the hearing instrument 1 into an output sound signal 34. The signal processing described here from the first filter bank 12 to the synthesis filter bank 28 preferably takes place on a correspondingly configured signal processor or a processor unit comprising such a signal processor. Such a processor unit is Figure 1 schematically shown as a control unit 35.

[0035] Due to the first filter bank 12 and the synthesis filter bank 28, the output signal 30 inevitably experiences a latency relative to the two input signals 8, 10, and thus relative to the ambient sound 7, which latency increases the higher the frequency resolution of the first filter bank 12. This latency is typically approximately 4 to 7 ms. While a significant reduction in latency can be achieved by reducing the frequency resolution, this also comes at the expense of the beamforming capabilities, as well as the ability to suppress noise components in the two input signals 8, 10 by means of the noise suppression 22, and, if necessary, to individually adapt the amplification in the amplification module 24 to the user of the hearing instrument 1.

[0036] In Figure 2 is a schematic block diagram of a modification of the hearing instrument 1 according to Figure 1which attempts to solve the described latency problems that occur as a result of the first filter bank 12 and the synthesis filter bank 28. The Figure 2 The signal processing shown is designed according to the state of the art. In the hearing instrument 1 according to Figure 2 A time-domain directional signal 38 is formed from the first input signal 8 and the second input signal 10 by means of a time-domain directional microphone module 36. The time-domain directional signal 38 is transformed into the frequency domain by the first filter bank 12, and a resulting transformed time-domain signal 40 is fed to the noise suppression 22 and subsequently to the amplification module 24. Frequency band-wise gain factors gj are determined, which are mapped into the time domain onto a time filter 44 by means of a mapping 42, which may in particular comprise a Fourier transformation. The control unit 35 according to Figure 1is not shown in Figure 35.

[0037] The time filter 44 thus implicitly incorporates the properties of the frequency-band-wise gain factors gj and their effects on the transformed time-directed signal 40, but now in the time domain. Accordingly, the time filter 44 is applied to the (original) time-directed signal 38 in the time domain to generate the output signal 30. In particular, the time filter 44 is defined as a minimum-phase filter.

[0038] While the results of Figure 2 described processing of the input signals 8, 10 reduces the latency compared to the embodiment according to Figure 1can be reduced, however, due to the necessarily broadband directional microphone technology in the time domain, which is applied to the input signals 8, 10 in the corresponding module 38, there are no options available to carry out frequency-selective signal processing (e.g. for noise suppression) already in the directional microphone technology.

[0039] In Figure 3 A block diagram of a hearing instrument 1 is shown schematically, in which latency is to be kept as low as possible for frequency-selective, direction-dependent signal processing. The first input signal 8 is used as a reference signal 46. The effects of frequency-selective directional microphony and noise suppression on the individual frequency bands are determined in a manner to be described below using the reference signal 46, and thus the time filter 44 to be applied to the reference signal 46 is determined.

[0040] The first input signal 8 as the reference signal 46 and the second input signal 10 are, similar to that based on Figure 1 illustrated embodiment, is transformed into the frequency domain by the first filter bank 12, thereby generating the transformed first input signal 14 as a frequency-domain reference signal 48 or the transformed second input signal 16. In the present embodiment, the directional microphone module 18 also subsequently generates the first frequency-domain directional signal 20 from the transformed first input signal 14—that is, the frequency-domain reference signal 48—and from the transformed second input signal.

[0041] To determine the time filter 44, which is preferably to be defined as a minimum-phase filter, first gain factors g1j are obtained frequency band by frequency band, which are determined by a spectral division 45 of the frequency-domain reference signal 48 and a second frequency-domain directional signal 50 derived from the first frequency-domain directional signal 20. In particular, the magnitudes of the frequency-domain reference signal 48 and the second frequency-domain directional signal 50 derived therefrom, or also quantities derived from the magnitudes, can be divided frequency band by frequency band to generate the first gain factors g1j.

[0042] Said second frequency-domain directional signal 50 is generated by feeding the first frequency-domain directional signal 20 to the noise suppression 22 and the amplification module 24, where the first frequency-domain directional signal 20 is amplified or attenuated frequency-band-wise by applying second gain factors g2j to the first frequency-domain directional signal 20, frequency-band-wise. The second gain factors g2j can, for example, be formed in each frequency band from the successive application of the individual factors determined in the noise suppression 22 and in the amplification module 24 for the respective frequency band.

[0043] The spectral division 45 de facto determines in what way the signal processing applied to the first frequency-domain directional signal 20, which takes place in the noise suppression 22 and in the amplification module 24, is to be modified or compensated if the input variable is not said first frequency-domain directional signal 20, but instead the frequency-domain reference signal 48. If the first amplification factors g1j resulting from the spectral division 45 were applied to the frequency-domain reference signal 48, the resulting signal would correspond in magnitude to the second frequency-domain directional signal 50, which results from the application of the noise suppression 22 and the amplification module 24 (or from the second amplification factors g2j determined there) to the first frequency-domain directional signal 20.

