EP3772861A1 - Method for directional signal processing for a hearing aid - Google Patents

Abstract

The invention refers to a method for directional signal processing for a hearing aid (1), a first input signal (E1) being generated by a first input transducer (2) of the hearing aid (1) from a sound signal (6) from the environment, with a second input transducer (4) of the hearing aid (1) from the sound signal (6) of the environment a second input signal (E2) is generated, a forward signal (Z1) and a backward signal (E2) from the first input signal (E1) and the second input signal (E2). Z2), a first directional parameter (a1) being determined as a linear factor of a linear combination of the forward signal (Z1) and the backward signal (Z2) in such a way that a first directional signal (R1) resulting from this linear combination moves in a first direction (24) has a maximum attenuation, a correction parameter (e) being determined in such a way that a second directional signal (R2) as one of the first directional signal (R1) and an omnidirectional onal signal (om) with the correction parameter (e) formed linear combination in the first direction (24) has a defined relative attenuation, the second directional signal (R2) from the forward signal (Z1) and the backward signal (Z2) based on the first directional parameter and of the correction parameter (e) or from the first directional signal (R1) and the omnidirectional signal (om) is generated using the correction parameter (e), and an output signal (out) of the hearing aid (1) is generated using the second directional signal (R2) .

Description

The invention relates to a method for directional signal processing for a hearing aid, a first input signal being generated from a sound signal of the surroundings by a first input transducer of the hearing aid, a second input signal being generated from the sound signal of the surroundings by a second input transducer of the hearing aid of the first input signal and a first directional signal is generated based on the second input signal, which has a maximum attenuation in a first direction, and an output signal of the hearing aid is generated based on the first directional signal.

In a hearing aid, ambient sound is converted into at least one input signal by means of at least one input transducer, which is processed depending on a hearing impairment of the wearer to be corrected in a frequency band-specific manner and in particular individually tailored to the wearer and is also amplified. The processed signal is converted into an output sound signal via an output transducer of the hearing aid, which is sent to the wearer's hearing.

Hearing aids with two or more input transducers, in which two or more corresponding input signals for further processing are generated from the ambient sound, represent an advantageous development. This further processing of the input signals generally includes directional signal processing, ie the formation of directional signals from the input signals, the different directional effects mostly being used to produce a Useful signal source - usually a speaker in the vicinity of the hearing aid wearer - to highlight and / or to suppress background noise.

The so-called adaptive directional microphone, in which a directional signal is generated in such a way that it has a maximum attenuation in the direction of an assumed, localizable interference signal source, is of particular importance. The assumption used for this is mostly that noises occurring from the area behind the wearer of the hearing aid, that is to say in his / her rear half-space, are basically to be treated as interfering noises. Based on this assumption, conventional directional microphone algorithms usually minimize the signal energy from the rear half-space in order to generate the directional signal with the desired attenuation properties. In the direction of the maximum attenuation, the directional signal has in particular a so-called "notch", i.e. a total ("infinite") attenuation. In the ideal case, the sound of the localized noise source is thus completely masked out of the directional signal.

In some cases, however, the assumption that a sound coming from the rear half-space can only be seen as background noise is not applicable, e.g. if the person wearing the hearing aid is being addressed from the side or from behind by another person. Certain everyday noises such as a siren from an ambulance must also be perceptible to the hearing aid wearer due to their warning effect if they come from the half-space behind them.

The invention is based on the object of specifying a method for signal processing for a hearing aid, by means of which, when directional microphones are used, potentially relevant sound signals from a non-frontal direction and in particular from the rear half-space are not completely eliminated.

According to the invention, the stated object is achieved by a method for directional signal processing for a hearing aid, with a first input transducer of the hearing aid from a sound signal from the environment Input signal is generated, a second input signal is generated from the sound signal of the environment by a second input transducer of the hearing aid, a forward signal and a backward signal are generated from the first input signal and the second input signal, and a first directional parameter as a linear factor of a linear combination of the forward signal and the backward signal is determined in such a way that a first directional signal resulting from this linear combination has a maximum attenuation in a first direction. It is provided that a correction parameter is determined in such a way that a second directional signal as a linear combination formed from the first directional signal and an omnidirectional signal with the correction parameter as a linear factor has a defined attenuation in the first direction, the second directional signal consisting of the forward signal and the The backward signal is generated based on the first directional parameter and the correction parameter or from the first directional signal and the omnidirectional signal based on the correction parameter, and an output signal of the hearing aid is generated based on the second directional signal, which is preferably converted into an output sound signal by an output transducer of the hearing aid. Advantageous and in part inventive embodiments are the subject matter of the subclaims and the following description.

An input transducer in this case includes in particular an electroacoustic transducer which is set up to generate a corresponding electrical signal from a sound signal. Preprocessing, e.g. in the form of a linear pre-amplification and / or an A / D conversion, preferably also takes place when the first or second input signal is generated by the respective input converter.

The generation of the forward signal or the backward signal from the first and the second input signal preferably includes that the signal components of the first and the second input signal are included in the forward signal and in the backward signal, and thus in particular the first and the second input signal not both at the same time for generating control parameters etc., which are applied to signal components of other signals. Preferably, at least the signal components of the first input signal, and particularly preferably also the signal components of the second input signal, are included linearly in the forward signal or in the reverse signal. The same applies to a generation of the second directional signal based on the forward signal and the backward signal, and possibly for further signals and their corresponding generation.

The generation of a signal, such as the second directional signal, can also take place from the generating signals, such as the forward signal and the backward signal, in such a way that one or more intermediate signals are initially formed from the said generating signals in the course of signal processing, from which then the generated signal (e.g. the second directional signal) is formed. The signal components of the generating signals, i.e. the forward and backward signals in the present example, are then initially included in the respective intermediate signal, and the signal components of the respective intermediate signal are then included in the generated signal, i.e. in the present case in the second directional signal, so that the signal components the generating signals (e.g. the forward and backward signal) are "passed through" via the respective intermediate signal to the generated signal (e.g. the second directional signal), and if necessary, amplified in frequency bands, partially delayed against each other or weighted differently from each other, etc. .

