Method for directional signal processing of signals of a microphone arrangement

Abstract

The invention relates to a method for directional signal processing of signals of a microphone arrangement comprising a first microphone generating a first input signal from ambient sound and a second microphone generating a second input signal from ambient sound, wherein a reference signal is formed from the first input signal, wherein the reference signal is transformed into the frequency domain and a frequency space reference signal is generated, wherein the first and second input signal are each transformed into the frequency domain and a first frequency space directional signal is formed in the frequency domain from the so transformed first and second input signal, wherein a frequency-dependent first amplification factor is generated from a frequency-resolved comparison of the frequency space reference signal with the first frequency space directional signal or a signal derived from the first frequency space directional signal in the frequency domain, wherein a time filter is generated in the time domain from the first amplification factor, and wherein the reference signal is filtered by means of the time filter and an output signal is generated from the filtered reference signal.

Description

Technical Field

[0001] The present invention relates to a method for directional signal processing of a signal from a microphone device, the microphone device comprising at least one first microphone for generating a first input signal based on ambient sound and a second microphone for generating a second input signal based on ambient sound, wherein the first input signal and the second input signal are respectively transformed into the frequency domain, and a first frequency spatial directional signal is formed in the frequency domain by means of the first input signal and the second input signal thus transformed, wherein the signal derived in the frequency domain generates a first frequency-dependent amplification factor, and wherein an output signal is generated based on the first amplification factor and based on the frequency spatial directional signal. Background Technology

[0002] Frequency-resolved signal processing for microphone devices with one or more microphones is typically performed in the frequency domain by decomposing the signals from one or more microphones into separate frequency bands. The signal components of each frequency band are processed separately, and amplified and / or compressed in particular, and may be combined into a directional signal if necessary. Subsequently, the individual signal components in the frequency bands are “synthesized” into a single output signal in the time domain.

[0003] Decomposing the signal into individual frequency bands (combined with the final synthesis) results in a latency (Latenz) for signal processing, which depends on the desired frequency resolution in the frequency domain and is typically around 5-8 ms.

[0004] In real-time applications such as hearing instruments, especially hearing aids, this latency often causes the delayed output signal (which is reproduced by the hearing instrument) to be superimposed on the direct sound of the surrounding environment, potentially resulting in a comb filtering effect. These effects are typically more pronounced the closer the direct sound of the environment is to the output signal processed and reproduced by the hearing instrument.

[0005] Comb filter effects are generally considered unpleasant and are therefore preferred to be avoided. One possibility is to significantly reduce or, as far as possible, eliminate latency in signal processing. However, this significantly limits the possibilities in frequency-resolved signal processing, as a sub-signal path (“analysis path”) is often used for this purpose, in which signal processing is determined, just as it would be applied to the signal components of the main signal path. In particular, in this scenario, it is difficult only to perform direction-dependent signal processing on the individual input signals generated by multiple microphones of a microphone device. Summary of the Invention

[0006] Therefore, the technical problem to be solved by the present invention is to provide a method for directional signal processing of a signal from a microphone device, wherein the method shall have the smallest possible waiting time for the resulting output signal relative to the input signal to be processed by the microphone device.

[0007] According to the present invention, the above-mentioned technical problem is solved by a method for performing directional signal processing on signals of a microphone device, the microphone device including at least one first microphone for generating a first input signal based on ambient sound and a second microphone for generating a second input signal based on ambient sound, wherein a reference signal is formed based on the first input signal, wherein the reference signal is transformed into the frequency domain, and thereby a frequency space reference signal is generated, wherein the first input signal and the second input signal are respectively transformed into the frequency domain, and a first frequency space directional signal is formed in the frequency domain based on the first input signal and the second input signal thus transformed.

[0008] This specification stipulates that a frequency-dependent first amplification factor is generated based on a comparison of the frequency resolution of a frequency space reference signal with a first frequency space orientation signal or with a signal derived from the first frequency space orientation signal in the frequency domain; a time filter is generated in the time domain based on the first amplification factor; and the reference signal is filtered by the time filter, and an output signal is generated based on the reference signal filtered by the time filter. Advantageous and inventive designs are inherently part of the subject matter of this invention.

