Method for determining the activity of a wearer's own voice, hearing aid and hearing aid system

The method and system in hearing aids use neural networks and fixed-coefficient spatial filters to automatically distinguish own voice from other sounds, addressing the altered perception issue and enhancing user experience without personalization.

DE102023205783A1Pending Publication Date: 2024-12-24SIVANTOS PTE LTD
View PDF 1 Cites 0 Cited by

Patent Information

Application Number
DE102023205783
Authority / Receiving Office
DE · DE
Patent Type
Applications
Current Assignee / Owner
Filing Date
2023-06-21
Publication Date
2024-12-24

AI Technical Summary

Technical Problem

Hearing aids often alter the perception of one's own voice, making it unfamiliar and undesirable, and existing self-voice detection methods are typically personalized and require adjustment by the user.

Method used

A method and system using a neural network-based signal processing in hearing aids to automatically distinguish between the wearer's own voice and other sounds, employing multi-frequency channel processing and spatial separation filters with fixed coefficients, without personalization, to enhance the perception of own voice.

Benefits of technology

Effectively differentiates between the wearer's voice and other sounds, improving the perception of own voice without user-adjustment, ensuring consistent and reliable operation.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure 00000013_0000
    Figure 00000013_0000
  • Figure 00000013_0001
    Figure 00000013_0001
  • Figure 00000014_0000
    Figure 00000014_0000
Patent Text Reader

Abstract

The invention relates to a method for determining the activity of a wearer's own voice of a hearing aid (4) using a signal processing unit (8) of the hearing aid (4), wherein a first input signal is generated by means of a first input converter (6) of the hearing aid (4) and a second input signal is generated by means of a second input converter (6) of the hearing aid (4), wherein the two input signals are supplied to a detection unit (22) of the signal processing unit (8), which has a neural network (24) and a pre-stage (26) upstream of the neural network (24), wherein information signals are generated by means of the pre-stage (26) based on the two input signals, wherein the information signals are evaluated by the neural network (24) and wherein a detection result is output by the detection unit (22) based on the evaluation of the information signals by the neural network (24).
Need to check novelty before this filing date? Find Prior Art

Description

[0001] The invention relates to a method for determining the activity of a hearing aid wearer's own voice. Furthermore, the invention relates to a hearing aid and a hearing aid system.

[0002] Hearing aids typically refer to traditional hearing aids designed to assist people with hearing loss. However, in a broader sense, this term also refers to devices designed to support people with normal hearing. Hearing aids designed to assist people with normal hearing are also known as "Personal Sound Amplification Products" or "Personal Sound Amplification Devices" (PSADs for short). Unlike traditional hearing aids, these hearing aids are not intended to compensate for hearing loss, but are specifically used to support and improve normal human hearing in specific listening situations.

[0003] Regardless of their intended purpose, hearing aids typically have at least one input transducer, a signal processing device, and an output transducer as essential components. The at least one input transducer is usually an acousto-electrical transducer, for example a microphone, or an electromagnetic receiver, such as an induction coil. In many cases, multiple input transducers are installed, for example one or more acousto-electrical transducers and an electromagnetic receiver. The output transducer is usually an electro-acoustic transducer, for example a miniature loudspeaker (also referred to as a "receiver"), or an electromechanical transducer, such as a bone conduction receiver.The signal processing device is usually implemented as an electronic circuit on a printed circuit board and usually includes an amplifier or amplifier unit. Hearing aids are also often equipped with a transmitting and / or receiving unit that enables wireless communication with other electronic devices, especially with other hearing aids.

[0004] In some applications, two such hearing aids form a hearing aid system, particularly a binaural hearing aid system. Typically, one of the hearing aids is designed for the left ear, and the other for the right ear.

[0005] It is well known, especially when using hearing aid systems, that the perception of one's own voice is altered, making it seem unfamiliar and alienated. Since such an effect is typically undesirable, the signal processing device prefers to use different signal processing methods depending on whether the underlying signals reflect one's own voice or not.

[0006] To enable such differentiated signal processing, corresponding hearing aids are configured for what is known as own-voice detection (OVD). They thus determine when the wearer's own voice is active and when it is not. Such own-voice detection is described, for example, in EP 3 222 057 B1. Especially when the own-voice detection is also individualized in some way, allowing recognition of a specific voice of a specific hearing aid wearer, this is often referred to as own-voice recognition.

[0007] Based on this, the invention is based on the object of providing an advantageous method for determining the activity of the individual voice of a hearing aid wearer. Furthermore, the invention is based on the object of providing an advantageously designed hearing aid and an advantageously designed hearing aid system.

