Method for operating a binaural hearing system

Abstract

The invention describes a method for operating a binaural hearing system (10) with a first hearing instrument (11) and a second hearing instrument (12), wherein a first input signal (E1) is generated from an ambient sound (18) by an electroacoustic first input transducer (M1) of the first hearing instrument (11), and a second input signal (E2) is generated by an electroacoustic second input transducer (M2) of the second hearing instrument (12), wherein a first instantaneous gain parameter (G1) is determined on the basis of the first input signal (E1), and a second instantaneous gain parameter (G2) is determined on the basis of the second input signal (E2), wherein a first parameter (P1) of an automatic gain control for the first input signal (E1) and / or a second parameter (P2) of an automatic gain control for the second input signal (E2) is adjusted such that, as a result of said adjustment (41,42) a difference between the first and the second instantaneous gain parameter (G1, G2) is reduced, and wherein in the first and second hearing instrument (11, 12) signal processing of the first and second input signal (E1, E2) is carried out with the first and second parameter (P1, P2) of the automatic gain control thus adapted.

Description

[0001] The invention relates to a method for operating a binaural hearing system with a first hearing instrument and a second hearing instrument, wherein a first input signal is generated from an ambient sound by an electroacoustic first input transducer of the first hearing instrument, and a second input signal is generated by an electroacoustic second input transducer of the second hearing instrument, wherein a first instantaneous gain parameter is determined on the basis of the first input signal, and a second instantaneous gain parameter is determined on the basis of the second input signal.

[0002] In a hearing aid, ambient sound is converted into an input signal by means of at least one electroacoustic input transducer (such as a microphone). This signal is then processed and amplified in frequency bands. This can occur, for example, in a hearing aid "in the narrower sense," to correct a wearer's hearing loss by individually tailoring the frequency band processing of the input signal to the wearer's audiological requirements. However, other hearing aids can also offer these functions to support the wearer in everyday life.

[0003] The processed input signal is converted into an output signal via an output converter in the hearing aid, which is then transmitted to the wearer's ear. During signal processing, automatic gain control (AGC) and dynamic compression are often applied to the input signal or to a pre-processed intermediate signal. With AGC, the input signal is typically amplified linearly only up to a certain threshold, and above this threshold, a lower amplification is applied to compensate for peak levels in the input signal. This is intended to prevent sudden, loud noises from resulting in an excessively loud output signal for the wearer due to the additional amplification in the hearing aid.

[0004] In a binaural hearing system with two separate hearing aids (worn on the left and right ear, respectively), compression in each instrument can cause a sound signal from a slightly lateral source to be amplified to varying degrees. In particular, background noise in the respective hemisphere (right or left) can lead to the local input signal of each hearing aid being compressed more strongly by the AGC (Automated Gain Control) and thus amplified less. This can result in a loss of the natural interaural level differences necessary for the accurate localization of sound sources. Consequently, the wearer of the binaural hearing system may perceive a sound source acoustically at a different location than where they see it.

[0005] The WO 2016 / 180 462 A1 mentions a binarual hearing aid system in which, based on an analysis of the input signals and in particular the interaural time or phase differences, the interaural level differences can be amplified by means of the two local AGCs.

[0006] US 2018 / 0 091 907 A1 refers to a change in the update rate of respective local AGCs in a binaural hearing system depending on the sound levels calculated by both AGCs on each side.

[0007] The WO 2018 / 038 821 A1 describes the generation of signal processing parameters for a binaural hearing system based on a comparison of the two signal amplitudes of the left and right local hearing aids of the binaural hearing system. In particular, local amplification parameters are coordinated and, if necessary, adjusted to each other.

[0008] US 3,509,289 A shows a binaural hearing system with an AGC, which is controlled by coupling the two local hearing aids of the binaural hearing system in order to prevent a drift apart of amplification factors for lateral sound signals, which would lead to interaural level differences via shadowing effects and thus to a different response behavior of the AGC.

[0009] In Dwyer Robert T. et al: "Synchronized Automatic Gain Control in Bilateral Cochlear Implant Recipients Yields Significant Benefit in Static and Dynamic Listening Conditions", Trends in hearing, Vol. 25, May 24, 2021, XP093034624, ISSN: 2331-2165, DOI: 10.1177 / 23312165211014139, a synchronization of the respective local AGCs for a binaural hearing system is published.

[0010] The invention is based on the objective of providing a method for signal processing in a binaural hearing system which, in particular in conjunction with AGC and dynamic compression, enables correct acoustic localization of sound sources.