[0009] The microphone, as used herein and hereinafter, shall include any electroacoustic input transducer configured to generate a corresponding input signal based on ambient sound, wherein sound pressure fluctuations of the ambient sound are converted into corresponding voltage or current fluctuations by the input transducer. The microphone arrangement shall accordingly be understood as any spatial arrangement of at least two such input transducers, wherein the spatial spacing between the input transducers enables direction-dependent signal processing.

[0010] The formation of a reference signal based on a first input signal includes two main scenarios: on the one hand, the reference signal is formed solely from the signal components of the first input signal, while other signal components, particularly those of the second input signal, are not included in the reference signal; or on the other hand, the reference signal can be formed from the signal components of two input signals, particularly by superimposing the time delays of the two input signals. The reference signal here is a signal in the time domain.

[0011] The aforementioned signal-to-frequency domain transformation, namely the transformation of the current reference signal and the first and second input signals, can be performed in particular by means of a suitably configured filter bank. In the following text, "frequency domain" also includes, in particular, the time-discrete time-frequency domain, in which the spectral components of each transformed signal are updated according to a time variable.

[0012] The first frequency spatial orientation signal can be generated in particular from the transformed first input signal and the transformed second input signal in such a way that the orientation characteristics formed by the first frequency spatial orientation signal vary across the frequency bands (and, for example, a linear combination of the two transformed input signals, or a linear combination of intermediate signals derived from the two input signals, such as cardioid / anticardioid signals, have different linear coefficients across the frequency bands).

[0013] The frequency space reference signal is compared in the frequency domain with a first frequency space oriented signal or a signal derived therefrom in a frequency-resolved manner, preferably band-by-band, thereby generating a frequency-dependent first amplification factor as a quantitative result of the comparison. This comparison can be performed, for example, by spectrale division (the first amplification factor can then be directly determined band-by-band as a corresponding quotient formed when the frequency space reference signal is divided by the first frequency space oriented signal). Other comparisons are also conceivable, for example, in the form of relative deviations and / or as a nonlinear, preferably monotonic, function of the band-by-band deviations between the two signals. To compare the frequency space reference signal with the signal derived in the frequency domain from the first frequency space oriented signal, a second frequency space oriented signal is preferably used, whose signal components can be directly derived from the first frequency space oriented signal, and preferably without the addition of other signal components, i.e., derived, for example, through band-by-band amplification of the first frequency space oriented signal.

[0014] The first amplification factor, a measure representing the deviation between the first frequency space orientation signal and the frequency space reference signal in the corresponding frequency band (and implicitly in the case of the derived signal if necessary), is used to generate a time filter in the time domain. Here, the time filter is preferably a filter with a finite impulse response (FIR). Specifically, the time filter can be generated here by transforming a transfer function from the frequency domain to the time domain, which represents the band-by-band application of the first amplification factor in the frequency domain. Particularly preferably, the time filter is generated here as a minimum phase filter, which therefore has the minimum possible latency for a given numerical frequency response.

[0015] The reference signal (in the time domain) is filtered using the resulting time filter, producing an output signal. This output signal can then be transmitted to a receiver for recording or further use. If a microphone is used in the hearing instrument, the output signal is preferably converted into an output sound signal via a loudspeaker (in the broadest sense, the electroacoustic output converter of the hearing instrument). Prior to this conversion, the output signal may undergo further signal processing, such as to suppress acoustic feedback that may occur between the loudspeaker and microphone.

[0016] Here, 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. Based on the mapping of the first amplification factor (obtained from the first frequency spatial orientation signal) to the time filter, information about the direction-dependent acoustic images in each frequency band can be realized in the time filter, as it is obtained by generating the first frequency spatial orientation signal. However, the difference here is that the directional signal processing of the first frequency spatial orientation signal differs from that of the reference signal (transformed to the frequency domain).