[0008] This object is achieved according to the invention by a method having the features of claim 1, by a hearing aid having the features of claim 14, and by a hearing aid system having the features of claim 15. Preferred developments are contained in the dependent claims. The advantages and preferred embodiments cited with regard to the method are also transferable mutatis mutandis to the devices, i.e., the hearing aid on the one hand and the hearing aid system on the other, and vice versa.

[0009] The method according to the invention serves to determine the activity of a person's own voice, namely the voice of a hearing aid wearer. The method expediently comprises at least one part of the method, namely determining the activity itself, with this part of the method being carried out by the hearing aid, specifically with the aid of a signal processing device of the hearing aid.

[0010] The hearing aid according to the invention, in turn, is designed and configured to carry out at least one method step of the method according to the invention in at least one operating mode. In particular, the hearing aid according to the invention is designed to carry out the aforementioned at least one method part, namely determining the activity itself. For this purpose, the hearing aid according to the invention has the aforementioned signal processing device. The hearing aid according to the invention is thus configured to automatically or automatically recognize in the at least one operating mode when the voice of the wearer of the hearing aid, i.e., the hearing aid according to the invention, is active and when it is not.

[0011] Finally, the hearing aid system according to the invention comprises two hearing aids of the aforementioned type. Preferably, one of these hearing aids is designed for a left ear, and the other of the two hearing aids is designed for a right ear. Furthermore, the hearing aid system is preferably designed as a so-called binaural hearing aid system.

[0012] In the course of carrying out the method, i.e. the method according to the invention, and in particular of at least one part of the method, a first input signal is generated by means of a first input transducer of the hearing aid, i.e. the hearing aid according to the invention, and a second input signal is generated by means of a second input transducer of the hearing aid.

[0013] The two input signals are then fed to a detection unit of the signal processing device, which comprises a neural network and a pre-stage connected upstream of the neural network. Alternatively, the two input signals are fed to a filter bank of the signal processing device, where they are split into sub-signals. In this case, the sub-signals are then fed to the detection unit. In other words, the two input signals are fed to the detection unit either split or split.

[0014] A corresponding filter bank is typically designed such that this filter bank specifies two or more frequency channels, frequency bands, or frequency ranges, namely n frequency channels, frequency bands, or frequency ranges, and at least approximately breaks them down into sub-signals such that each sub-signal contains all frequency components of an input signal that can be assigned to one of the frequency channels / frequency bands / frequency ranges k from k=1...n. Hearing aids with such a filter bank are also referred to as multi-frequency channel hearing aids. If the hearing aid is designed as a multi-frequency channel hearing aid, it is expediently designed as an n-frequency channel hearing aid, where n is, for example, 2, 4, 8, 16, 32, or 48. The frequency spectrum is thus divided into n frequency bands, frequency ranges, or frequency channels.

[0015] Furthermore, the pre-stage generates information signals based on the two input signals—the undissociated and decomposed input signals. These information signals are then evaluated by the neural network. Based on the neural network's evaluation of the information signals, the evaluation unit outputs a result. This result is typically a simple binary signal that indicates whether the hearing aid wearer's own voice, i.e., the voice of the hearing aid user, is currently active or not.

[0016] Based on this determination result, an output signal is then preferably generated for an output transducer of the hearing aid. Different signal processing methods or parameterizations are typically used to generate the output signal depending on whether the determination result indicates that the wearer's own voice is active or not.

[0017] Furthermore, the method described here preferably dispenses with personalization of the detection unit. This means that the eigenvoice detection realized using the method is not and will not be adapted to a specific wearer, i.e., a specific person. Therefore, it is further preferred that the detection unit is not adapted after it has been issued or delivered by the manufacturer. In this case, not only the hardware of the detection unit is specified by the manufacturer, but also the software, the programming, and / or the settings of the detection unit. In particular, therefore, no subsequent changes to parameters or parameter values ​​take place, at least none in which the wearer participates in any way.

[0018] The aforementioned neural network is expediently a so-called artificial neural network, or ANN for short. Depending on the application, the artificial neural network is designed, for example, as a so-called feedforward network. Preferably, it is designed as a so-called recurrent network. Irrespective of this, the artificial neural network preferably has two or more so-called layers. It is therefore particularly designed as a multi-layer neural network (deep neural network, DNN). Alternatively, only one layer is formed. Suitable topologies include artificial neural networks designed according to the type of LSTM (long short-term memory), GRU (gated recurrent unit), or CNN (convolutional neural network), or neural networks that have a mixed topology.

[0019] To implement the method or at least the aforementioned at least one part of the method, namely determining the activity itself, the pre-stage preferably further comprises two filters, namely a first filter and a second filter, wherein the first filter corresponds to a first filter type and the second filter to a second filter type. In the at least one operating mode of the hearing aid, a first filter signal is generated by means of the first filter, and a second filter signal is generated by means of the second filter.