[0044] The first gain factors g1j resulting from the spectral division 45 are now mapped from the frequency domain to the time-domain filter 44, which is preferably provided by an FIR filter, using the mapping 42. The time-domain filter 44 is thus the time-domain equivalent of the just-described "modification" or "compensation" of the signal processing of the first frequency-domain directional signal 20, which must be applied to the frequency-domain reference signal 46. In this respect, the influence of the transformed input signal 16 on the second frequency-domain directional signal 50 is also incorporated into the time-domain filter 44 via the spectral division 45. Accordingly, the output signal is generated from an application of the time-domain filter 44 to the frequency-domain directional signal 48.

[0045] The time filter 46 in the time domain allows the latency to be kept very low, since latencies which arise, for example, from the first filter bank 12, do not affect the propagation of the reference signal 46 through the signal flow, but merely result in the time filter 44, which is applied to the reference signal 46, no longer being "up-to-date" by the amount of the latency, which, however, is a trade-off compared to the significantly reduced latency of the output signal 30 compared to the embodiment according to Figure 1 can be accepted.

[0046] In Figure 4 is a schematic block diagram of an alternative embodiment of the Figure 3 described signal processing, which also includes elements of the embodiment according to Figure 2 by applying the time filter 44 to a directional signal in the time domain in a manner to be shown.

[0047] A time-domain directional signal 38 is first generated from the first input signal 8 and the second input signal 10 by means of the time-domain directional microphone module 36. This can be achieved, for example, by delaying one of the two input signals 8, 10 relative to the other, which may vary over time but always has the same effect on all signal components (and is therefore, in particular, frequency-independent). In particular, the time-domain directional signal 38 can also be generated by applying an all-pass filter with a frequency-dependent delay to one of the two input signals 8, 10 in the time-domain directional microphone module 36, so that the time-domain directional signal 38 itself can already exhibit a certain frequency dependence with regard to its directivity.

[0048] The first and second input signals 8, 10 are also transformed into the frequency domain by means of the first filter bank 12, and the first frequency domain directional signal 20 is generated from the thus generated transformed first and second input signals 14, 16 by means of the directional microphone module 18 in the frequency domain.

[0049] The time-domain directional signal 36 generated as described above serves in the present embodiment as the reference signal 46, which is transformed into the frequency domain by means of a second filter bank 52, whereby the transformed time-domain directional signal 40 is generated as a frequency-domain reference signal 48. This and the first frequency-domain directional signal 20 are subjected to spectral division 45 for comparison with each other, whereby the first gain factors g1j for the respective frequency bands are determined.

[0050] For the first frequency-domain directional signal 20 generated as described above, second amplification factors g2j are determined by the noise suppression 22 and by the amplifier module 24, which second amplification factors would have to be applied to the first frequency-domain directional signal 20 accordingly in order to achieve the noise suppression effect of the noise suppression 22 or the amplification effect of the amplification module 24 for the first frequency-domain directional signal 20.

[0051] In contrast to the embodiment according to Figure 3However, this noise suppression effect or amplification effect is not achieved directly in the first frequency-domain directional signal 20. Rather, the second amplification factors g2j corresponding to said effects, which were determined in the noise suppression 22 and in the amplifier module 24, together with the first amplification factors g1j, which were obtained from the spectral division 45 of the frequency-domain reference signal 48 and the first frequency-domain directional signal 20, are now mapped into the time domain by the mapping 42 onto the time filter 44. The time filter 44 thus determined, which is also preferably designed as an FIR filter here, is then applied to the reference signal 46 in the time domain—that is, to the time-domain directional signal 38—and thereby generates the output signal 30. Finally, the output signal 30 is converted into the output sound signal 34 by the loudspeaker 32.

[0052] By the spectral division 45 in the embodiment according to Figure 4determines the extent to which the frequency-selective beamforming of the directional microphone module 18 (frequency domain) differs from the broadband beamforming of the time-domain directional microphone module 36, so that the first gain factors g1j generated thereby de facto represent the amount in the instantaneous gain by which the transformed time-directional signal 40 is to be compensated frequency band by frequency in order to obtain the intrinsic, direction-sensitive sound behavior inherent in the first frequency-domain directional signal 20.This direction-sensitive sound behavior characterized by the first gain factors g1j is thus mapped to the time filter 44 together with the noise suppression and amplification effect of the noise suppression 22 and the amplification module 24 characterized by the second gain factors g2j in the time domain, so that the said sound behavior and the said effects can be achieved by applying the time filter 44 to the correspondence of the transformed time-directional signal 40 in the time domain, i.e. exactly to the time-directional signal 38.

[0053] By applying the time filter 44 as described above, the latency of the output signal 30 relative to the two input signals 8, 10 and thus relative to the ambient sound 7 can also be reduced in this embodiment compared to the embodiment according to Figure 1 be kept very low.