[0017] Therefore, based on a comparison between the frequency space reference signal and the first frequency space directional signal, or preferably a signal directly derived from it, it is determined which band-wise amplification factors can be applied to the frequency space reference signal (which actually only represents the reference signal transformed into the frequency domain) in order to approximate as well as possible the band-wise acoustic characteristics present for the first frequency space directional signal, especially in terms of signal volume and level. Here, the application of time filters, especially FIR filters, in the time domain corresponds to the application of frequency- and time-dependent attenuation or amplification factors in each band and subsequent signal synthesis. Thus, the band-wise directional correlation encoded in the first frequency space directional signal is "encoded" as a first amplification factor for the frequency space reference signal. By mapping these first amplification factors (and thus the encoded information about the directional effect described) onto the time filter, this directional information can therefore be provided to the reference signal with a very low latency.

[0018] Preferably, the comparison of the frequency resolution of the frequency space reference signal with the first frequency space orientation signal or with the signal derived from the first frequency space orientation signal in the frequency domain is performed by spectral division, and a frequency-dependent first amplification factor is generated according to the spectral division. The comparison by spectral division is particularly efficient on the one hand, and on the other hand, it takes into account that the first amplification factor is therefore a transfer function between the two mentioned signals in the frequency domain. Here, the frequency-dependent first amplification factor is generated, in particular, by dividing the value of the frequency space reference signal and the value of the first frequency space orientation signal or the signal derived from the first frequency space orientation signal in the frequency domain.

[0019] Advantageously, as a signal derived in the frequency domain from the first frequency spatial orientation signal, a second frequency spatial orientation signal is generated for frequency-resolved comparison with a frequency spatial reference signal. This second frequency spatial orientation signal is generated by applying a frequency-dependent second amplification factor to the first frequency spatial orientation signal. The first frequency spatial orientation signal can be additionally noise-suppressed, for example, by means of a Wiener filter, and, if necessary, dynamically compressed, so that the frequency spatial reference signal used for the first amplification factor is compared with the formed second frequency spatial orientation signal.

[0020] Suitablely, the time filter is formed here based on a mapping from the frequency-dependent first amplification factor to the time domain. The term "mapping" here specifically refers to the transformation from the frequency domain to the time domain of the transfer function corresponding to the first amplification factor, or the transfer function corresponding to the product of the first amplification factor and another band-wise amplification factor. In this way, it can be effectively ensured that a filter with the correct characteristics in the time domain—that is, a filter with the characteristics given by the first amplification factor in the frequency domain—is applied to the reference signal. Here, the time filter is generated, in particular, as an FIR filter.

[0021] Advantageously, a frequency-dependent second amplification factor for the first frequency spatial directional signal is determined, wherein a time filter is formed based on a common mapping of the first and second amplification factors in the time domain. Specifically, the frequency-dependent second amplification factor for the first frequency spatial directional signal is determined here based on noise suppression and / or dynamic compression and / or the hearing impairment to be corrected at the receiver of the output signal. Here, the receiver of the output signal is particularly a user of a hearing instrument including a microphone.

[0022] In other words, this means that noise suppression and / or dynamic compression applied to the first frequency spatial directional signal in relation to the frequency band, corresponding to a preset for the signal-to-noise ratio (SNR) or a preset for the maximum signal level in the frequency band (which can also be individually coordinated with the hearing loss of the user of the hearing aid including the microphone), determine the instantaneous second amplification factor for the first frequency spatial directional signal. However, these are not applied to the first frequency spatial directional signal. Instead, the first amplification factor, formed by comparing the first frequency spatial directional signal with a transformed reference signal, is mapped together with the second amplification factor into the time domain to form a time filter. Therefore, the characteristics of noise suppression or dynamic compression do not enter the time filter through a comparison of the transformed reference signal with a second frequency spatial directional signal (which is generated by applying noise suppression or dynamic compression to the first frequency spatial directional signal), but directly into the time filter through the mapping of the second amplification factor determined for noise suppression or dynamic compression into the time domain.