[0020] Both a filter of the first filter type and a filter of the second filter type are typically filters for spatial separation, in particular a so-called beamformer. They are therefore not filters for pure frequency separation, as used to implement a filter bank mentioned above.

[0021] Particularly if the hearing aid is designed as a multi-frequency channel hearing aid as described above, it is further advantageous if the preamplifier is designed as a multi-frequency channel preamplifier. This means that partial signals are either fed to the preamplifier in the manner described above, or that the preamplifier itself has a filter bank of the type described above, with which the supplied undivided input signals are decomposed into partial signals. Typically, the preamplifier is designed as an n-frequency channel preamplifier, where n is, for example, 2, 4, 6, 8, 16, 32, or 48. Thus, the frequency spectrum is divided into n frequency bands, frequency ranges, or frequency channels. For each of the input signals, n partial signals are then present in the preamplifier, at least after the filter bank. The further processing of the partial signals in the preamplifier is then typically carried out frequency channel by frequency channel. This means that, typically, in frequency channel k, k = 1...n only the partial signal of the first input signal with the frequency components from the frequency band k is further processed and the partial signal of the second input signal with the frequency components from the frequency band k.

[0022] In an advantageous development, the multi-frequency channel preamplifier has two filters for each frequency channel k from k=1...n, namely a first frequency channel k filter and a second frequency channel k filter, wherein the first frequency channel k filter corresponds to the first filter type and the second frequency channel k filter corresponds to the second filter type. In the at least one operating mode, two filter signals are generated in each frequency channel k, namely a first frequency channel k filter signal with the filter of the first type and a second frequency channel k filter signal with the filter of the second type.

[0023] As already explained above, both a filter of the first filter type and a filter of the second filter type are typically spatial separation filters. In this case, each of the two filter types is preferably designed such that a common directional characteristic is realized for the two aforementioned microphones of the hearing aid, namely a first directional characteristic by the first filter type and a second directional characteristic by the second filter type. Thus, two different directional characteristics are preferably realized with the two filter types.

[0024] The two directional characteristics differ particularly with regard to the specified spatial directions in which the sensitivity of the respective directional characteristics reaches its maximum. The corresponding two spatial directions are expediently specified relative to a frontal direction, namely, in particular, by an angle for each of the two spatial directions relative to the angle of 0° specified for the frontal direction. If the hearing aid is then worn by the wearer, the frontal direction typically points in the direction in which the face of the hearing aid wearer is facing. If the wearer then also looks straight ahead, the frontal direction also points in the wearer's line of sight.

[0025] For the first filter type, a spatial direction is then preferably specified that points into a frontal hemisphere relative to the frontal direction, in particular starting from the hearing aid wearer and their head orientation. Further preferably, a spatial direction is specified that lies in the range between 270° and 90° relative to the frontal direction, and in particular in the range between 340° and 20°, i.e., for example, at approximately 0°.

[0026] For the second filter type, a spatial direction is preferably specified that points into a posterior hemisphere relative to the frontal direction, in particular starting from the hearing aid wearer and their head orientation. Further preferably, a spatial direction is specified that lies between 95° and 265° relative to the frontal direction, and in particular between 160° and 200°, i.e., for example, approximately 180°.

[0027] Also useful are variants in which the first and second filter types are designed such that the first filter type suppresses self-voices less than the second filter type. As a result, signal components that reproduce self-voices are suppressed more strongly by the second filter type than by the first filter type during operation.

[0028] Also advantageous are variants in which the first filter type is designed to extract or emphasize useful signals from defined useful signal sources, particularly from defined useful signal sources positioned directly in front of the hearing aid wearer. The defined useful signal sources are typically conversation partners.

[0029] The second filter type, at least in some advantageous embodiments, is designed to mask out or suppress self-voices and thus also to mask out or suppress the self-voices of the hearing aid wearer, in particular the current wearer of the hearing aid. Signal components that reproduce self-voices are masked out or suppressed more strongly than all other signal components.

[0030] It is also advantageous if both filter types are designed as static filters, each with fixed filter coefficients. In this case, the filter coefficients are preferably specified by the manufacturer and cannot be changed after the manufacturer has specified them.

[0031] According to an alternative variant, only one of the two filter types is designed as a static filter with fixed filter coefficients. The other is then typically designed as an adaptive filter. A variant in which both filter types are designed as adaptive filters is also useful. In such adaptive filters, the filter coefficients are not fixed. Instead, the filter coefficients are adjusted by the hearing aid during operation.

[0032] As already explained above, the first filter type, according to at least one preferred embodiment, is designed as a spatial separation filter with a predetermined spatial direction with regard to the directional characteristic and is configured to extract or emphasize useful signals from defined useful signal sources. The following should be noted in this regard: As soon as the predetermined spatial direction deviates from the direction in which the useful signal source is located, the spatial separation filter will usually also at least partially suppress the useful signal. This is a typical result with common filter algorithms. The extent of this suppression depends primarily on the extent to which the spatial direction and the direction of the useful signal source differ from each other.