A forward signal here includes in particular a directional signal with a non-trivial directional characteristic which, on average, has a higher sensitivity to a standardized test sound of a predetermined level in a front half space of the hearing aid than in a rear half space. The direction of maximum sensitivity of the forward signal is preferably also in the front half space, in particular in the forward direction (i.e. at 0 ° with respect to a preferred direction of the hearing aid), while a direction of minimum sensitivity of the forward signal is in the rear half space, in particular in the backward direction (i.e. at 180 ° with respect to a preferred direction of the hearing aid). Preferably, the same applies to the reverse signal, with interchangeability of the front and rear half-spaces or the forward and backward directions. The front and rear half-spaces and the forward and backward directions of the hearing aid are preferably defined by a preferred direction of the hearing aid which, when the hearing aid is worn by the wearer as intended, preferably coincides with its frontal direction. Deviations from this due to inaccurate adjustment when wearing should remain unaffected.

In particular, the forward and reverse signals are symmetrical to one another with respect to a plane of symmetry perpendicular to said preferred direction. The directional characteristic of the forward signal is given, for example, in an advantageous embodiment by a cardioid, while in this embodiment the directional characteristic of the backward signal is given by an anti-cardioid.

In order to determine the first directional parameter, it is not absolutely necessary that the first directional signal is actually generated for further signal processing of its signal components. Rather, the first directional parameter a1 can be determined, for example by minimizing the signal energy of the linear combination Z1 + a1 · Z2 (with Z1 as the forward signal and Z2 as the backward signal) or by other methods of optimization or adaptive directional microphone, without this from the linear combination resulting signal, which corresponds to the first directional signal, would experience further use in the course of the further process. In this case, the second directional signal is generated directly from the forward signal and the reverse signal. The first directional parameter is set by the aforementioned minimization of the signal energy or by other methods of optimization such that the resulting first directional signal, even if it is not used any further, has the maximum attenuation in the first direction as required, in particular if this is due to the Direction of a dominant sound source is given.

A maximum attenuation of the first directional signal is to be understood here in particular as meaning that the relevant directional characteristic has a sensitivity in the respective direction which is local, preferably global Has minimum. In other words, the first directional signal thus has a non-trivial directional characteristic and thus a sensitivity that is variable across the room to a normalized test sound predetermined level. The first directional signal preferably has a “notch” in the first direction with a total or quasi-total attenuation, that is to say by at least 15 dB, preferably by at least 20 dB. In contrast to this, however, the omnidirectional signal preferably has an angle-independent sensitivity to a standardized test sound.

Likewise, in order to determine the correction parameter it is not absolutely necessary that the second directional signal is actually formed as a linear combination, in particular a convex superimposition of the first directional signal and the omnidirectional signal with the correction parameter as a linear factor or convexity parameter. Rather, the correction parameter is selected such that a second directional signal generated as required has the required, defined relative attenuation in the first direction.

The actual generation of the second directional signal, the signal components of which are included in the output signal, takes place in particular by the described linear combination or convex superimposition of the omnidirectional signal with the first directional signal using the correction parameter, or alternatively by a linear combination of the forward signal and the reverse signal.

mit dem Korrekturparameter e als Konvexitätsparameter, om als omnidirektinoalem Signal und dem ersten Richtsignal R1. Die Abhängigkeit des zweiten Richtsignals vom ersten Richtparameter erfolgt in diesem Fall implizit über das erste Richtsignal.A convex superposition for the second directional signal R2 is to be understood here in particular as a superposition of the shape $$\mathrm{R.}\mathrm{2}=\left(\mathrm{1}-\mathrm{e}\right)\cdot \mathrm{om}+\mathrm{e}\cdot \mathrm{R.}\mathrm{1},$$

with the correction parameter e as the convexity parameter, om as the omnidirectineal signal and the first directional signal R1. The dependence of the second directional signal on the first directional parameter takes place implicitly in this case via the first directional signal.

auf, wobei a2 ein zweiter Richtparameter ist, welcher vom ersten Richtparameter a1 und vom Korrekturparameter e abhängt.The alternative generation of the second directional signal R2 from the forward signal Z1 and the backward signal Z2 on the basis of the correction parameter e and the first directional parameter has the shape in particular $$\mathrm{R.}\mathrm{2}=\mathrm{Z}\mathrm{1}+\mathrm{a}\mathrm{2}\cdot \mathrm{<figure-callout\; id="2"\; label="Z"\; filenames="imgf0001.png,imgf0002.png"\; state="\{\{state\}\}">Z</figure-callout>}\mathrm{2}\mathrm{with\; a}\mathrm{2}=\mathrm{f}\left(\mathrm{a}\mathrm{1},\mathrm{e}\right)$$

where a2 is a second directional parameter which depends on the first directional parameter a1 and the correction parameter e.

With a suitable choice of the forward signal Z1 and the backward signal Z2, for example as a cardioid and anti-cardioid signal, the omnidirectional signal om and the first directional signal R1 from equation (i) can also be represented using the forward and backward signals (for generate the omnidirectional signal om) or by means of adaptive directional microphones (for the first directional signal R1 = Z1 + a1 · Z2). In this case, there are two mutually equivalent possibilities or representations for generating the second directional signal R2, which are given by equation (i) and equation (ii).

The defined relative attenuation that the second directional signal has in the first direction (the first directional signal has precisely the maximum attenuation in this direction) is to be understood here in particular as meaning that the second directional signal has a sensitivity in the first direction that is around a factor determined in particular by the correction parameter is lower than the maximum sensitivity. The defined relative attenuation thus means, in particular, an attenuation by a factor or in dB, which can preferably be specified immediately if the correction parameter is known.

For example, if the first direction in the rear half-space is 120 ° (zero degrees in the frontal direction), and if the second directional signal is mixed in equal parts from the omnidirectional signal and the first directional signal, then the value of the relative attenuation of the second directional signal is also 120 ° - so the first direction - set towards a maximum sensitivity of the signals.

In the event that, for example, the second directional signal according to equation (i) is generated from the omnidirectional signal and the first directional signal, or for an actual generation which is carried out according to equation (ii) from the forward and reverse signals, at least one equivalent representation exists according to equation (i), the correction parameter e directly indicates the arithmetical component of the first directional signal in the second directional signal. Since its attenuation in the first direction is in the ideal case total, i.e. infinite, the sensitivity of the second directional signal in the first direction is in the ideal case completely determined by the component (1-e) of the omnidirectional signal om. For example, if you want a suppression in the first direction of only 6 dB, then as a result of the complete suppression in the first direction by the first directional signal in a second directional signal formed according to equation (i) (or a second directional signal equivalent to it) the component of the omnidirectional signal will be used choose 50%, i.e. e = 0.5. If the attenuation of the first directional signal in the first direction is finite, e.g. 15 dB or 20 dB, the calculation can be adapted accordingly if the value of the attenuation in the first direction is known.