[0023] It has proven advantageous that the reference signal is formed solely from the signal components of the first input signal. This specifically means that no signal components other than those of the first input signal enter the reference signal, and preferably, the first input signal is used as the reference signal either directly or after single-channel signal processing. Therefore, the role of the directional microphone is to form a first frequency-space directional signal in the frequency domain (which indeed contains the frequency-space reference signal, i.e., the transformed first input signal in the frequency domain), which is compared with the transformed reference signal. The resulting first amplification factor then carries full-spectrum information about how the directional microphone affects the transformed reference signal. This spectral information is then transmitted to the reference signal via a time filter generated by mapping these first amplification factors into the time domain.

[0024] In an advantageous design, the reference signal is formed into a time-directed signal in the time domain using a directional microphone based on the first and second input signals. Here, the time-domain directional microphone can be implemented by time-delaying and, if necessary, superimposing the two input signals with different weights (wherein the time delay is implemented in the time domain and is identical for all spectral components of both input signals). By using this time-directed signal as the reference signal, the overall directivity can also be enhanced, where, for example, strongly localized broadband and / or dominant interference noise can be removed from the time-directed signal. The directional microphone is then fine-tuned band-by-band by a time filter formed thereby, based on a comparison between the transformed time-directed signal (i.e., the frequency-space reference signal) and the first frequency-space directional signal.

[0025] Preferably, the microphone device for performing the method further includes a third microphone for generating a third input signal based on ambient sound, wherein the third input signal is transformed into the frequency domain, and a first frequency spatial orientation signal is also formed in the frequency domain based on the transformed third input signal. In particular, the microphone device may correspondingly include a fourth microphone or additional microphones. The method described herein can be readily adapted to such microphone devices having three or more microphones.

[0026] The present invention also proposes a method for directional signal processing in a hearing instrument, wherein the hearing instrument includes a microphone device having at least one first microphone for generating a first input signal based on ambient sound and a second microphone for generating a second input signal based on ambient sound, and further includes a control unit, wherein, according to the above method, an output signal for reproduction by the hearing instrument is generated based on the first and second input signals. The advantages of the method for directional signal processing of signals from a microphone device and its improvements can be comparatively applied to the method for directional signal processing in a hearing instrument. Therefore, the method for directional signal processing of signals from a microphone device is particularly applicable to the input signals of the microphone device, which is part of the hearing instrument.

[0027] The present invention also proposes a hearing instrument having a microphone device comprising at least one first microphone for generating a first input signal based on ambient sound and a second microphone for generating a second input signal based on ambient sound, and further comprising a control unit configured to perform the above-described method based on the first and second input signals. The hearing instrument shares the advantages of the method for directional signal processing of the signals from the microphone device. The advantages described herein for the method and its improvements can be comparatively applied to the hearing instrument. In particular, the hearing instrument can be designed as a hearing aid, configured and designed to address hearing loss. Attached Figure Description

[0028] Embodiments of the present invention will now be described in detail with reference to the accompanying drawings. The following are schematic illustrations:

[0029] Figure 1 A block diagram illustrates a hearing instrument with a microphone and directional signal processing according to the prior art.

[0030] Figure 2 A block diagram illustrates the prior art. Figure 1 A design scheme to replace the signal processing of hearing instruments.

[0031] Figure 3 The diagram illustrates the basis for reduced latency. Figure 1 The design scheme for directional signal processing of hearing instruments, and

[0032] Figure 4 The block diagram illustrates directional signal processing with reduced latency compared to... Figure 3 The alternative design scheme.

[0033] Corresponding parts and dimensions have the same reference numerals in all the accompanying drawings. Detailed Implementation

[0034] exist Figure 1 The diagram schematically illustrates a hearing instrument 1, which includes a microphone device 2 having a first microphone 4 and a second microphone 6. Figure 1 The signal processing described below for the hearing instrument 1 shown is designed according to the prior art. A first microphone 4 is configured to generate a first input signal 8 based on ambient sound 7 hitting the microphone device 2. Correspondingly, a second microphone 6 is configured to generate a second input signal 10 based on ambient sound 7. Possible pre-amplification and / or digitization of the first or second input signals 8, 10 are included herein in the respective first or second microphones 4, 6. The first and second input signals 8, 10 are respectively fed to a first filter bank 12, and there are respectively transformed into the frequency domain to generate a transformed first input signal 14 or a transformed second input signal 16.