[0033] Such specification of filter coefficients is typically not part of the aforementioned at least one method part, hereinafter referred to as the second method part, but rather of a further method part, hereinafter referred to as the first method part, which is usually not performed by the hearing aid. Part of this further method part, i.e., the first method part, is preferably a method step in which suitable filter coefficients are determined. These determined suitable filter coefficients are then specified for the filters and thus for the second method part, namely, the determination of the activity itself, i.e., stored in particular in the hearing aid.

[0034] Suitable filter coefficients are preferably determined by analyzing recorded acoustic signals. The recording is typically performed using a hearing aid prototype or a hearing aid dummy, and the analysis is more preferably performed using a separate processing unit. The corresponding hearing aid prototype, hearing aid dummy, or processing unit preferably also has adaptive test filters or simulates appropriate adaptive test filters, which are then used to determine suitable filter coefficients for the hearing aid's static filters.

[0035] To determine suitable filter coefficients for the second filter type, typically recorded acoustic signals are used in which the original voices of different test wearers are recorded. This means that different people act as test wearers, in particular as test wearers of a previously mentioned hearing aid prototype or hearing aid dummy, and that acoustic signals with their original voices are evaluated in order to find suitable filter coefficients for the second filter type and then use these as fixed filter coefficients. To determine suitable filter coefficients for the first filter type, in contrast, typically recorded acoustic signals are used in which various wanted signal sources are recorded, which are positioned in particular frontally in front of a previously mentioned test wearer or an artificial head. The wanted signal sources are typically conversation partners.In both cases, the preferred signals are recorded acoustic signals without background noise.

[0036] In the aforementioned first part of the process, the neural network is preferably trained, particularly without the involvement of the future hearing aid wearer. The training is typically carried out by the hearing aid manufacturer.

[0037] Furthermore, the neural network is further preferably trained using a number of recorded acoustic signals. These typically include the recorded acoustic signals used to determine the appropriate filter coefficients. Alternatively or additionally, the number of recorded acoustic signals for training the neural network includes recorded acoustic signals with different background noises.

[0038] For training the neural network according to one embodiment, recorded acoustic signals are then used, for example, in which the individual voices of different test wearers are recorded together with different background noises, and / or recorded acoustic signals in which different wanted signal sources are recorded together with different background noises, and / or recorded acoustic signals in which the individual voices of different test wearers are recorded together with different wanted signal sources and together with different background noises.

[0039] In another variant, the own voices of different test carriers, different useful signal sources and different background noises are recorded separately and subsequently combined for training the neural network, if necessary with different weightings.

[0040] It is also advantageous if, during training, a previously mentioned hearing aid prototype or a previously mentioned hearing aid dummy is used to record the aforementioned acoustic signals. In some cases, the recorded signals are further used to initially train a neural network, which is implemented in the hearing aid prototype or hearing aid dummy or by a separate computing unit. The training of the neural network in the hearing aid is then preferably carried out by transferring the training state from the neural network in the hearing aid prototype or hearing aid dummy or the separate computing unit to the neural network in the hearing aid.

[0041] The pre-stage of the determination unit then preferably generates two or more filter signals in the manner described above, i.e., the first filter signal and the second filter signal, or the first frequency channel n filter signals and the second frequency channel n filter signals. In this case, each of the filter signals is then further preferably combined in the pre-stage to generate at least one information signal with a reference signal based on at least one of the two input signals.

[0042] In an advantageous development, a real-valued attenuation value is determined by linking one of the filter signals with one of the reference signals, wherein an information signal is expediently generated based on the attenuation value. According to one alternative, a complex-valued attenuation value is determined. Two embodiments are then provided for this alternative. In the case of the first embodiment, an information signal is generated based on the complex-valued attenuation value. In the case of the second embodiment, however, two information signals are generated based on the complex-valued attenuation value. In this case, separate information signals are generated, for example, for a real part and an imaginary part or for an amplitude and a phase.

[0043] As already explained above, the information signals generated in the preliminary stage are evaluated by the neural network, with the neural network expediently outputting an evaluation result based on the evaluation. The evaluation of the information signals, and in particular of the attenuation variables, is further preferably carried out in such a way that the neural network outputs a prediction value, namely a type of probability value, as the evaluation result, in particular a value in the range from 0 to 1.

[0044] Further preferably, the evaluation result of the neural network is fed to a final stage of the determination unit, which preferably has a comparator unit and is expediently connected downstream of the neural network. In the comparator unit, the evaluation result is then preferably compared with a reference, in particular a threshold value, thereby generating a determination result. If, for example, the aforementioned probability value is above the aforementioned threshold value, the determination result is determined to be that the wearer's own voice is active. If, on the other hand, the probability value is below the threshold value, the determination result is determined to be that the wearer's own voice is not active. This determination result is then preferably output by the determination unit.