The correction parameter is determined in particular as a function of acoustic parameters, which can be monitored using the two input signals or signals derived from the input signals, such as the forward and backward signals, and generally using a signal that characterizes the sound signal of the environment , and which have, in particular, quantifiable information about the background noise character of a non-frontal sound signal, that is to say in particular also for a sound signal from the rear half-space.

Such a meaningfulness can be given, for example, by a background noise level, by a signal-to-noise ratio (SNR), or by a stationarity of the noise to be examined, with an examination Stationarity is preferably accompanied by an examination of the half-space in which a dominant, non-frontal sound source is located.

If the first directional signal is now formed from the forward signal and the backward signal by means of adaptive directional microphones in such a way that the first direction - i.e. the direction of maximum attenuation of the first directional signal - is in the direction of a dominant, localized sound source in the rear half-space, the method can be used to mix take place with the omnidirectional signal in such a way that the resulting second directional signal is weakened by a defined factor in the first direction, and thus the sound of the sound source is no longer maximally or completely suppressed, but remains audible to the wearer of the hearing aid.

If, for example, it is determined on the basis of the backward signal that there is a considerable non-stationary signal, which also has a significant sound level and is well above the determined noise floor, i.e. also a high SNR, this can be as an indication that the dominant sound source is given by a speaker. In this case, the mixing of the omnidirectional signal with the first directional signal can be configured in such a way that a particularly high proportion of the former is included in the second directional signal in order not to suppress the signal contributions of this speaker speaking behind the wearer by the first directional signal. This applies in particular when the first directional signal is designed for dynamic or adaptive adaptation of the first direction to the direction of such a dominant sound source.

If, on the other hand, the SNR is rather low, it can nevertheless be advantageous not to allow too high a proportion of such a signal to enter the second directional signal, since this could otherwise worsen the SNR of the second directional signal in an undesirable manner. If, on the other hand, there is a significantly stationary signal with a high SNR and a comparatively high level in the rear half-space, it can be assumed that the noise is localized. The proportion of the Omnidirectional signal at the second directional signal in favor of a better suppression of the background noise, as it occurs in the first directional signal, reduced.

In the borderline case, the second directional signal can also be generated entirely without the additional addition of signal components of the first directional signal in order to prevent a strongly directional sound source in the rear half-space from being extinguished. Conversely, the second directional signal can also emerge completely from the first directional signal, that is to say without an additional addition of signal components of the omnidirectional signal, if it is decided to suppress a directional sound signal from the rear half-space as far as possible. These borderline cases are in particular formed by the end points of the value range of the correction parameter. In other words, the second directional signal can in particular be represented by a mixture of the omnidirectional signal with the first directional signal (even if the specific signal generation may possibly take place in a different but equivalent manner), the mixture also including the borderline cases that the signal components of one of the two generating signals are completely hidden.

mit a1 als erstem Richtparameter,
so kann das zweite Richtsignal R2 erzeugt werden als $$\mathrm{R}\mathrm{2}=\mathrm{Z}\mathrm{1}+\mathrm{a}\mathrm{2}\cdot \mathrm{Z}\mathrm{2}\mathrm{mit\; a}\mathrm{2}=\mathrm{f}\left(\mathrm{a}\mathrm{1},\mathrm{e}\right)$$

als zweitem Richtparameter (vgl. Gleichung ii).The second directional signal is advantageously generated by a linear combination of the forward signal and the backward signal with a second directional parameter as a linear factor, the second directional parameter being determined by a predetermined functional relationship from the first directional parameter and the correction parameter in such a way that the second directional signal in the first direction has the defined relative attenuation. If, for example, the first directional signal R1 is determined from the forward signal and the backward signal Z1 or Z2 by an adaptive directional microphone, that is to say in the form $$\mathrm{R.}\mathrm{1}=\mathrm{<figure-callout\; id="1"\; label="Z"\; filenames="imgf0002.png,imgf0005.png"\; state="\{\{state\}\}">Z</figure-callout>}\mathrm{1}+\mathrm{a}\mathrm{1}\cdot \mathrm{<figure-callout\; id="2"\; label="Z"\; filenames="imgf0001.png,imgf0002.png"\; state="\{\{state\}\}">Z</figure-callout>}\mathrm{2}$$

with a1 as the first straightening parameter,
so the second directional signal R2 can be generated as $$\mathrm{R.}\mathrm{2}=\mathrm{<figure-callout\; id="1"\; label="Z"\; filenames="imgf0002.png,imgf0005.png"\; state="\{\{state\}\}">Z</figure-callout>}\mathrm{1}+\mathrm{a}\mathrm{2}\cdot \mathrm{<figure-callout\; id="2"\; label="Z"\; filenames="imgf0001.png,imgf0002.png"\; state="\{\{state\}\}">Z</figure-callout>}\mathrm{2}\mathrm{with\; a}\mathrm{2}=\mathrm{f}\left(\mathrm{a}\mathrm{1},\mathrm{e}\right)$$

as the second guide parameter (see equation ii).

In this case, the forward signal Z1 and the backward signal Z2 are preferably generated symmetrically to a preferred plane of the hearing aid (in particular the frontal plane of the wearer), with the omnidirectional signal om being particularly preferably reproducible through these signals, e.g. as om = Z1-Z2. In particular, Z1 is given by a cardioid and Z2 by an anti-cardioid. By generating the second directional signal in this way, the generation can take place on the level of the forward and backward signals, while the first directional signal R1 is only used to determine the first directional parameter a1 (of which the second directional parameter a2 of the second directional signal is functionally called a2 = f (a1, e) with a defined function f) is required.

mit e < 1 als Korrekturparameter, wobei die Werte für den Korrekturparameter e und den Offset d beispielswese als tabellierte Werte im Hörgerät hinterlegt sein können, um in Abhängigkeit vom der ersten Richtung dort durch eine gewünschte relative Abschwächung durch eine entsprechende Parameterwahl für e und d erreichen zu können. Durch die dargestellte funktionale Abhängigkeit des zweiten Richtparameters vom ersten Richtparameter lässt sich eine relative und dabei in definiertem Ausmaß begrenzte Abschwächung in der ersten Richtung besonders einfach erreichen. Vorzugsweise ist hierbei für den Fall, dass das Vorwärts- und das Rückwärtssignal durch ein Kardioid- bzw. Anti-Kardioid-Signal gegeben sind, der Offset d zu e - 1 gewählt.The second straightening parameter is expediently derived from the first straightening parameter by scaling by the correction parameter and by a predetermined offset. This means that $$\mathrm{a}\mathrm{2}=\mathrm{f}\left(\mathrm{a}\mathrm{1},\mathrm{e}\right)=\mathrm{e}\cdot \mathrm{a}\mathrm{1}+\mathrm{d},$$

with e <1 as the correction parameter, whereby the values for the correction parameter e and the offset d can be stored in the hearing aid as tabulated values, for example, in order to achieve a desired relative attenuation there through a corresponding parameter selection for e and d, depending on the first direction to be able to. As a result of the functional dependence of the second directional parameter on the first directional parameter shown, a relative weakening in the first direction that is limited to a defined extent can be achieved particularly easily. In this case, the offset d to e − 1 is preferably selected for the case that the forward and backward signals are given by a cardioid or anti-cardioid signal.