[0035] In the directional microphone module 18, a first frequency spatial directional signal 20 is formed in the frequency domain from the transformed first input signal 14 and the transformed second input signal 16. To form the first frequency spatial directional signal 20, any algorithm suitable for forming a band-by-band directional signal can be used in the directional microphone module 18, particularly delay-and-sum-beamforming, delay-and-subtract-beamforming, adaptive differential directional microphones, etc. The first frequency spatial directional signal 20 undergoes noise suppression 22, wherein, in particular, useful signal components and interference signal components are estimated separately for each frequency band, and the amplification factor for each frequency band is determined based on the useful signal components or interference signal components, such that frequency bands with high useful signal components are relatively increased and frequency bands with high interference signal components are relatively decreased.

[0036] After noise suppression 22, the resulting signal is fed to amplification module 24, which may include, in particular, an AGC for dynamically compressing the signal components per frequency band, and a mechanism to compensate for the individual hearing loss of the user of the hearing instrument 1 by corresponding adaptation of the amplification factor per frequency band.

[0037] A processed second frequency spatial orientation signal 26 is generated by amplification module 24. This second frequency spatial orientation signal 26 is fed to a synthesis filter bank 28, which combines the frequency-band-wise signal components of the second frequency spatial orientation signal 26 and transforms them into the time domain. The output signal 30 formed by this transformation is converted into an output sound signal 34 by the speaker 32 of the hearing instrument 1. The signal processing described herein from the first filter bank 12 to the synthesis filter bank 28 is preferably performed on a correspondingly configured signal processor or a processor unit including such a signal processor. Such a processor unit... Figure 1 The control unit 35 is schematically shown in the diagram.

[0038] Through 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 therefore relative to the ambient sound 7. The higher the frequency resolution of the first filter bank 12, the greater this latency. This latency is typically about 4 to 7 ms. Although a significant reduction in latency can be achieved by reducing the frequency resolution, this comes at the cost of sacrificing the possibility of beamforming and suppressing interfering signal components in the two input signals 8, 10 by means of noise suppression 22, and, if necessary, adapting the amplification in the amplification module 24 separately to the user of the hearing instrument 1.

[0039] exist Figure 2 The diagram illustrates the situation based on... Figure 1 A variant of the hearing instrument 1, which attempts to address the described latency problem arising from the first filter bank 12 and the synthesis filter bank 28. Figure 2 The signal processing shown herein is also designed according to existing technology. Figure 2 In the hearing instrument 1, a time-oriented signal 38 is formed from the first input signal 8 and the second input signal 10 using a time-domain directional microphone module 36. The time-oriented signal 38 is transformed into the frequency domain by a first filter bank 12, and the resulting transformed time-oriented signal 40 is sent to a noise suppression 22 and then to an amplification module 24. Here, band-wise amplification factors gj are determined, and these amplification factors are mapped onto a time filter 44 in the time domain using a mapping 42, which may include, in particular, a Fourier transform. Figure 2 According to (not shown in the text) Figure 1 Control unit 35.

[0040] Therefore, the time filter 44 implicitly incorporates the characteristics of the band-by-band amplification factor gj and its effect on the transformed time-oriented signal 40, precisely in the time domain. Accordingly, the time filter 44 is applied to the (original) time-oriented signal 38 in the time domain to thereby generate the output signal 30. In particular, the time filter 44 is determined to be a minimum-phase filter.

[0041] Although by means of Figure 2 The processing of the input signals 8 and 10 described can be relative to... Figure 1 The embodiment reduces the latency, but here, due to the necessary wide-bandwidth directional microphone in the time domain (which is applied to the input signals 8, 10 in the corresponding module 38), there is no possibility that frequency-selective signal processing (e.g. for noise suppression) has already been performed in the directional microphone.

[0042] exist Figure 3 The hearing instrument 1 is schematically shown in a block diagram, in which the latency should be kept as low as possible for the most frequency-selective, direction-dependent signal processing. Here, the first input signal 8 is used as the reference signal 46. The effects of the frequency-selective directional microphone and noise suppression on the various frequency bands are determined here based on the reference signal 46 in a manner further described, and thus the time filter 44 to be applied to the reference signal 46 is determined.