[0045] According to another embodiment, the evaluation result is again compared with the reference in the comparator unit, but this only generates a preliminary determination result. The preliminary determination result is then preferably evaluated together with another preliminary determination result in the final stage, for example, in an additional logic unit of the final stage, which then determines the determination result, i.e., the final determination result. This determination result is then typically output by the determination unit.

[0046] The further preliminary determination result is preferably determined by a further determination unit in a further hearing aid and transmitted by this hearing aid. This variant of the method is particularly advantageous in hearing aid systems with two hearing aids of the type described above. In such a hearing aid system, each of the two hearing aids preferably has a previously described signal processing device with a determination unit. In the at least one operating mode, a preliminary determination result is then preferably generated in each determination unit. The preliminary determination result of one hearing aid is then typically transmitted wirelessly to the other hearing aid, where it is evaluated together with its own preliminary determination result. The final determination result determined in this way is output by the corresponding determination unit. This is then available internally.In the other hearing aid, the same process preferably runs in parallel. Here, too, the two preliminary assessment results are evaluated, and a final assessment result is determined.

[0047] Irrespective of this, each hearing aid according to the invention comprises at least the two aforementioned input transducers, the aforementioned signal processing device, and the aforementioned output transducer as essential components. Each of these input transducers is generally embodied by an acousto-electrical transducer, for example, a microphone. The output transducer is usually an electro-acoustic transducer, for example, a miniature loudspeaker, or an electromechanical transducer, for example, a bone conduction receiver. The signal processing device is generally embodied by an electronic circuit implemented on a printed circuit board and, independently thereof, comprises the detection unit and usually also an amplifier or amplifier unit.In addition, each hearing aid according to the invention is typically equipped with a transmitting and / or receiving unit that enables wireless communication with other electronic devices, in particular with another hearing aid according to the invention.

[0048] Embodiments of the invention are explained in more detail below with reference to a schematic drawing. In the drawing: Fig. 1 shows a block diagram of a binaural hearing aid system with a left hearing aid and a right hearing aid, Fig. 2 shows a block diagram of a signal processing device of the left hearing aid with a detection unit, Fig. 3 shows a block diagram of the detection unit with a pre-stage with four channels, Fig. 4 in a block diagram of one of the four channels with two measuring modules, Fig. 5 in a block diagram representation of one of the two measuring modules with a filter and Fig. 6 shows a test arrangement in a plan view.

[0049] Corresponding parts are provided with the same reference numerals in all figures.

[0050] A sample described below and in Fig. The hearing aid system 2 schematically illustrated in Figure 1 consists of two hearing aids 4 and is preferably designed as a binaural hearing aid system 2. Each of the two hearing aids 4, one of which is typically designed for a left ear and one for a right ear, has, in the exemplary embodiment, two input transducers 6, a signal processing device 8, an output transducer 10, and an antenna unit 12.

[0051] In this exemplary embodiment, each of the input transducers 6 is embodied as a microphone. Miniature loudspeakers serve as output transducers 10, for example. The signal processing devices 8 are preferably each implemented as an electronic circuit on a printed circuit board. The antenna units 12 serve for wireless communication and, in particular, for data exchange between the two hearing aids 4.

[0052] The signal processing devices 8 of the two hearing aids 4 are essentially identical in the exemplary embodiment. One of these signal processing devices 8 is Fig. 2 is schematically illustrated in a block diagram and is described in more detail below. It comprises several units, namely a transmitting and receiving unit 14, two analog-to-digital converters 16, a main unit 18, a digital-to-analog converter 20, and a detection unit 22. At least one of these units further comprises several subunits, namely the detection unit 22.

[0053] For each of the units and each of the subunits, and in particular each subunit mentioned below, depending on the application, it is implemented by a separate electronic circuit forming only the corresponding unit or subunit, or merely by a separate program module / software module. A typical embodiment of the signal processing device 8 is particularly typical in which some units or subunits are implemented by a separate electronic circuit and some merely by a separate program module.

[0054] Independently of this, the determination unit 22 is designed and configured for a so-called self-voice determination in at least one operating mode of the hearing aid 4. Therefore, the corresponding hearing aid 4 is configured to automatically detect, in at least one operating mode, when the voice of a wearer of the hearing aid 4 is active and when it is not. In this at least one operating mode, the determination unit 22 then outputs a determination result, which is preferably a simple binary signal indicating when the self-voice, i.e., the voice of the wearer of the hearing aid 4, is currently active or not.