mit dem Korrekturparameter e als Konvexitätsparameter. Dieser wird bevorzugt in Abhängigkeit eines Grundrauschpegels und/oder eines SNR und/oder einer Stationarität des Schallsignals der Umgebung ermittelt.It also proves to be advantageous if the second directional signal is generated as a convexity parameter by convex superimposition of the first directional signal and the omnidirectional signal with the correction parameter. The second Directional signal R2 then reads as a function of the omnidirectional signal om and the first directional signal R1: $$\mathrm{R.}\mathrm{2}=\left(\mathrm{1}-\mathrm{e}\right)\cdot \mathrm{om}+\mathrm{e}\cdot \mathrm{R.}\mathrm{1}\left(\mathrm{see}\mathrm{.}\mathrm{Equation\; i}\right),$$

with the correction parameter e as the convexity parameter. This is preferably determined as a function of a background noise level and / or an SNR and / or a stationarity of the sound signal of the environment.

und somit $$\mathrm{a}\mathrm{2}=\left(\mathrm{e}+\mathrm{e}\cdot \mathrm{a}\mathrm{1}-\mathrm{1}\right)$$

The forward signal and the backward signal are preferably generated symmetrically to a preferred plane of the hearing aid (in particular the frontal plane of the wearer), by which the omnidirectional signal om is particularly preferably reproducible, for example as om = Z1-Z2. In this case, the omnidirectional signal om and the first directional signal R1 can be represented in the above equation (i) by means of the forward and backward signals Z1, Z2 $$\mathrm{R.}\mathrm{2}=\mathrm{<figure-callout\; id="1"\; label="Z"\; filenames="imgf0002.png,imgf0005.png"\; state="\{\{state\}\}">Z</figure-callout>}\mathrm{1}+\left(\mathrm{e}+\mathrm{e}\cdot \mathrm{a}\mathrm{1}-\mathrm{1}\right)<figure-callout\; id="2"\; label="\cdot \; Z"\; filenames="imgf0001.png,imgf0002.png"\; state="\{\{state\}\}">\cdot </figure-callout>\mathrm{Z}\mathrm{}\mathrm{2},$$

and thus $$\mathrm{a}\mathrm{2}=\left(\mathrm{e}+\mathrm{e}\cdot \mathrm{a}\mathrm{1}-\mathrm{1}\right)$$

It can be seen from equation (vi) that the first straightening parameter a1 is scaled by the factor e <1 and shifted by an offset of e − 1. In this case, the forward signal Z1 is preferably given by a cardioid signal, and the backward signal Z2 by an anti-cardioid signal.

It also proves to be advantageous if a second direction is generated by pivoting the first direction by an angle tabulated as a function of 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 wherein the second directional parameter is determined in such a way that the second directional signal has a maximum attenuation in the second direction. This means: First, the first direction is determined in which the first directional signal, from the forward and the Backward signal preferably formed by means of adaptive directional microphone, has a maximum attenuation. The correction parameter is then determined, for example as a function of a background noise level, an SNR or a stationarity of the sound signal in the environment.

The first direction is then shifted depending on the correction parameter and possibly the first direction itself by a tabulated angle in such a way that the second directional signal, which is generated analogously to the first directional signal, in the second directional signal resulting from the shifting of the first direction by the said angle Direction has the maximum attenuation, and in the first direction the defined relative attenuation. The second directional signal is generated by means of a preferably tabulated second directional parameter which, with the linear combination of the forward and backward signals, results in precisely the required attenuation properties for the second directional signal.

The first directional parameter is advantageously generated by means of adaptive directional microphones of the linear combination of the forward signal and the reverse signal, in particular by minimizing the signal energy. In this way it can be ensured in a particularly simple manner that the first direction lies in the direction of a dominant sound source. A first directional signal generated in this way is used in many methods for directional noise suppression in hearing aids, so that the method described here is particularly suitable for preventing excessive or even complete extinction of non-stationary sound sources, especially in the rear hemisphere of the wearer of the hearing aid.

The correction parameter is advantageously determined as a function of at least one of the following variables characterizing the sound signal: a background noise level and / or an SNR and / or a stationarity parameter and / or direction information. The correction parameter is preferably determined in such a way that, for a comparatively high background noise level or a comparatively low SNR, the second directional signal is from a comparatively low one Correction of the first directional signal results, and for a comparatively low background noise level or a comparatively high SNR, the second directional signal has a comparatively low directivity. In particular, the criteria mentioned can also be applied in stages, so that, for example, for a high SNR even with a high background noise level, the second directional signal still has a considerable difference from the first directional signal. The noise floor level, the SNR and the stationarity parameter can in particular be determined using at least one of the two input signals or using the forward signal and / or the backward signal.