[0043] Similar to according to Figure 1 In the illustrated embodiment, the first input signal 8 and the second input signal 10, serving as reference signal 46, are transformed into the frequency domain by the first filter bank 12, thereby generating a transformed first input signal 14 as either the frequency space reference signal 48 or the transformed second input signal 16. In this embodiment, a first frequency space directional signal 20 is subsequently generated by the directional microphone module 18 based on the transformed first input signal 14 (i.e., the frequency space reference signal 48) and the transformed second input signal.

[0044] To determine the time filter 44 (which is preferably determined to be a minimum phase filter), a first amplification factor g1j is obtained band-by-band. This first amplification factor is determined by a spectral division 45 of the frequency space reference signal 48 and the second frequency space orientation signal 50 derived from the first frequency space orientation signal 20. In particular, the values ​​of the frequency space reference signal 48 and the derived second frequency space orientation signal 50, or parameters derived from these values, can also be divided band-by-band to produce the first amplification factor g1j.

[0045] The second frequency spatial orientation signal 50 is generated as follows: the first frequency spatial orientation signal 20 is fed to the noise suppression 22 and the amplification module 24, where the first frequency spatial orientation signal 20 is amplified or reduced band-by-band by applying a band-by-band second amplification factor g2j to the first frequency spatial orientation signal 20. The second amplification factor g2j can, for example, be formed in each band by successively applying various coefficients, which have been determined for the corresponding band in the noise suppression 22 and the amplification module 24, respectively.

[0046] If the input parameter is not the first frequency space orientation signal 20, but instead the frequency space reference signal 48, then the signal processing applied to the first frequency space orientation signal 20 in the noise suppression 22 and amplification module 24 is effectively determined by the spectrum division 45, showing how the signal processing is modified or compensated. If the first amplification factor g1j obtained by the spectrum division 45 is applied to the frequency space reference signal 48, the resulting signal is numerically identical to the second frequency space orientation signal 50, which is obtained by applying the noise suppression 22 and amplification module 24 (or the second amplification factor g2j determined therein) to the first frequency space orientation signal 20.

[0047] The first amplification factor g1j generated by the spectrum division 45 is mapped from the frequency domain to a time filter 44 in the time domain via mapping 42, which is preferably provided by an FIR filter. Therefore, the time filter 44 corresponds in the time domain to the previously described "modification" or "compensation" of the signal processing of the first frequency space-oriented signal 20, which must be applied to the frequency space reference signal 46. In this regard, the effect of the transformed input signal 16 on the second frequency space-oriented signal 50 also enters the time filter 44 via the spectrum division 45. Accordingly, the output signal is generated by applying the time filter 44 to the frequency space-oriented signal 48.

[0048] The time filter 46 in the time domain can maintain a very small latency because, for example, the latency generated by the first filter bank 12 does not affect the propagation of the reference signal 46 through the signal stream, but only causes the value of the time filter 44 applied to the reference signal 46 with respect to the latency to no longer be "current". However, this can be used as a reference to... Figure 1 The significantly reduced latency of the output signal 30 in the embodiment is accepted as a trade-off.

[0049] exist Figure 4 The diagram illustrates the underlying principles. Figure 3 The described alternative design scheme for signal processing, which also includes, in accordance with Figure 2An element of one embodiment, namely, is that time filter 44 is applied to the directional signal in the time domain in a manner further shown.

[0050] First, a time-directed signal 38 is generated from the first input signal 8 and the second input signal 10 using a time-domain directional microphone module 36. This can be achieved, for example, by delaying one of the two input signals 8 and 10 relative to the other, whereby the delay, although varying with time, always acts equally on all signal components (and therefore acts particularly independently of frequency). In particular, the time-directed 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 and 10 in the time-domain directional microphone module 36, such that the time-directed signal 38 itself already possesses a certain frequency dependence in terms of directionality.

[0051] Furthermore, the first and second input signals 8 and 10 are transformed into the frequency domain by means of the first filter bank 12, and the transformed first and second input signals 14 and 16 thus generated generate a first frequency spatial directional signal 20 in the frequency domain by means of the directional microphone module 18.