[0055] Based on this determination result or depending on this determination result, which in the embodiment according to Fig. 2 is fed to the main unit 18, an output signal for the output transducer 10 of the hearing aid 4 is then typically generated by the main unit 18. Different signal processing processes or signal processing programs are then used to generate the output signal, depending on whether the determination result indicates that the wearer's own voice is currently active or not.

[0056] Regardless of the signal processing process currently in use, the output signal is generated in such a way that the primary function of the hearing aid 4, namely the amplification of acoustic signals, is fulfilled. This occurs according to a known principle by processing input signals generated by the input converters 6. These input signals are digitized in the analog-to-digital converters 16 and fed as digital input signals to the main unit 18. Here, the digital input signals are processed using the various signal processing processes mentioned above, generating a digital output signal. This digital output signal is then converted in the digital-to-analog converter and finally fed to the output converter 10.

[0057] In parallel, the digital input signals are also fed to the detection unit 22, as shown in Fig. 2. Here, the digital input signals are then used for the aforementioned eigenvoice determination. This eigenvoice determination is carried out with the help of two subunits of the determination unit 22, namely a neural network 24, i.e., an artificial neural network, and a pre-stage 26 connected upstream of the neural network 24. The pre-stage 26 is typically designed as a multi-frequency channel pre-stage.

[0058] This means that the two digital input signals are fed to a filter bank (not shown) of the signal processing device 8 and are each broken down into sub-signals. The corresponding filter bank is designed such that two or more frequency ranges are specified by this filter bank and at least approximately a breakdown into sub-signals takes place such that each sub-signal has all frequency components of a digital input signal that can be assigned to exactly one of the frequency ranges. By way of example, the multi-frequency channel pre-stage is designed as a 4-frequency channel pre-stage, as shown in Fig. 3 is indicated.

[0059] Corresponding filter banks are known in principle, and hearing aids with such a filter bank are usually referred to as multi-frequency-channel hearing aids. In the exemplary embodiment, the hearing aids 4 of the hearing aid system 2 are designed as 4-frequency-channel hearing aids and each have a 4-frequency-channel preamplifier. Thus, the frequency spectrum is divided into 4 frequency ranges, also called frequency channels. Consequently, the two digital input signals are then fed to the detection unit 22, broken down into sub-signals. These sub-signals are then available to the preamplifier 26, where they are further processed frequency-channel-by-frequency.

[0060] The frequency channel-wise further processing is carried out with the help of subunits of the pre-stage 26, namely channel modules 28, whereby for each frequency channel k from k=1...4, i.e. each frequency range k, a channel module 28 is realized. An example of such a channel module 28 is shown in Fig. 4 shown.

[0061] Each channel module 28, in turn, has a first sub-module A as a first sub-unit and a second sub-module B as a second sub-unit. Each sub-module A further has, as an essential component, a filter a, which corresponds to a first filter type. Each sub-module B, in turn, has, as an essential component, a filter b, which corresponds to a second filter type. The pre-stage 26 therefore has a total of four filters of filter type a and four filters of filter type b.

[0062] Both a filter a of the first filter type and a filter b of the second filter type are spatial separation filters. Each of the two filter types is designed in such a way that a common directional characteristic is realized for the two aforementioned microphones of the hearing aid 4, namely a first directional characteristic by the first filter type and a second directional characteristic by the second filter type.

[0063] The two directional characteristics differ particularly with regard to the specified spatial directions in which the sensitivity of the respective directional characteristics reaches its maximum. The corresponding two spatial directions are conveniently specified relative to a frontal direction at 0°, in which the face of the hearing aid wearer typically points, as in Fig. 6. For the first filter type, a spatial direction of 0° is specified in the exemplary embodiment, and for the second filter type, a spatial direction of 180° is specified.

[0064] Preferably, the two filter types are further configured as static filters, each with fixed filter coefficients. The filter coefficients are preferably selected and specified by the manufacturer.

[0065] In the preamplifier 26 configured in this way, two partial signals are then fed to each of the two sub-components A and B of the k-th channel component 28 in frequency channel k, namely the k-th partial signal of one digital input signal and the k-th partial signal of the other digital input signal. A filter signal is then generated in both sub-component A of the k-th channel component 28 and sub-component B of the k-th channel component 28, and a real-valued attenuation value is determined based on this filter signal.

[0066] In total, eight real-valued attenuation variables are determined, and each of these attenuation variables is then passed on to the neural network 24 with its own information signal. Thus, in the exemplary embodiment, eight information signals are fed to the neural network.

[0067] In Fig. 5 shows the k-th sub-component A as an example. However, the following explanations can be applied analogously to any other sub-component, whether sub-component A or sub-component B. According to Fig. 5 the k-th partial signal of one digital input signal and the k-th partial signal of the other digital input signal are now fed to the k-th sub-component A, hereinafter referred to as the first partial signal and the second partial signal.