[0011] The aforementioned problem is solved according to the invention by a method for operating a binaural hearing system with a first hearing instrument and a second hearing instrument, wherein a first input signal is generated from ambient sound by an electroacoustic first input transducer of the first hearing instrument, and a second input signal is generated by an electroacoustic second input transducer of the second hearing instrument, and wherein a first instantaneous gain parameter is determined on the basis of the first input signal, and a second instantaneous gain parameter is determined on the basis of the second input signal, wherein the first and the second instantaneous gain parameters are determined in such a way that level peaks of the ambient sound are attenuated in the first input signal and the second input signal, respectively.

[0012] The method involves adjusting a first parameter of an AGC for the first input signal and / or a second parameter of an AGC for the second input signal such that, as a result of said adjustment, the difference between the first and second instantaneous gain parameters is reduced, and signal processing of the first or second input signal in the first or second hearing instrument is carried out using the adjusted first or second parameter of the AGC. Advantageous and, in some cases, inventive embodiments are the subject of the dependent claims and the following description.

[0013] The first and second hearing instruments are to be worn by the wearer on the left and right ear, respectively, during the intended use of the binaural hearing system (without this constituting a mandatory assignment of the first or second hearing instrument to a specific ear).

[0014] A hearing instrument, in this context, generally encompasses any device designed to generate an electrical input signal from ambient sound and, through appropriate processing, to produce an output signal, which is then converted into an output sound signal by an output converter. This output sound signal is then delivered to the ear of the wearer of the device. Specifically, a hearing instrument includes headphones (e.g., earplugs), headsets, smart glasses with speakers, etc. However, a hearing aid in the narrower sense is also considered a hearing instrument; that is, a device for treating a wearer's hearing impairment, in which an input signal generated from an ambient sound by means of a microphone is processed into an output signal and, in particular, amplified depending on the frequency band, and an output signal is produced from the output signal by means of a loudspeaker or similar device.The generated output sound signal is suitable, in particular, to at least partially compensate for the hearing impairment of the wearer in a user-specific manner.

[0015] An electroacoustic input transducer, in this context, refers specifically to a transducer designed to generate a corresponding electrical signal from ambient sound. In particular, the generation of the first and / or second input signals by the respective input transducer may also involve preprocessing, for example, in the form of linear pre-amplification and / or analog-to-digital conversion. The resulting input signal is, in particular, an electrical signal whose current and / or voltage fluctuations essentially represent the sound pressure fluctuations in the air.

[0016] The first and second instantaneous gain parameters are to be determined such that peak levels of the ambient sound are attenuated in the respective first and second input signals, thereby preventing clipping, and preferably amplifying quiet sound events in the ambient noise. In particular, the instantaneous gain parameters can be determined using an AGC, for example, by means of a suitable compression characteristic.

[0017] The first and second parameters of the AGC to be adjusted can be directly given by the first and second instantaneous gain parameters; in this case, the difference can be directly reduced by adjustment, for example by matching the instantaneous gain parameters to each other.

[0018] However, the first or second parameter of the AGC can also be a compression ratio, a knee point of a compression characteristic, an attack time, and / or a release time of a compression. In this case, the adjustment results in a smaller difference in the instantaneous gain parameters when the input signals are recalculated based on the adjusted AGC. This means, in particular, that natural volume differences in the ambient sound on both sides of the binaural hearing system, which are represented accordingly in the two input signals, are better preserved. Specifically, only the first parameter of the AGC can be adjusted, or both the first and second parameters of the AGC can be adjusted, preferably with the two parameters moving closer together in the latter case.

[0019] In a first of two variants of the invention, the adjustment of the first or second parameter of the AGC can be carried out by transferring the input signal of the respective other hearing instrument - or a derived transmission signal (which may have a lower sample rate and / or a lower dynamic range than the input signal in question and / or contain only some frequency bands of the input signal) - to the "local" hearing instrument, and determining both instantaneous gain parameters and the corresponding adjustment of the relevant parameter of the AGC (or both parameters) in a hearing instrument based on both input signals - i.e., the "local" and the "off-axis" (of the other hearing instrument).

[0020] However, in the second variant of the invention, the adjustment of the first or second parameter of the AGC can also be carried out by determining the respective instantaneous gain parameter locally in each hearing instrument based on the local input signal, and only this parameter is transferred to the other hearing instrument, then the adjustment of the local parameter of the AGC is carried out based on both instantaneous gain parameters in at least one of the hearing instruments.

[0021] The adjustment of the first and second parameters of the AGC is carried out in the respective hearing instrument, preferably on the basis of regulations that are identically defined in advance for both hearing instruments, i.e. in particular depending on the two instantaneous gain parameters and possibly other variables, whereby the regulation specifies in advance how the respective parameter of the AGC is to be adjusted in which of the hearing instruments, and in what way (i.e. by lowering or by raising).