mit dem oberen Grenzwert Th_Hi für den Grundrauschpegel NP (in dB) sein. Eine andere funktionale Abhängigkeit als die in der zweiten Zeile von Gleichung (vii) gezeigt lineare Relation zwischen e und NP ist ebenso möglich, sofern der Anstieg dabei monoton verläuft. Insbesondere kann auch ein unterer Grenzwert Th_Lo für den Grundrauschpegel angegeben werden, unterhalb dessen e = 0 gesetzt wird, also für NP ≤ Th_Lo. In diesem Fall ist für Th_Lo < NP < Th_Hi e = (NP - Th_Lo)/(Th_Hi - Th_Lo) zu setzen.The correction parameter is advantageously formed by a monotonic function of the background noise level characterizing the sound signal, the monotonic function above an upper limit value mapping the background noise level to a first end point of the value range of the correction parameter for which the second directional signal changes into the first directional signal. The function of the background noise level NP can here for the correction parameter e ∈ [0, 1], for example, from the form $$\begin{array}{ll}\mathrm{e}& =\mathrm{1}\mathrm{f}\ddot{\mathrm{u}}\mathrm{r\; NP}\ge {\mathrm{Th}}_{\mathrm{Hi},}\\ \mathrm{e}& =\mathrm{NP}/{\mathrm{Th}}_{\mathrm{Hi}}\mathrm{f}\ddot{\mathrm{u}}\mathrm{r\; NP}<{\mathrm{Th}}_{\mathrm{Hi},}\end{array}$$

with the upper limit value Th _Hi for the noise floor level NP (in dB). A functional dependency other than the linear relationship between e and NP shown in the second line of equation (vii) is also possible, provided that the increase is monotonic. In particular, a lower limit value Th _Lo can also be specified for the background noise level, below which e = 0 is set, i.e. for NP ≤ Th _Lo . In this case, Th _Lo <NP <Th _Hi e = (NP - Th _Lo ) / (Th _Hi - Th _Lo ).

mit e_max z.B. 0.7 oder 0.5, wenn der eigentliche Wertebereich von e für SNR < Th_SNR von 0 bis 1 geht. Dies bedeutet: Für SNR < Th_SNR wird e nach der normalen funktionalen Abhängigkeit von NP bestimmt, z.B. nach Gleichung (vii). Für SNR ≥ Th_SNR wird der Wertebereich von e bei e_max begrenzt, sodass insbesondere auch das zweite Richtsignal in diesem Fall immer noch einen erheblichen Unterschied zum ersten Richtsignal aufweist, wenn das zweite Richtsignal gemäß Gleichung (i) erzeugt wird.The monotonic function of the background noise level characterizing the sound signal is preferably corrected as a function of the SNR and / or as a function of the stationarity parameter in conjunction with the directional information. For example, one possibility of such a correction is that a function defined according to equation (vii) - possibly with a different functional, monotonic dependency for the range NP <Th _Hi than the linear one specified there - at a sufficiently high SNR, e.g. for SNR ≥ Th _SNR with a correspondingly defined limit value Th _SNR for the SNR, in its range of values for e is reduced, for example
(viii) for $$\mathrm{SNR}\ge {\mathrm{Th}}_{\mathrm{SNR}}:\mathrm{e}\le {\mathrm{e}}_{\mathrm{Max}}$$

with e _max e.g. 0.7 or 0.5, if the actual value range of e for SNR <Th _SNR goes from 0 to 1. This means: For SNR <Th _SNR , e is determined according to the normal functional dependence on NP, eg according to equation (vii). For SNR ≥ Th _SNR , the value range of e is limited at e _max , so that in particular the second directional signal in this case also still has a considerable difference from the first directional signal when the second directional signal is generated according to equation (i).

A stationarity parameter is used in particular in the context of a suppression of stationary interfering noises and can thus be taken from such a parameter and can alternatively also be determined via an autocorrelation function. Such a parameter usually has a value range between zero (completely non-stationary) and one (completely stationary). If such a stationarity parameter S1 is below a corresponding limit value, i.e. S1 ≤ Th _S , and it can be identified on the basis of the directional information that the corresponding noise primarily comes from the rear half-space, then a correction of the monotonous function that changes the background noise level to the Correction parameters maps, in a middle range for the latter, that is, for example for 0.4 e 0.6, preferably also for 0.25 e 0.27, the slope of the monotonic function can be chosen to be flatter. In particular, such a correction can be combined with a correction according to equation (viii), as continuously as possible in e.

It has also proven to be advantageous if the second end point of the value range of the correction parameter is in a defined vicinity of a second end point Directional signal is superimposed with a third directional signal, which is designed to simulate a natural directional effect of a human ear, and the superimposition merges into the third directional signal when the correction parameter takes the second end point of its value range. This means that, for example, for e ≤ M, with M = 0.1 (another value, e.g. 0.05, is possible), an output signal out is formed as follows: $$\mathrm{out}=\left(\mathrm{e}/\mathrm{M.}\right)\cdot \mathrm{R.}\mathrm{2}+\left[\left(\mathrm{M.}-\mathrm{e}\right)/\mathrm{M.}\right]\cdot \mathrm{R.}\mathrm{3.}$$

At a second end of the value range of the correction parameter, which preferably corresponds to the range for which the second directional signal has the lowest possible proportion of the first directional signal or has the lowest possible directivity, the second directional signal is thus increasingly superimposed with the third directional signal, and preferably completely changes into the third directional signal at the second end point for the correction parameter. As a result, the wearer of the hearing aid has the natural spatial hearing impression that a pinna produces in a person with normal hearing. This can be done in particular because in this area it is assumed for the correction parameter that the noise floor level is sufficiently low and / or the SNR is sufficiently high.

The forward signal is preferably generated on the basis of a time-delayed superposition of the first input signal with the second input signal implemented by means of a first delay parameter, and / or the backward signal is generated on the basis of a time-delayed superposition of the second input signal with the first input signal implemented by means of a second delay parameter. In particular, the first and second delay parameters can be selected to be identical to one another, and in particular the forward signal with respect to a preferred plane of the hearing aid can be generated symmetrically to the backward signal, the preferred plane being preferably assigned to the frontal plane of the wearer when the hearing aid is worn. Aligning the first directional signal to the frontal direction of the wearer facilitates signal processing, since this takes into account the wearer's natural line of sight.

It proves to be advantageous if the forward signal is generated as a forward cardioid directional signal and the backward signal is generated as a backward cardioid directional signal (anti-cardioid). A cardioid directional signal can be generated by superimposing the two input signals on one another with the acoustic transit time delay corresponding to the distance between the input transducers. As a result - depending on the sign of this transit time delay at the superposition - the direction of the maximum attenuation lies in the frontal direction (backward cardioid directional signal) or in the opposite direction (forward cardioid directional signal).

The direction of maximum sensitivity is opposite to the direction of maximum attenuation. This facilitates further signal processing, since such an intermediate signal is particularly suitable for adaptive directional microphones due to the maximum attenuation in or against the frontal direction. In addition, the omnidirectional signal can be represented or reproduced through a difference between the forward cardioid directional signal and the backward cardioid directional signal, so that the method can run on the level of the cardoid and anti-cardioid signals, and the first directional signal only for the determination of the corresponding adaptive directional parameter is generated.

The first directional signal is expediently generated by means of adaptive directional microphones. In this way it can be achieved in a particularly simple manner that the first direction, in which the first directional signal has the maximum attenuation, coincides with a direction of a dominant sound source located in the rear half-space.