[0068] The first partial signal is then preferably fed unprocessed to filter a of the kth sub-component A. The second partial signal, however, is preferably fed processed to filter a of the kth sub-component A, with the second partial signal being processed for this purpose in an equalizer unit 30. The previously described spatial separation is then implemented by means of filter a and equalizer unit 30.

[0069] Based on the two partial signals, a filter signal is then generated by filter a and fed to a measuring unit 32. The unprocessed first partial signal is also fed to this measuring unit 32, and based on the filter signal and the first partial signal, one of the aforementioned attenuation variables is then finally determined in the measuring unit 32.

[0070] The previously described implementation of the kth sub-component A represents only a convenient implementation. In this implementation, the equalizer unit 30 typically has the function of multiplying the incoming signal by a frequency-dependent, complex-valued scalar weight. In an alternative implementation, the filter a takes over this function, and the equalizer unit 30 is then omitted.

[0071] To determine the attenuation value, auxiliary quantities are first determined in the measuring unit 32, for example, namely a first auxiliary quantity based on the first partial signal and a second auxiliary quantity based on the filter signal. These auxiliary quantities are then compared with each other. One such auxiliary quantity is, for example, power.

[0072] According to an alternative not explicitly shown, a complex-valued attenuation quantity is determined in the measuring unit 32 instead of a real-valued attenuation quantity. For this purpose, the measuring unit 32 typically evaluates amplitude and phase information in the signals, i.e., in the first partial signal on the one hand and in the filter signal on the other.

[0073] Two implementation variants are then provided for this alternative. In the first implementation variant, a total of eight complex-valued damping variables are present, and each of the eight complex-valued damping variables is passed on to the neural network 24 with its own information signal.

[0074] In the case of the second embodiment, there are a total of two sets of eight variables, and each of the sixteen variables is transmitted to the neural network 24 with its own information signal. In this case, for example, the real part and imaginary part, or amplitude and phase, are transmitted to the neural network 24 in separate information signals.

[0075] Furthermore, in some applications it is useful to supply one or more further information signals to the neural network 24 in addition to the previously mentioned information signals, i.e. information signals with other information.

[0076] In the exemplary embodiment shown, however, for the sake of simplicity, exactly eight information signals are supplied to the neural network 24, with eight real-valued attenuation variables being determined in the manner described above and with each of these attenuation variables being passed on to the neural network 24 with its own information signal.

[0077] This neural network 24 now has a number of artificial neurons (not explicitly shown) arranged in one or more layers. These neurons then evaluate the determined attenuation values, thereby determining an evaluation result. This evaluation result is finally output by the neural network 24. In the exemplary embodiment, an evaluation signal with a type of probability value, i.e., a value in the range from 0 to 1, is output as the evaluation result.

[0078] The evaluation signal is then fed to a comparator unit (not explicitly shown) of an output stage 34 of the determination unit 22, where it is compared with a reference, namely a reference value. The reference value is fixed and typically lies in the value range from 0.3 to 0.7, and in particular from 0.4 to 0.6, for example, 0.5. This typically generates a determination result from the comparator unit, which indicates whether or not the determination unit 22 has determined an individual voice.

[0079] The previously described process for determining the own voice typically runs continuously in the background in at least one operating mode of the hearing aid 4, and preferably this applies to each of the two hearing aids 4 of the hearing aid system 2. Thus, a previously described determination result is then generated independently in each of the two hearing aids 4. Based on the respective determination result, the respective output signal is then generated in the respective hearing aid 4.

[0080] However, a preferred embodiment is one in which only a preliminary determination result is generated by the aforementioned comparator units of the two hearing aids 4. Further preferably, each hearing aid 4 transmits its preliminary determination result to the other hearing aid 4 using the transmitting and receiving unit 14 and the associated antenna unit 12. In each of the two hearing aids 4, the two preliminary determination results of the two determination units 22 of the two hearing aids 4 are then jointly evaluated. This typically occurs in a logic unit (not explicitly shown) of the output stage 34. This ultimately generates the determination result, i.e., the final determination result.

[0081] As already explained above, the two filter types of filters a, b of sub-components A, B are preferably designed as static filters, each with fixed filter coefficients. The filter coefficients are preferably specified by the manufacturer. Further preferably, at least the detection units 22 of the two hearing aids 4 of the hearing aid system 2 are not personalized or individualized.

[0082] The corresponding filter coefficients are also preferably determined using a coefficient determination method. The determination is preferably carried out without the intervention of the future wearer, i.e., the end customer. It is also useful to use a hearing aid dummy to determine these filter coefficients during the coefficient determination method. The hearing aid dummy has at least two Fig. 6. During the coefficient determination process, input signals are then generated, preferably using the test input transducers 36, in various tests. These are expediently recorded and evaluated, for example, in a computer (not shown in detail).