[0022] By adjusting either the first or second parameter of the AGC, or both parameters, natural volume differences in the ambient sound are better preserved. This can be achieved, for example, by aligning the two parameters, particularly in the case of a level-related AGC parameter such as a compression ratio or an instantaneous gain parameter (or even a knee point of the compression characteristic). This is reflected in the fact that the difference between the instantaneous gain parameters decreases when they are recalculated based on the input signals processed with the adjusted AGC parameters.This reduced difference in volume now allows for a significantly improved acoustic localization of sound sources based on output signals, which were each generated from the respective input signal in the hearing instrument using the adjusted parameters of the AGC, since the different compression, which distorts interaural level differences, can at least be partially eliminated.

[0023] According to the invention, the direction of an ambient sound source is determined at least approximately based on the first and second input signals, with the adjustment of the first and second parameters of the AGC also being based on the determined direction of the sound source. For example, the direction of the sound source can be determined to within a few degrees (e.g., + / - 5° or + / - 10°), and the adjustment of the respective AGC parameter, provided or deemed necessary based on the two instantaneous gain parameters, will be stronger the closer the sound source is to the wearer's frontal direction (the frontal direction, in the intended use of the binaural hearing system, lies in the plane of symmetry between the two hearing instruments). The further the sound source is located laterally, the easier it is to locate it based on the interaural level differences, so that the AGC adjustment on both sides may be more precise.can be softened in favor of a gentler sound.

[0024] According to the invention, determining the direction of the sound source is combined with an analysis of a useful sound signal. This means that, particularly in a situation with multiple sound sources (some of which may be directional), an analysis can be performed to determine which sound signal from each source should be considered a useful sound signal (e.g., a speech signal), so that the direction of the sound source can be approximately determined for this useful sound signal. The detection of the useful sound signal can be achieved, in particular, by analyzing the modulation and / or spectral contributions of the first and / or second input signal.

[0025] Advantageously, for the approximate determination of the direction of the sound source, a focus half-space containing the sound source and a background half-space facing away from the focus half-space are determined, wherein the focus half-space and the background half-space are defined with respect to the said plane of symmetry of the binaural hearing system (in intended operation), and wherein, for the adjustment of the first or second parameter of the AGC, in particular the corresponding first or second instantaneous gain parameter of the focus half-space and / or a signal level in the focus half-space are also used.

[0026] In other words, the direction of the sound source is determined only with respect to the lateral hemisphere in which the sound source is located. The two hemispheres are defined with respect to the plane of symmetry of the hearing instruments (as they are to be worn during normal operation), with the hemisphere containing the sound source being referred to as the focus hemisphere and the remaining hemisphere as the background hemisphere. The adjustment of the first and second parameters of the AGC, and thus the tuning of the dynamic signal processing in both hearing instruments, then occurs depending on the signal level and / or the instantaneous gain parameter in the hemisphere containing the sound source (i.e., the focus hemisphere). If the sound level in the focus hemisphere is higher than the sound level in the background hemisphere, this can determine, for example, how the gain parameters can be further increased.whether, in particular, the instantaneous gain parameter of the focus half-space (or a compression ratio there) needs to be increased for the adjustment.

[0027] In an advantageous embodiment, a first correction parameter and / or a second correction parameter is determined for adjusting the first or second parameter of the automatic gain control (AGC). The adjusted first or second parameter is formed based on a convex combination of the first parameter with the first correction parameter or of the second parameter with the second correction parameter, respectively. The first or second correction parameter preferably corresponds to a complete "one-sided" adjustment of the respective parameter; that is, for example, for the first parameter of the AGC, it is determined depending on the two instantaneous gain parameters and, optionally, the direction of a sound source and / or a signal level.), a parameter value is determined which the first parameter should preferably take in order to adjust the signal processing with respect to preserving the interaural level differences solely via the first parameter. This parameter value, the first correction parameter PC1, can then either be used directly for signal processing, or combined convexly with the original value P1 of the first parameter (which was determined according to the first instantaneous gain parameter) to obtain the parameter value Pout1 of the first parameter ultimately used in signal processing, i.e. . Pout 1 = w 1 ⋅ PC 1 + 1 − w 1 ⋅ P 1 with 0 ≤ w1 ≤ 1.

[0028] A similar approach can be taken for the second parameter, i.e. Pout 2 = w 2 ⋅ PC 1 + 1 − w 2 ⋅ P 2 with 0 ≤ w2 ≤ 1, the second correction parameter PC2, the original value of the second parameter P2, and the finally used value Pout2 of the second parameter.