[0054] Although the invention has been illustrated and described in detail by the preferred embodiment, the invention is not limited by the disclosed examples and other variations may be derived therefrom by those skilled in the art without departing from the scope of the invention, which scope is defined by the claims. List of reference symbols

[0055] 1 Hearing instrument 2 Microphone array 4 First microphone 6 Second microphone 7 Ambient sound 8 First input signal 10 Second input signal 12 First filter bank 14 Transformed first input signal 16 Transformed second input signal 18 Directional microphone module 20 First frequency-domain directional signal 22 Noise suppression 24 Amplification module 26 Second frequency-domain directional signal 28 Synthesis filter bank 30 Output signal 32 Loudspeaker 34 Output sound signal 35 Control unit 36 Time-domain directional microphone module 38 Time-domain directional signal 40 Transformed time-domain directional signal 42 Illustration 44 Time filter 45 Spectral division 46 Reference signal 48 Transformed reference signal 50 Second frequency-domain directional signal 52 Second filter bank 1 First gain factors g2jsecond gain factors gjfrequency band gain factors

Claims

1. Method for directional signal processing of signals in a microphone array (2), which comprises at least one first microphone (4) for generating a first input signal (8) from an ambient sound (7) and a second microphone (6) for generating a second input signal (10) from the ambient sound (7), wherein the first input signal (8) is used to form a reference signal (46), and wherein a time filter (44) is generated in the time domain, characterized in that - the reference signal (46) is transformed into the frequency domain, thereby generating a frequency-space reference signal (48), - the first input signal (8) and the second input signal (10) are each transformed into the frequency domain, and a first frequency-space directional signal (20) is formed in the frequency domain using the transformed first input signal (14) and the transformed second input signal (16), - by means of a frequency-resolved comparison of the frequency-space reference signal (48) with the first frequency-space directional signal (20) or a signal derived from the first frequency-space directional signal (20) in the frequency domain, first frequency-dependent first gain factors (g1j) are generated, - in that the time filter (44) in the time domain is generated by means of the first gain factors (glj), and - the reference signal (46) is filtered by means of the time filter (44), and an output signal (30) is generated from the reference signal (46) filtered by means of the time filter (44).

2. Method according to Claim 1, wherein the frequency-resolved comparison of the frequency-space reference signal (48) with the first frequency-space directional signal (20) or the signal derived from the first frequency-space directional signal (20) in the frequency domain is carried out using a spectral division (45), on the basis of which the frequency-dependent first gain factors (g2j) are generated.

3. Method according to Claim 1 or Claim 2, wherein as a signal derived from the first frequency-space directional signal (20) in the frequency domain for the frequency-resolved comparison with the frequency-space reference signal (48), a second frequency-space directional signal (50) is generated by applying frequency-dependent second gain factors (g2j) to the first frequency-space directional signal (20).

4. Method according to Claim 3, wherein the time filter (44) is formed using a mapping (42) of the frequency-dependent first gain factors (g1j) into the time domain.

5. Method according to Claim 1 or Claim 2, wherein frequency-dependent second gain factors (g2j) are determined for the first frequency-space directional signal (20), and wherein the time filter (44) is formed using a common mapping (42) of the first gain factors (g1j) and the second gain factors (g2j) into the time domain.

6. Method according to any one of Claims 3 to 5, wherein the frequency-dependent second gain factors (g2j) for the first frequency-space directional signal (20) are determined by means of noise suppression (22) and / or dynamic compression and / or a hearing impairment to be corrected of a recipient of the output signal (30).

7. Method according to any one of Claims 1 to 4 or Claim 6, wherein the reference signal (46) is formed from signal components of the first input signal (8) only.

8. Method according to any one of Claims 1, 2, 5 or 6, wherein the reference signal (46) is formed in the time domain from the first input signal (8) and the second input signal (10) as a time directional signal (38) by means of directional microphony.

9. Method according to any one of the previous claims, wherein in order to carry out the method the microphone array (2) also comprises a third microphone for generating a third input signal from the ambient sound (7), wherein the third input signal is transformed into the frequency domain, and the first frequency-space directional signal (20) is also formed in the frequency domain from the transformed third input signal.

10. Method for directional signal processing in a hearing instrument (1), wherein the hearing instrument (1) comprises a microphone array (2) having at least one first microphone (4) for generating a first input signal (8) from an ambient sound (7) and a second microphone (6) for generating a second input signal (10) from the ambient sound (7), as well as a control unit (35), and wherein from the first input signal (8) and the second input signal (10), an output signal (30) of the hearing instrument (1) intended for reproduction is generated according to the method according to any one of the previous claims.

11. Hearing instrument (1) having - a microphone array (2), which comprises at least one first microphone (4) for generating a first input signal (8) from an ambient sound (7) and a second microphone (6) for generating a second input signal (10) from the ambient sound (7), and - a control unit (35), wherein the control unit (35) is configured to carry out the method according to Claim 10 using the first and the second input signal (8, 10).