[0052] The time-oriented signal 36 generated as described above is used as a reference signal 46 in this embodiment. This reference signal is transformed into the frequency domain by means of a second filter bank 52, thereby generating a transformed time-oriented signal 40 as a frequency space reference signal 48. The frequency space reference signal 48 and the first frequency space orientation signal 20 are subjected to a spectral division 45 for comparison with each other, thereby determining a first amplification factor g1j for the corresponding frequency band.

[0053] For the first frequency spatial orientation signal 20 generated as described above, a second amplification factor g2j is determined by noise suppression 22 and by amplifier module 24. This second amplification factor will be applied to the first frequency spatial orientation signal 20 accordingly so as to achieve the noise suppression effect of noise suppression 22 or the amplification effect of amplifier module 24 on the first frequency spatial orientation signal 20.

[0054] However, according to Figure 3Unlike other embodiments, the noise suppression or amplification effect is not directly achieved with respect to the first frequency spatial directional signal 20. More precisely, the second amplification factor g2j (which has already been determined in noise suppression 22 and in amplifier module 24) corresponding to the effect, together with the first amplification factor g1j (which has been obtained from the spectral division 45 of the frequency spatial reference signal 48 and the first frequency spatial directional signal 20), is mapped in the time domain to a time filter 44 via mapping 42. This time filter 44 (which is also preferably designed here as an FIR filter) is then applied to the reference signal 46 in the time domain, i.e., to the time-directional signal 38, thereby generating the output signal 30. Finally, the output signal 30 is converted into an output sound signal 34 by the speaker 32.

[0055] By dividing the spectrum by 45, according to Figure 4 In the embodiment, the frequency-selective beamforming (frequency domain) of the directional microphone module 18 differs from the broadband beamforming of the time-domain directional microphone module 36 to a certain degree. The resulting first amplification factor g1j effectively represents a value in the instantaneous gain (this value is compensated for band-by-band by the transformed time-directional signal 40) to obtain intrinsic, direction-sensitive sound characteristics (which are inherent to the first frequency-domain directional signal 20). Therefore, the direction-sensitive sound characteristics characterized by the first amplification factor g1j, together with the noise suppression 22 and the noise suppression and amplification effects of the amplification module 24 characterized by the second amplification factor g2j, are mapped in the time domain to the time filter 44. Thus, the sound characteristics and effects can be achieved by applying the time filter 44 to the time-domain correspondence of the transformed time-directional signal 40, i.e., precisely to the time-directional signal 38.

[0056] Through the application of the time filter 44 as described above, in this embodiment, the output signal 30, relative to the two input signals 8, 10 and therefore relative to the ambient sound 7, is delayed by a time equal to the time according to... Figure 1 The implementation can also keep it very low.

[0057] Although the present invention has been described and illustrated in detail with reference to preferred embodiments, the present invention is not limited to the disclosed examples, and those skilled in the art can derive other variations therefrom without departing from the scope of protection of the present invention.

[0058] List of reference numerals

[0059] 1. Hearing instruments

[0060] 2. Microphone device

[0061] 4 First microphone

[0062] 6 Second microphone

[0063] 7. Ambient sound

[0064] 8 First Input Signal

[0065] 10 Second Input Signal

[0066] 12 First Filter Bank

[0067] 14. The first input signal after transformation

[0068] 16. The transformed second input signal

[0069] 18 directional microphone modules

[0070] 20 First Frequency Spatial Directional Signal

[0071] 22 Noise Suppression

[0072] 24 Amplification Module

[0073] 26 Second frequency spatial orientation signal

[0074] 28 Synthetic Filter Bank

[0075] 30 Output signal

[0076] 32 speakers

[0077] 34 Output sound signal

[0078] 35 Control Unit

[0079] 36 Time-Domain Directional Microphone Module

[0080] 38 Time-Directional Signal

[0081] 40 Transformed Time-Oriented Signal

[0082] 42 Mapping

[0083] 44 Time Filter

[0084] 45. Spectrum Division

[0085] 46 Reference Signal

[0086] 48. Transformed reference signal

[0087] 50 Second frequency spatial orientation signal

[0088] 52 Second Filter Bank

[0089] g1j First magnification factor

[0090] g2j Second amplification factor

[0091] gj is the amplification factor per frequency band.