[0029] The "continuous" variation of the adjustment described here can be performed independently for each hearing instrument. Advantageously, this can also be done in conjunction with the detected speech content in the input signals. If a high speech content is detected in one of the input signals, a high degree of modification of the first (and possibly the second) parameter for the adjustment can be made (w1 and possibly w2 are then chosen close to 1), since localization of the speech source is considered important.

[0030] On the other hand, a connection between the adjustments on both sides can also be established, particularly through a functional linkage of the weighting factors w1 and w2, i.e., as w2 = f(w1), specifically w2 = 1 - w1. In this case, the adjustment of the parameter on one side is made more strongly the weaker it is on the other side.

[0031] This approach is particularly advantageous in combination with the determination of the focus and background hemispheres (and corresponding signal level of the focus hemisphere), since the total extent of the adjustment can be distributed on both sides depending on the signal volume in the focus hemisphere.

[0032] Advantageously, the first input signal, or a derived first transmission signal, is transmitted from the first hearing instrument to the second hearing instrument, with the first and second instantaneous gain parameters being determined locally in the second hearing instrument. Depending on these first and second instantaneous gain parameters in the second hearing instrument, the second parameter of the automatic gain control for processing the second input signal is adjusted. This approach is particularly advantageous when not only the two instantaneous gain parameters, but also, for example, the direction of a sound source, are to be used for adjustment.

[0033] Preferably, the second input signal or a second transmission signal derived therefrom is also transmitted from the second hearing instrument to the first hearing instrument, wherein both instantaneous gain parameters are determined locally in the first hearing instrument, and wherein, depending on the two instantaneous gain parameters in the first hearing instrument, the first parameter of the automatic gain control for signal processing of the first input signal is adjusted.

[0034] Advantageously, a dedicated, hard-wired circuit is used for the local determination of the first and second instantaneous gain parameters in the second (or even the first) hearing instrument. This means, in particular, that the first instantaneous gain parameter is determined on a dedicated hardware circuit (e.g., an ASIC) within the hearing instrument, and the second instantaneous gain parameter on another hardware circuit within the same hearing instrument. Some hearing instruments already have two such dedicated circuits: one for AGC of the audio signals generated within the hearing instrument, and the other for AGC of a streaming signal that the hearing instrument receives, for example, from a multimedia device, a telephone, or similar source. In this case, for example...An AGC ASIC originally intended for the streaming signal in the second hearing instrument is used to determine the first instantaneous gain parameter.

[0035] The method is preferably applied frequency-band by frequency, meaning that AGC is performed separately for each frequency band in each hearing instrument, and the corresponding adjustment of the relevant first and second parameters of the AGC is also carried out separately for individual frequency bands. However, this adjustment can also be limited to individual frequency bands or, in particular, to a contiguous frequency range (which is especially important for interaural level differences for localization) consisting of several frequency bands. This also allows for an energy-efficient implementation of the method (especially with regard to battery power). In this case, preferably only lower frequency bands (within the aforementioned frequency range) of the first input signal are transmitted to the second hearing instrument as the first transmission signal, thereby further improving energy efficiency.

[0036] The invention further describes a binaural hearing system comprising a first hearing instrument and a second hearing instrument, wherein the binaural hearing system is configured to carry out the method described above. The binaural hearing system according to the invention shares the advantages of the method according to the invention. The advantages specified for the method and its further developments can be transferred analogously to the binaural hearing system.

[0037] Preferably, for the implementation of the method, the first and second hearing instruments each have a first and second input converter, respectively, for generating the first and second input signals of the method. Preferably, the binaural hearing system includes a signal processing unit in at least one of the hearing instruments for carrying out the signal processing steps of the method, which in particular comprises at least one signal processor. Most preferably, both hearing instruments each have such a signal processing unit.

[0038] An embodiment of the invention is explained in more detail below with reference to the drawings. The drawings schematically depict: Fig. 1a in a top view of a conversation situation, Fig. 1b in a top view the effect of dynamic compression in the binaural hearing system on the spatial hearing perception of the conversation situation according to Fig. 1a through the carrier, and Fig. 2 in a block diagram the sequence of a procedure for a binaural hearing system to improve spatial hearing perception.

[0039] Corresponding parts and sizes are marked with the same reference symbols in all figures.

[0040] In Fig. 1a The diagram schematically depicts a conversation in a top-down view, in which a first person (1) is conversing with a second person (2) who is facing the first person (1) directly in front of them. A third person (3) is positioned diagonally behind the first person (1) and may also make comments or interjections, but does not participate in the conversation between the first person (1) and the second person (2). For the purposes of the following explanations, the third person (3) could be replaced by any other directional source of background noise.