[0083] In the various tests, different scenarios are preferably simulated with different test persons. Furthermore, different test persons preferably act as test subjects 38 for the test device and also preferably as useful signal sources 40 in the sense of a conversation partner, wherein the useful signal sources 40 are preferably positioned directly in front of the test subjects 38, as shown in Fig. 6 is indicated.

[0084] The calculator then preferentially determines suitable filter coefficients for each test and each simulated scenario, particularly for both the previously described filters a and b. This results in a data set with a large number of suitable values ​​for each filter coefficient.

[0085] For example, a so-called k-means algorithm is then applied to the corresponding data sets in order to obtain the filter coefficients that are ultimately specified for the filters a,b of the sub-modules A,B. List of reference symbols 2 hearing aid system 4 Hearing aid 6 input converters 8 Signal processing device 10 output converters 12 Antenna unit 14 Transmitting and receiving unit 16 analog-to-digital converters 18 Main unit 20 digital-to-analog converters 22 Investigation Unit 24 neural network 26 Pre-stage 28 channel module 30 Equalizer unit 32 measuring units 34 power amplifier 36 test input converters 38 test carriers 40 Useful signal source A sub-module B Submodule a filter b Filter QUOTES CONTAINED IN THE DESCRIPTION

[0000] This list of documents submitted by the applicant was generated automatically and is included solely for the convenience of the reader. This list is not part of the German patent or utility model application. The DPMA assumes no liability for any errors or omissions. Cited patent literature

[0000] EP 3 222 057 B1

[0006]

Claims

[1] Method for determining an activity of an own voice of a wearer of a hearing aid (4) with the aid of a signal processing device (8) of the hearing aid (4), wherein - a first input signal is generated by means of a first input transducer (6) of the hearing aid (4) and a second input signal is generated by means of a second input transducer (6) of the hearing aid (4), - the two input signals are fed to a detection unit (22) of the signal processing device (8), which has a neural network (24) and a pre-stage (26) connected upstream of the neural network (24), - information signals are generated by means of the pre-stage (26) based on the two input signals, - the information signals are evaluated by the neural network (24) and - based on the evaluation of the information signals by the neural network (24), a determination result is output by the determination unit (22). [2] Method according to claim 1, wherein personalization of the determination unit (22) is dispensed with. [3] Method according to claim 1 or 2, wherein the neural network (24) is trained with the aid of a number of recorded acoustic signals in which the eigenvoices of different test carriers are recorded. [4] Method according to one of claims 1 to 3, wherein the pre-stage (26) has two filters (a,b), namely a first filter (a) and a second filter (b), wherein the first filter (a) corresponds to a first filter type and the second filter (b) to a second filter type and wherein a first filter signal is generated by means of the first filter (a) and a second filter signal is generated by means of the second filter (b). [5] Method according to claim 4, wherein the first filter type is designed to extract useful signals from useful signal sources and wherein the second filter type is designed to mask out eigenvoice. [6] Method according to claim 4 or 5, wherein the two filter types are designed as static filters (a,b) with fixed predetermined filter coefficients. [7] Method according to claim 6, wherein the filter coefficients for the second filter type are predetermined based on an evaluation of recorded acoustic signals in which eigenvoices of different test carriers are recorded. [8] Method according to claim 7, wherein at least one adaptive test filter is used for the evaluation. [9] Method according to one of claims 4 to 8, wherein each of the filter signals is combined with a reference signal based on at least one of the two input signals to generate an information signal. [10] Method according to claim 9, wherein by linking one of the filter signals with one of the reference signals an attenuation value is determined, which is transmitted to the neural network (24) with the corresponding information signal. [11] Method according to one of claims 1 to 10, wherein the information signals are evaluated by the neural network (24) and wherein a prediction value is output by the neural network (24) as an evaluation. [12] Method according to claim 11, wherein the information signals are evaluated by the neural network (24), which is designed as a recurrent neural network (24), and wherein the prediction value is output by the neural network (24) as an evaluation. [13] Method according to one of claims 1 to 12, wherein a preliminary determination result is determined by the determination unit (22), wherein the preliminary determination result is evaluated together with a further preliminary determination result in the determination unit (22), which is determined by a further determination unit (22) in a further hearing aid (4) and transmitted by the further hearing aid (4), and wherein the determination result in the determination unit (22) is determined by evaluating the two preliminary determination results. [14] Hearing aid (4) configured to carry out at least one method step of a method according to one of the preceding claims in at least one operating mode. [15] Hearing aid system (2) comprising two hearing aids (4) according to the previous claim.

Citation Information

Patent Citations

  • Hearing device or system comprising a user identification unit

    US20200294508A1