In an advantageous embodiment, when the first directional signal is generated, a first directional parameter is determined which characterizes a superposition of the first intermediate signal with the second intermediate signal for generating the first directional signal, the second directional signal being generated by superimposing the first intermediate signal with the second intermediate signal, Which is characterized by a second directional parameter, and wherein the second directional parameter is determined on the basis of the first directional parameter in such a way that the second directional signal has a relative attenuation defined with respect to the maximum sensitivity in the first direction.

The invention also calls a hearing system with a hearing aid, which has a first input transducer for generating a first input signal from a sound signal of the environment and a second input transducer for generating a second input signal from the sound signal of the environment, and a control unit which is configured to do the Carry out method according to one of the preceding claims. In particular, the control unit can be integrated in the hearing aid. In this case, the hearing system is given directly by the hearing aid. The hearing system shares the advantages of the method according to the invention. The advantages mentioned for the method and for its further developments can be applied analogously to the hearing system.

Fig. 1

in einem Blockschaltbild ein Hörgerät nach Stand der Technik, in welchem mittels adaptiver Richtmikrofonie ein Richtsignal mit einer maximalen Abschwächung in einer ersten Richtung erzeugt wird,

Fig. 2

in einem Blockschaltbild eine erfindungsgemäße Abwandlung des Hörgerätes nach Fig. 1, wobei in der ersten Richtung die Abschwächung in definierter Weise verringert wird,

Fig. 3

in einem Funktionsdiagramm einen Korrekturparameter für die Verringerung der Abschwächung gemäß Fig. 2 in Abhängigkeit eines Grundrauschpegels,

Fig. 4

in einem Blockschaltbild eine alternative Ausgestaltung des Hörgerätes nach Fig. 2, und

Fig. 5

in einem Diagramm die Richtung maximaler Abschwächung für ein erstes Richtsignal und ein gemäß Fig. 2 oder Fig. 4 abgewandeltes Richtsignal in Abhängigkeit des Richtparameters.

An exemplary embodiment of the invention is explained in more detail below with reference to a drawing. Here each show schematically:

Fig. 1

In a block diagram, 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 microphone,

Fig. 2

in a block diagram a modification of the hearing aid according to the invention Fig. 1 , whereby in the first direction the attenuation is reduced in a defined way,

Fig. 3

in a function diagram, a correction parameter for reducing the attenuation according to FIG Fig. 2 depending on a background noise level,

Fig. 4

an alternative embodiment of the hearing aid according to in a block diagram Fig. 2 , and

Fig. 5

in a diagram the direction of maximum attenuation for a first directional signal and according to Fig. 2 or Fig. 4 modified directional signal depending on the directional parameter.

Corresponding parts and sizes are provided with the same reference symbols in each of the figures.

In Fig. 1 a method for directional signal processing in a hearing aid 1 according to the prior art is shown schematically in a block diagram. 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 from a sound signal 6 from the surroundings, and can each be given, for example, by a microphone. The first input transducer 2 is arranged further forward than the second input transducer 4 with respect to a frontal direction 7 of the hearing aid 1 (which is defined by the intended wearing during operation).

The second input signal E2 is now delayed by a first delay parameter T1, and the second input signal delayed in this way is subtracted from the first input signal E1 in order to generate a forward signal Z1. In a similar manner, 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 reverse signal Z2. The first delay parameter T1 and the second delay parameter T2 are given by the transit time T, which corresponds exactly to the spatial sound path d between the first input transducer 2 and the second input transducer 4, except for possible quantification errors during digitization. The forward signal Z1 is thus given by a forward cardioid signal 16, and the backward signal Z2 is given by a backward cardioid signal 18 (that is, an anti-cardioid).

By means of an adaptive directional microphone 20, a first directional signal R1 is now obtained from the forward signal Z1 and the backward signal Z2 by minimizing the signal energy of the signal Z1 + a1 · Z2 via a first directional parameter a1. The first directional signal R1 has a directional characteristic 22 with a maximum attenuation in a first direction 24. As a result of the selection of the first directional parameter a1 by means of the adaptive directional microphone 20, the first direction 24 falls in the direction of a dominant, localized sound source 25 in the rear half-space 26. In the in FIG Fig. 1 In the example shown, the first direction is rotated by approx. 120 ° against the frontal direction 7, which coincides with a frontal direction of the wearer of the hearing aid 1 (not shown) when the hearing aid 1 is worn as intended. A maximum attenuation here means that the sound coming from the first direction 24 is ideally completely extinguished (ie “infinitely” attenuated). In other words, the first directional signal 1 has a so-called “notch” in the first direction 24.

An output signal out is now generated from the signal contributions of the first directional signal R1, possibly also by further, non-directional signal processing 29, which is converted into an output sound signal 34 by an output transducer 32 of the hearing aid 1. In the present case, the output transducer 32 can be provided by a loudspeaker or also by a bone conduction receiver.

If the dominant sound source 25 in the rear half-space 26 is given by a speaker, for example, the maximum attenuation of his speech contributions that occurs in the present case 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 which does not have a maximum attenuation in the first direction 24.

A corresponding method that can achieve this goal is based on Fig. 2 shown. A hearing aid 1 is shown in a block diagram, which follows the hearing aid up to the generation of the first directional signal R1 Fig. 1 equal is. Using the forward signal Z1 and the backward signal Z2, the example according to FIG Fig. 2 an omnidirectional signal om is formed, which is superimposed on the first directional signal R1 in accordance with a rule to be described. This superimposition takes place in accordance with a correction parameter e, which can be determined as a function of the background noise level NP and the SNR of the sound signal 6, but can also be determined for the sound signal 6 using a stationarity parameter S1 and directional information IR. The said variables can be determined either from the input signals E1 and E2 or from the forward and backward signals Z1, Z2.