[0041] In Fig. 1b The conversation situation is schematically depicted in a top-down view. Fig. 1a The first person 1 is represented here by a wearer 5 of a state-of-the-art binaural hearing system 10, which comprises a first hearing instrument 11 and a second hearing instrument 12. In particular, the hearing instruments 11 and 12 can each be a hearing aid (in the narrower sense). The second person 2 is, according to the conversation situation, Fig. 1a Simultaneously, a conversation partner 6 of the carrier 5. The third person 3 is to be regarded as a source of disturbance 7 in the context of the conversation situation between the carrier 5 and his conversation partner 6.

[0042] In hearing aids, dynamic compression is typically applied to the input signals to reduce the dynamic range that the hearing aid microphones can theoretically resolve (i.e., from the minimum audible sound level to clipping) to a range that is acceptable and preferably comfortable for the wearer. The lower limit of this range is preferably defined by the wearer's hearing threshold, and the upper limit is preferably defined by the discomfort threshold. This is intended to ensure optimal amplification (or attenuation) for all possible or realistically expected input levels at the microphones.

[0043] In the binaural hearing system, 10 after Fig. 1b The described dynamic compression is applied independently in both hearing instruments 11 and 12; that is, for each of the two hearing instruments 11 and 12, an AGC determines an optimal gain factor for the corresponding input signal. This gain factor is usually different for both hearing instruments 11 and 12, since, for example, the second hearing instrument 12, worn by the wearer 5 on his right ear, registers a higher sound level due to the interference source 7 than the first hearing instrument 11 (which, in this case, is worn by the wearer 5 on his left ear, and thus the interference source 7 is shaded by the wearer 5's head).

[0044] In dynamic compression, a higher sound level is usually assigned a lower gain factor than a lower sound level, with the assignment being determined, for example, by a compression characteristic curve (which describes the relationship between input and output levels). This means that in a conversation situation, after... Fig. 1a or 1b, that the amplification factor, which is determined in the first hearing instrument 11 as a first instantaneous amplification parameter G1 for application to the input signal there (or, in the case of several, to the input signals there), is greater than a second instantaneous amplification parameter G2 determined in the second hearing instrument 12 for application there.

[0045] As a result of the different amplification of the input signals to the two auditory instruments 11, 12 by the instantaneous amplification parameters G1, G2, the contributions of the conversation partner 6 are also amplified differently in the two auditory instruments 11, 12, and consequently reproduced at different volumes for the user 5. This leads to the user perceiving the "left" contributions of the conversation partner 6 (which are recorded and processed by the first auditory instrument 11) as louder than the "right" contributions (which are recorded and processed by the second auditory instrument 12) due to G1 > G2.

[0046] However, this distorts the interaural level differences used by the ear to localize sound sources when the input signals to the hearing instruments 11 and 12 are reproduced. This can also distort the localization itself, meaning the wearer may perceive a sound source acoustically at a different location in the room than its actual position.

[0047] In the present example, according to Fig. 1b Therefore, if the noise source 7 emits a loud noise, the second instantaneous gain factor G2 is reduced due to dynamic compression resulting from the increased sound level (compared to the case where the noise source 7 is silent), thus the case G1 > G2 described above occurs. However, since the conversation partner 6 is positioned frontally 14 to the wearer 5, and thus their contributions to the conversation arrive at the wearer 5 at the same volume, these contributions are reproduced more loudly by the first hearing instrument 11 worn on the left by the wearer 5 than by the second hearing instrument 12 worn on the right.In the perception of the wearer 5, this "artificial" level difference resulting from the different amplification is perceived as a shadowing effect and thus as an interaural level difference, so that the conversation partner 6 is no longer "heard" (i.e., perceived) in the frontal direction 14, but in a direction 15 that is slightly rotated to the left from the frontal direction 14. Fig. 1b This is schematically illustrated by a graphical shift of conversation partner 6 (see thick arrow) towards direction 15.

[0048] The different volume levels resulting from the different instantaneous amplification parameters G1 > G2 can also have the same effect on other sound sources in the vicinity of carrier 5. This applies in particular to the interference source 7. In realistic situations, such an interference source could also pose a danger to carrier 5 (e.g., an approaching vehicle in traffic), which is why spatially distorted perception is also a safety concern.