Claims

1. A method for directional signal processing of a signal from a microphone device (2), the microphone device comprising at least one first microphone (4) and a second microphone (6), the first microphone (4) being used to generate a first input signal based on ambient sound (7), and the second microphone (6) being used to generate a second input signal based on ambient sound (7). in, A reference signal (46) is formed based on the first input signal, and Among them, a time filter is generated in the time domain (44). Its features are, The reference signal (46) is transformed into the frequency domain, thereby generating a frequency space reference signal (48). The first input signal and the second input signal are transformed into the frequency domain, and a first frequency spatial orientation signal (20) is formed in the frequency domain by means of the first input signal and the second input signal transformed in this way. A frequency-dependent first amplification factor (g1j) is generated by comparing the frequency resolution of the frequency space reference signal (48) with the first frequency space orientation signal (20) or the signal derived from the first frequency space orientation signal (20) in the frequency domain. The time filter (44) in the time domain is generated based on the first amplification factor (g1j), and The reference signal (46) is filtered by the time filter (44), and an output signal (30) is generated based on the reference signal (46) filtered by the time filter (44).

2. The method according to claim 1, in, The frequency resolution comparison between the frequency space reference signal (48) and the first frequency space orientation signal (20) or the signal derived from the first frequency space orientation signal (20) in the frequency domain is performed according to the spectrum division (45), and the frequency-related first amplification factor (g1j) is generated according to the spectrum division.

3. The method according to claim 1 or 2, in, A second frequency spatial orientation signal (50) is generated as a signal derived in the frequency domain from the first frequency spatial orientation signal (20) for frequency-resolved comparison with a frequency spatial reference signal (48). The second frequency spatial orientation signal is generated by applying a frequency-related second amplification factor (g2j) to the first frequency spatial orientation signal (20).

4. The method according to claim 3, in, The time filter (44) is formed based on the mapping of the frequency-related first amplification factor (g1j) in the time domain.

5. The method according to claim 1 or 2, in, Determine the frequency-dependent second amplification factor (g2j) for the first frequency spatially oriented signal (20), and The time filter (44) is formed based on the common mapping of the first amplification factor (g1j) and the second amplification factor (g2j) in the time domain.

6. The method according to claim 3, in, The frequency-dependent second amplification factor (g2j) for the first frequency spatial orientation signal (20) is determined based on the noise suppression (22) and / or dynamic compression and / or the hearing impairment to be corrected of the receiver of the output signal (30).

7. The method according to claim 1 or 2, in, The reference signal (46) is formed only by the signal components of the first input signal.

8. The method according to claim 1 or 2, The reference signal (46) is formed in the time domain as a time-directional signal (38) by means of a directional microphone based on the first input signal and the second input signal.

9. The method according to claim 1 or 2, in, The microphone device (2) for performing the method further includes a third microphone for generating a third input signal based on the ambient sound (7). The third input signal is transformed into the frequency domain, and the first frequency spatial orientation signal (20) is also formed in the frequency domain based on the transformed third input signal.

10. A method for directional signal processing in a hearing instrument (1), in, The hearing instrument (1) includes a microphone device (2) having at least one first microphone (4) and a second microphone (6) and a control unit (35). The first microphone (4) is used to generate a first input signal based on ambient sound (7), and the second microphone (6) is used to generate a second input signal based on ambient sound (7). In this method according to any one of claims 1 to 9, the output signal (30) of the hearing instrument (1) is generated based on the first input signal and the second input signal for reproduction.

11. A hearing instrument (1), comprising: - A microphone device (2), the microphone device comprising at least one first microphone (4) and a second microphone (6), the first microphone (4) being used to generate a first input signal based on ambient sound (7), and the second microphone (6) being used to generate a second input signal based on ambient sound (7). - Control unit (35) in, The control unit (35) is configured to perform the method according to claim 10 based on the first input signal and the second input signal.

Images