A second directional signal R2 results from said superposition $$\mathrm{R.}\mathrm{2}=\left(\mathrm{1}-\mathrm{e}\right)\cdot \mathrm{om}+\mathrm{e}\cdot \mathrm{R.}\mathrm{1}\left(\mathrm{see}\mathrm{.}\mathrm{Equation\; i}\right)$$

On the basis of the second directional signal R2, if necessary by further, non-directional signal processing 29, which can include, among other things, a frequency band-dependent amplification and / or compression, analogous to FIG Fig. 1 The illustrated procedure generates the output signal out, which is converted by the output transducer 32 into the output sound signal 34. The directional characteristic 38 of the second directional signal R2 now exhibits its maximum attenuation in a second direction 40, while a relative attenuation 42 is present in the first direction 24.

beschrieben werden kann. Eine andere Kennlinie als die hier dargestellte lineare Relation ist ebenso denkbar, solange der monotone Anstieg für f(NP) zwischen Th_Lo und Th_Hi gewahrt bleibt.In Fig. 3 a function f is shown, which the background noise level NP to the correction parameter e des based on Fig. 2 illustrated procedure (solid line). Above an upper limit value Th _Hi , which in the example according to Fig. 3 when Th _Hi = 80 dB, any noise floor is mapped to e = 1. This means: Im in Fig. 2 For a background noise level NP of 80 dB and more, the first directional signal R1 is always completely converted into the second directional signal R2. Below a lower limit value Th _Lo which in the example according to Fig. 3 when Th _Lo = 40 dB is chosen, any noise floor is mapped to e = 0. This means: Im in Fig. 2 For a basic noise level NP of 40 dB and less, the illustrated method is always completely converted into the second directional signal R2. In the range Th _Lo <NP <Th _Hi , the function f has a linear slope, which by $$\mathrm{e}=\mathrm{f}\left(\mathrm{NP}\right)=\left(\mathrm{NP}-{\mathrm{Th}}_{\mathrm{Lo}}\right)/\left({\mathrm{Th}}_{\mathrm{Hi}}-{\mathrm{Th}}_{\mathrm{Lo}}\right)$$

can be described. Another characteristic than the linear relationship shown here is also conceivable, as long as the monotonic increase for f (NP) between Th _Lo and Th _{Hi is} maintained.

If the SNR is now above a predetermined limit value Th _SNR , that is to say SNR Th Th _SNR , the characteristic curve given by the function f (NP) is cut off, resulting in a new function f '(dashed line). In this case, this means: If the SNR is above Th _SNR , the behavior to the original function f is identical for comparatively low values of the background noise level NP. Above approx. NP = 65 dB, however, e is always mapped to the value e = 0.675. This takes into account the fact that with a high SNR, even with a high background noise level NP, the directional noise suppression does not have to be fully implemented, and a higher proportion of omnidirectional signal om can remain mixed for reasons of better spatial hearing perception.

If it is also determined that the sound signal 6 on the one hand is sufficiently non-stationary - e.g. on the basis that the statinar parameter S1 falls below an upper limit Th _S - and, moreover, a considerable proportion comes from the rear half-space (which is recognized on the basis of the direction information IR, which e.g. indicates the half-space of the first direction 24, which results from the adaptive directional microphone 20), the slope of the function f is reduced in a range above 55 dB for the background noise level NP (dotted line), whereby e = 1 only for a background noise level NP is reached above the limit value Th _Hi (assuming SNR <Th _SNR , since otherwise the function f 'is used immediately).

auf einen zweiten Richtparameter a2 abgebildet, welcher durch eine Skalierung des ersten Richtparameters a1 um den Faktor e (den Konvexitätsparameter gemäß Fig. 2) sowie eine Verschiebung um den Offset e - 1 gebildet wird. Das zweite Richtsignal R2 wird dann, analog zum ersten Richtsignal R1, aus dem Vorwärtssignal Z1 und dem Rückwärtssignal Z2 gebildet zu $$\mathrm{R}\mathrm{2}=\mathrm{Z}\mathrm{1}+\mathrm{a}\mathrm{2}\cdot \mathrm{Z}\mathrm{2}\left(\mathrm{vgl}\mathrm{.}\mathrm{Gleichungen\; v\; und\; vi}\right)$$

One based on Fig. 2 The procedure is analogous to the procedure described in Fig. 4 shown. This shows in a block diagram a hearing aid 1, which is similar to that shown in FIG Fig. 2 shown hearing aid 1 is modeled. Here, however, 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 results when the first directional signal R1 is generated by the adaptive directional microphone 20, is in accordance with the rule $$\mathrm{a}\mathrm{2}=\mathrm{e}+\mathrm{e}\cdot \mathrm{a}\mathrm{1}-\mathrm{1}\left(\mathrm{see}\mathrm{.}\mathrm{Equation\; vi}\right)$$

mapped onto a second directional parameter a2, which by scaling the first directional parameter a1 by the factor e (the convexity parameter according to FIG Fig. 2 ) and a shift by the offset e - 1 is formed. The second directional signal R2 is then, analogously to the first directional signal R1, formed from the forward signal Z1 and the backward signal Z2 $$\mathrm{R.}\mathrm{2}=\mathrm{<figure-callout\; id="1"\; label="Z"\; filenames="imgf0002.png,imgf0005.png"\; state="\{\{state\}\}">Z</figure-callout>}\mathrm{1}+\mathrm{a}\mathrm{2}\cdot \mathrm{<figure-callout\; id="2"\; label="Z"\; filenames="imgf0001.png,imgf0002.png"\; state="\{\{state\}\}">Z</figure-callout>}\mathrm{2}\left(\mathrm{see}\mathrm{.}\mathrm{Equations\; v\; and\; vi}\right)$$

The directional characteristic 38 is, since the in Fig. 4 procedure for in Fig. 2 The procedure shown is analogous under the same conditions, with the exception of an expansion described below for e 0.1, corresponding to the directional characteristic of the second directional signal R2 after Fig. 2 . The maximum weakening now takes place in a second direction 40, while a defined relative weakening 42 is present in the first direction 24.

In the event that the calculation of the correction parameter e according to Fig. 3 a value close to zero, i.e. e smaller than a predetermined limit value e _Lo = M with e.g. M = 0.1, the output signal out is generated by adding a third directional signal R3 to the second directional signal R2, e.g. according to the following formula: $$\mathrm{out}=\left(\mathrm{e}/\mathrm{M.}\right)\cdot \mathrm{R.}\mathrm{2}+\left[\left(\mathrm{M.}-\mathrm{e}\right)/\mathrm{M.}\right]\cdot \mathrm{R.}\mathrm{3}\left(\mathrm{see}\mathrm{.}\mathrm{Equation\; xi}\right)\mathrm{.}$$

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 that do not have the above linear relationship in e are also conceivable.