[0049] To do this using the following methods: Fig. 1b To resolve the described problem, a procedure is provided for the binaural hearing system 10, the process of which is based on Fig. 2This is shown in a block diagram. The first hearing instrument 11 has an electroacoustic first input transducer M1, which is configured to generate a first input signal E1 from an ambient sound 18, and which in this case is provided by a microphone. The first hearing instrument 11 can also have a further input transducer (not shown), by means of which a further input signal is generated from an ambient sound 18, so that directional processing of the local input signals can take place in the hearing instrument 11.

[0050] The second hearing instrument 12 has an electroacoustic second input transducer M2, which is configured to generate a second input signal E2 from the ambient sound 18 and which is also provided by a microphone. The second hearing instrument 12 can also have a further input transducer (not shown) for local directional processing.

[0051] Based on the first input signal E1, a first transmission signal T1 is generated, which is transmitted from the first hearing instrument 11 to the second hearing instrument 12. The first transmission signal can, for example, be generated from a frequency range of contiguous frequency bands of the first input signal E1. In the aforementioned case of directional processing of two input signals in the first hearing instrument 11, the first transmission signal can also be given by the resulting directional signal (or frequency bands thereof). However, the first transmission signal T1 can also be directly given by the complete first input signal E1. Analogously, based on the second input signal E2, a second transmission signal T2 is generated, which is transmitted from the second hearing instrument 12 to the first hearing instrument 11.

[0052] In the first hearing instrument, a first instantaneous gain parameter G1 for the first input signal E1 is determined frequency-band by frequency-wise using a first local AGC 21-L. This parameter is preferably determined such that the first instantaneous gain parameter G1 achieves optimal gain for the ambient sound 18 represented in the first input signal E1, with respect to the dynamic range of the hearing instrument 11 and the hearing of the wearer 5. Likewise, in the first hearing instrument 11, a second instantaneous gain parameter G2 for the second input signal E2 is determined frequency-band by frequency-wise using a first remote AGC 21-R, according to the same rules as the first instantaneous gain parameter G1 was determined based on the first input signal E1.The second instantaneous amplification parameter G2 thus provides the optimal amplification for the second input signal E2 in the respective frequency band with regard to the dynamics and hearing ability of the carrier 5.

[0053] It should be noted that in the present example, the second input signal E2 is identical to the second transmission signal T2 in the relevant frequency bands (i.e., in those in which T2 is not zero). In the case, not shown, of two input signals per hearing instrument, which are locally pre-processed into corresponding directional signals, the aforementioned locally generated directional signals preferably replace the two input signals E1 and E2. This means, in particular, that the instantaneous gain parameters G1 and G2 are preferably generated frequency-band by frequency from the corresponding directional signals (where, in particular, the respective directional signal, possibly limited to some of its frequency bands, also serves as the transmission signal).

[0054] Similarly, in the second hearing instrument 12, the second instantaneous gain parameter G2 is determined frequency-bandwise from the second input signal E2 by a second local AGC 22-L, and the first instantaneous gain parameter G1 is determined from the first transmission signal T1 by a second remote AGC 22-R. Due to the frequency-band-identical signal components used in both hearing instruments 11 and 12 to determine the first and second instantaneous gain parameters G1 and G2, respectively, and due to the identical algorithms in the local first AGC 21-L and the remote second AGC 22-R (as well as in the remote first AGC 21-R and the local second AGC 22-L), the first instantaneous gain parameters G1 determined in both hearing instruments 11 and 12 are identical to each other (and the second instantaneous gain parameters G2 determined in each case are identical to each other).

[0055] In a first source determination Q1 of the first hearing instrument 11, the direction 25 of a sound source 30 in the ambient sound 18 is determined at least approximately, frequency-wise, based on the first input signal E1 and the second transmission signal T2. This approximate determination can, for example, ascertain a polar angle (possibly accurate to 5°, 10°, etc.) of the sound source with respect to the frontal direction 14, or it can simply determine a half-space with respect to a plane of symmetry 28 of the binaural hearing system 10 containing the frontal direction 14, in which the sound source 30 lies. This half-space is referred to here as the focus half-space 31. Analogously, the direction 25 is also determined at least approximately, frequency-wise, in a second source determination Q2 of the second hearing instrument 12, based on the second input signal E2 and the first transmission signal T1.Since the same signal components in the frequency bands are used in the first and second listening instruments 11, 12 (i.e., that the respective input signal E1 or E2 is identical to its transmission signal T1 or T2 in the frequency bands used), the same direction 25 is determined in both source determinations Q1, Q2. In the case not shown, where two input signals are present in each listening instrument 11, 12, which are each locally pre-processed into corresponding directional signals, the frequency-band-wise directional signals are preferentially fed to the first and second source determinations Q1, Q2, respectively.

[0056] Based on the direction 25, the focus half-space 31 in which the sound source 30 is located is determined in each of the two hearing instruments 11, 12 (unless this has already been done by the approximate determination of the direction 25), as well as, resulting from this, the half-space opposite the focus half-space 31, which shall be referred to here as the background half-space 32.

[0057] Based on the first and second instantaneous gain parameters G1, G2, and the knowledge of the focus half-space 31, a first adjustment 41 of a first parameter P1 of the AGC is performed in the first hearing instrument 11. This parameter is used locally in the first hearing instrument 11 for signal processing (i.e., in the present embodiment, no "external" parameter of the second hearing instrument 12 is adjusted in the first hearing instrument 11). Additionally or alternatively, a second adjustment 42 of a second parameter P2 of the AGC is performed in the second hearing instrument 12. This parameter is also used locally in the first hearing instrument 12 for signal processing.

[0058] In this case, the first parameter P1 is given by the first instantaneous gain parameter G1, and the second parameter P2 by the second instantaneous gain parameter G2. If, for example, G1 < G2 (e.g., if the sound source 30 in the focus hemisphere 31 results in a higher sound level than in the background hemisphere 32 due to shadowing effects, and there is no excessively loud noise source present there), the adjustment can, for example, consist of simply using the value of the first instantaneous gain parameter G1 for the second parameter P2 via the second adjustment 42, so that the gain of the input signals E1, E2 is the same in both hearing instruments 11, 12. Such a reduction only attenuates additional background noise in the background hemisphere 32. Conversely, if G1 > G2 (e.g.,If, due to a loud interference source in the background hemisphere 32 and a simultaneous useful signal from the sound source 30, the parameters P1 and P2 (i.e., the two instantaneous gain parameters G1 and G2) can be adaptively adjusted depending on additional speech recognition (not shown) of the input signals E1 and E2 (or the transmission signals T1 and T2). In short signal segments (such as frames or other time bins of suitable length) containing speech, the adjustment can be suspended to prevent the speech signal from becoming unintelligible by increasing G2 (amplifying the background noise) or decreasing G1 (and thus the speech signal). The adjustment (e.g., by matching the instantaneous gain parameters G1 and G2 to the values ​​of parameters P1 and P2) is then limited to cases where no speech signal is present.

[0059] Other previously described types of adjustments – especially of both parameters P1 and P2 simultaneously – can also be carried out.

[0060] The first input signal E1 is then processed in the first hearing instrument 11 with the appropriately adjusted first parameter P1 to produce a second output signal Ou1, and the second input signal E2 is processed in the second hearing instrument 12 with the appropriately adjusted second parameter P2 to produce a second output signal Ou2 (where, as mentioned, adjusting only one of the two parameters may have a non-trivial effect). The two output signals Ou1 and Ou2 can then be subjected to further signal processing steps not described in detail (e.g., additional noise suppression and / or acoustic feedback, etc.) and are subsequently converted by an electroacoustic first and second output converter L1 and L2, respectively, into a first and second output sound signal 51 and 52, respectively.

[0061] Although the invention has been illustrated and described in detail by the preferred embodiment, the invention is not limited by the disclosed examples and other variations can be derived by the person skilled in the art without leaving the scope of protection of the invention. Reference symbol list

[0062] 1 First person 2 Second person 3 Third person 5 Wearer (of the binaural hearing system) 6 Conversation partner 7 Source of interference 10 Binaural hearing system 11 First hearing instrument 12 Second hearing instrument 14 Frontal direction 15 Direction 18 Ambient sound 21-L / -Rest first local or off-axis AGC 22-L / -Rest second local or off-axis AGC 25 Direction (of the sound source) 28 Plane of symmetry 30 Sound source 31 Focus hemisphere 32 Background hemisphere 41 First fitting 42 Second fitting 51 First output sound signal 52 Second output sound signal E1, E2 First or second input signal G1, G2 First or second instantaneous gain parameter L1, L2 (electroacoustic) first or second output transducer M1, M2 (electroacoustic) first or second input converter Ou1, Ou2 first or second output signal P1, P2 first or second parameter (of an AGC) Q1, Q2 first or second source determination T1, T2 first or second transmission signal