In Fig. 5 is a diagram schematically showing 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 Fig. 4 shown. The functional relationship is a2 = 0.7 · a1 - 0.3. The symbols below are used in the in Fig. 5 shown example formed by the respective first direction 24 to the parameter value of the first directional parameter a1, while the symbols above are given by the second direction to the given parameter value for a1, i.e. by the angle at which the second direction 40, i.e. sets the direction of maximum attenuation after applying the mapping of the first directional parameter R1 to the second directional parameter a2. For a given value of a1, it can be determined that said angle increases, with clipping occurring due to the axial symmetry of the directional characteristics with respect to the frontal direction in the angular direction of 180 °, which is opposite to the frontal direction. As a result of the shown pivoting of the direction of maximum attenuation during the transition from the first to the second directional signal, in the first direction in which the maximum attenuation still took place in the first directional signal, a relative attenuation is now set compared to the maximum sensitivity, which is defined by the correction parameter e is controlled.

Although the invention has been illustrated and described in more detail by the preferred exemplary embodiment, the invention is not restricted by the disclosed examples and other variations can be derived therefrom by the person skilled in the art without departing from the scope of protection of the invention.

List of reference symbols

1

Hearing aid

2

first input transducer

4th

second input transducer

6

Sound signal of the environment

7th

Frontal direction

16

forward cardioid (signal)

18th

backward facing cardioid (signal)

20th

adaptive directional microphone

22nd

Directional characteristic

24

first direction

25th

dominant sound source

26th

rear half-space

29

non-directional signal processing

32

Output converter

34

Output sound signal

38

Directional characteristic

40

second direction

42

relative weakening

a1

first straightening parameter

a2

second straightening parameter

e

Correction parameters

E1

first input signal

E2

second input signal

IR

Direction information

om

omnidirectional signal

out

Output signal

NP

Noise floor

R1

first directional signal

R2

second directional signal

R3

third directional signal

S1

Stationarity parameters

SNR

Signal-to-noise ratio

Th _Lo

lower limit value (for the background noise level NP)

Th _Hi

upper limit value (for the background noise level NP)

Th _S

upper limit (for the SNR)

Z1

Forward signal

Z2

Reverse signal

Claims (13)

Method for directional signal processing for a hearing aid (1),
a first input signal (E1) being generated by a first input transducer (2) of the hearing aid (1) from a sound signal (6) from the surroundings,
with a second input transducer (4) of the hearing aid (1) generating a second input signal (E2) from the sound signal (6) of the surroundings,
wherein a forward signal (Z1) and a reverse signal (Z2) are generated from the first input signal (E1) and the second input signal (E2),
wherein a first directional parameter (a1) is determined as a linear factor of a linear combination of the forward signal (Z1) and the backward signal (Z2) such that a first directional signal (R1) resulting from this linear combination has a maximum attenuation in a first direction (24),
wherein a correction parameter (e) is determined in such a way that a second directional signal (R2) defines a linear combination in the first direction (24) formed from the first directional signal (R1) and an omnidirectional signal (om) with the correction parameter (e) exhibits relative weakening,
wherein the second directional signal (R2) from the forward signal (Z1) and the backward signal (Z2) based on the first directional parameter and the correction parameter (e) or from the first directional signal (R1) and the omnidirectional signal (om) based on the correction parameter (e) is generated, and
an output signal (out) of the hearing aid (1) being generated on the basis of the second directional signal (R2). Method according to claim 1,
wherein the second directional signal (R2) is generated by a linear combination of the forward signal (Z1) and the backward signal (Z2) with a second directional parameter (a2) as a linear factor, and
wherein the second directional parameter (a2) is determined by a predetermined functional relationship from the first directional parameter (a1) and the correction parameter (e) such that the second directional signal (R2) in the first direction (24) the defined relative attenuation (42) having. Method according to claim 2,
wherein the second straightening parameter (a2) emerges from the first straightening parameter (a1) by scaling by the correction parameter (e) and by a predetermined offset. Method claim 1,
wherein the second directional signal (R2) is generated by a convex superimposition of the first directional signal (R1) and the omnidirectional signal (om) with the correction parameter (e) as convexity parameter. Method according to claim 1,
whereby a second direction (40) is generated by pivoting the first direction (24) by an angle tabulated as a function of the correction parameter (e),
wherein the second directional signal (R2) is generated by a linear combination of the forward signal (Z1) and the backward signal (Z2) with a second directional parameter (a2) as a linear factor, and
wherein the second directional parameter (a2) is determined in such a way that the second directional signal (R2) has a maximum attenuation in the second direction (40). Method according to one of the preceding claims,
wherein the first directional parameter (a1) is generated by means of adaptive directional microphones of the linear combination of the forward signal (Z1) and the backward signal (Z2), in particular by minimizing the signal energy. Method according to one of the preceding claims,
wherein the correction parameter (e) is determined as a function of at least one of the following variables characterizing the sound signal (6): A background noise level (NP) and / or signal-to-noise ratio (SNR) and / or a starionarity parameter (S1) and / or direction information (IR). Method according to claim 7,
wherein the correction parameter (e) is formed by a monotonic function of the background noise level (NP) characterizing the sound signal (6), the monotonic function above an upper limit value (Th _Hi ) reducing the background noise level (36) to a first end point of the value range of the correction parameter ( e) maps for which the second directional signal (R2) merges into the first directional signal (R1). Method according to claim 8,
the monotonic function of the noise floor level (36) characterizing the sound signal (6) being corrected as a function of the signal-to-noise ratio (SNR) and / or as a function of the stationarity parameter (S1) in conjunction with the directional information (IR). Method according to one of the preceding claims,
wherein in a defined area of a second end point of the value range of the correction parameter (e), the second directional signal (R2) is superimposed with a third directional signal (R3) which is designed to simulate a natural directivity of a human ear, and the superimposition in the third directional signal (R3) passes over when the correction parameter (e) assumes the second end point of its value range. Method according to one of the preceding claims,
wherein the forward signal (Z1) is generated on the basis of a time-delayed superposition of the first input signal (E1) with the second input signal (E2) implemented by means of a first delay parameter (T1), and / or
wherein the backward signal (Z2) is generated on the basis of a time-delayed superposition of the second input signal (E2) with the first input signal (E1) implemented by means of a second delay parameter (T2). Method according to claim 11,
wherein the forward signal (Z1) is generated as a forward cardioid directional signal (16), and
wherein the backward signal (Z2) is generated as a backward cardioid directional signal (18). Hearing system with - A hearing aid (1) which has a first input transducer (2) for generating a first input signal (E1) from a sound signal (6) from the environment and a second input transducer (4) for generating a second input signal (E2) from the sound signal (6 ) of the surroundings, and - A control unit which is set up to carry out the method according to one of the preceding claims.

Images