Claims

1. Method for operating a binaural hearing system (10) having a first hearing instrument (11) and a second hearing instrument (12), wherein an electroacoustic first input transducer (M1) of the first hearing instrument (11) generates a first input signal (E1), and an electroacoustic second input transducer (M2) of the second hearing instrument (12) generates a second input signal (E2), from ambient sound (18), wherein the first input signal (E1) is used to ascertain a first instantaneous gain parameter (G1), and the second input signal (E2) is used to ascertain a second instantaneous gain parameter (G2), wherein the first and the second instantaneous gain parameter (G1, G2) are ascertained in such a way that in each case level peaks in the ambient sound (18) are attenuated in the first input signal (E1) or second input signal (E2), wherein a first parameter (P1) of an automatic gain control for the first input signal (E1) and / or a second parameter (P2) of an automatic gain control for the second input signal (E2) is / are adjusted so that said adjustment (41, 42) results in a difference between the first and the second instantaneous gain parameter (G1, G2) being decreased, and wherein a signal processing for the first or second input signal (E1, E2) using the thus adjusted first or second parameter (P1, P2) of the automatic gain control is performed in the first or second hearing instrument (11, 12), wherein the first or second parameter (P1, P2) of the automatic gain control that is adjusted is the first or second instantaneous gain parameter (G1, G2), and / or wherein the first or second parameter (P1, P2) of the automatic gain control that is adjusted is a compression ratio and / or a knee of a compression characteristic curve and / or an attack time and / or a release time of a compression, such that there is a smaller difference between the two instantaneous gain parameters (G1, G2) as a result of the adjustment when recalculating the two instantaneous gain parameters (G1, G2) on the basis of the input signals (E1, E2) processed according to the two adjusted parameters (P1, P2) of the automatic gain control, wherein, for the adjustment, - the first input signal (E1) or a first transmission signal (T1) derived therefrom is transmitted from the first hearing instrument (11) to the second hearing instrument (12), the first and the second instantaneous gain parameter (G1, G2) are ascertained locally in the second hearing instrument (12), and the first and the second instantaneous gain parameter (G1, G2) are taken as a basis for adjusting the second parameter (P2) of the automatic gain control for the signal processing of the second input signal (E2) in the second hearing instrument (12), or - the respective instantaneous gain parameter (G1, G2) is ascertained locally in each of the two hearing instruments (11, 12) on the basis of the respective local input signal (E1, E2), and this gain parameter is transmitted to the respective other hearing instrument (11, 12), and the first or second parameter (P1, P2) of the automatic gain control is adjusted in at least one of the two hearing instruments (11, 12) on the basis of both instantaneous gain parameters (G1, G2), characterized in that the first input signal (E1) and the second input signal (E2) are used to at least approximately determine a direction (25) of a sound source (30) relating to the ambient sound (18), the determination of the direction of the sound source (30) being associated with an analysis with regard to a useful sound signal, such that the direction of the sound source (30) is approximately determined for a useful sound signal, and the adjustment (41, 42) to the first or second parameter (P1, P2) of the automatic gain control is also made on the basis of the ascertained direction (25) of the sound source (30).

2. Method according to Claim 1, wherein the adjustment (41, 42) to the first or second parameter (P1, P2) of the automatic gain control is made to a greater degree the closer the ascertained direction (25) of the sound source (30) is to a frontal direction of the binaural hearing system (10).

3. Method according to Claim 1, wherein the approximate determination of the direction (25) of the sound source (30) involves ascertaining a focus half-area (31), containing the sound source (30), and a background half-area (32), which is remote from the focus half-area (31), the focus half-area (31) and the background half-area (32) being defined relative to a plane of symmetry (28) of the binaural hearing system (10), and wherein the adjustment (41, 42) to the first or second parameter (P1, P2) of the automatic gain control involves in particular also using the applicable first or second instantaneous gain parameter (G1, G2) relating to the focus half-area (31), or a signal level in the focus half-area (31).

4. Method according to one of the preceding claims, wherein the adjustment (41, 42) to the first or second parameter (P1, P2) of the automatic gain control involves ascertaining a first correction parameter and / or a second correction parameter, wherein an adjusted first or second parameter (P1, P2) is formed on the basis of a convex combination of the first parameter (P1) with the first correction parameter, or of the second parameter (P2) with the second correction parameter, respectively.

5. Method according to one of the preceding claims, wherein the local ascertainment of the first and the second instantaneous gain parameter (G1, G2) in the second hearing instrument (12) involves using a respective specifically dedicated hardwired circuit.

6. Method according to one of the preceding claims, wherein the method is applied frequency band by frequency band.

7. Method according to Claim 6, wherein the first transmission signal (T1) transmitted to the second hearing instrument (12) is only low frequency bands of the first input signal (E1).

8. Binaural hearing system (10) having a first hearing instrument (11) and a second hearing instrument (12), wherein the binaural hearing system (10) is configured to carry out the method according to one of the preceding claims.

Images