WO2008107359A1 - Hearing system with distributed signal processing and corresponding method - Google Patents

Abstract

The power consumption of hearing devices of a hearing system is to be reduced. To this end, it is provided to distribute complex algorithms to the processors of a hearing device (10) and an additional device (20) of the hearing system. For example, an input signal is transmitted from the hearing device (10) to the additional device (20), or vice versa, and processed in a partial process (22) of the total signal processing process. A result of the partial process (22) is transmitted (32) to the hearing device (10). Thereafter, an output signal resulting from the input signal is transmitted back to the device (20) from which the input signal originates. Finally, the received output signal is further processed (14, 16, 24, 26) in the respective devices (10, 20).

Description

description

Hearing system with distributed signal processing and corresponding method

The present invention relates to a hearing system with a hearing device that is portable at one ear and a second device that has a separate housing relative to the hearing device, wherein the hearing device and the second device each have a signal processor and are coupled to each other by a data transmission device for data transmission , Moreover, the present invention relates to a corresponding method for carrying out a signal processing process for a corresponding hearing system. A hearing device is understood to mean, in particular, a device that can be worn on the ear, in particular a hearing device, a headset, headphones and the like.

Hearing aids are portable hearing aids that are used to care for the hearing impaired. In order to meet the numerous individual needs, different types of hearing aids such as behind-the-ear hearing aids (BTE) and in-the-ear hearing aids (ITO), e.g. also Concha hearing aids or canal hearing aids (CIC), provided. The exemplarily listed hearing aids are worn on the outer ear or in the ear canal. In addition, bone conduction hearing aids, implantable or vibrotactile hearing aids are also available on the market. The stimulation of the damaged hearing takes place either mechanically or electrically.

Hearing aids have in principle as essential components an input transducer, an amplifier and an output transducer. The input transducer is usually a sound receiver, z. As a microphone, and / or an electromagnetic receiver, for. B. an induction coil. The output transducer is usually used as an electroacoustic transducer, z. As miniature speaker, or as an electromechanical transducer, z. B. bone conduction, realized. The amplifier is usually way integrated into a signal processing unit. This basic structure is shown in FIG. 1 using the example of a behind-the-ear hearing device. In a hearing aid housing 1 for carrying behind the ear, one or more microphones 2 for receiving the sound from the environment are installed. A signal processing unit 3, which is also integrated in the hearing aid housing 1, processes the microphone signals and amplifies them. The output signal of the signal processing unit 3 is transmitted to a loudspeaker or earpiece 4, which outputs an acoustic signal. The sound is optionally transmitted via a sound tube, which is fixed with an earmold in the ear canal, to the eardrum of the device carrier. The power supply of the hearing device and in particular of the signal processing unit 3 is carried out by a likewise integrated into the hearing aid housing 1 battery. 5

Blind source separation enables statistical signal processing of at least two microphone signals to separate source signals without prior knowledge of the geometric arrangement of the sources. This procedure brings in the

Use in hearing aids Advantages over conventional directional microphone approaches, where the preferred direction is set to the front (0 °).

In principle, with the blind source separation method (BSS) with N microphones N sources can be separated, i. Generate N output signals. In choosing which of the N output signals is supplied to the hearing aid wearer, there are various possibilities.

In addition, the combination of a BSS algorithm with other algorithms of statistical signal processing, such. B. a situation detection conceivable. Algorithms of this kind are based on several complex arithmetic operations that require a high level of computing power. However, computing power is directly related to power consumption and thus a critical value in hearing aids. Since BSS algorithms usually require very high computing power, the criterion of low power consumption in blind source separation for hearing aids with conventional technology can only be achieved to a certain extent.

From the document WO 99/43185 a binaural, digital hearing aid system is known. The input signals of two hearing aids of the hearing aid system are mutually exchanged. Both hearing aids have signal processors that can process their own and the input signal received from the other hearing aid. It therefore arises in both hearing aids according to high computational complexity.

Furthermore, document WO 02/28143 A2 discloses a method for operating a hearing aid system with at least two hearing aid devices. In the hearing aid devices, sound field characteristics are generated and transmitted to different hearing situations between the hearing aids to adapt the signal processing units. Both hearing aids are thus always operated in the same hearing program.

Furthermore, document WO 00/00001 discloses a synchronization method for a binaural hearing system. The hearing system again comprises two hearing aids. In order to operate the hearing aids, each of which is switchable in situ in at least two transmission modes from the respective microphone to the respective output transducer arrangement, to operate binaurally, the respectively active transmission modes of the hearing aids are synchronized via a wireless connection between the hearing aids.

Furthermore, DE 10 2004 053 790 A1 discloses a method for generating stereo signals for separate sources. For this purpose, a blind source separation of at least two microphone signals for the extraction of BSS filters is performed. After proper filtering, two stereo signals are obtained for each microphone signal. A method for reducing interference power in a microphone is shown in the document DE 10 2004 052 912 A 1. In this case, the microphone signals of several microphones are adaptively filtered as a function of at least one parameter. The directivity of the directional microphone obtained in this case is adjusted by changing the at least one parameter such that the sum of noise power including microphone noise is reduced.

The object of the present invention is a

Provide hearing system that shows the lowest possible power consumption with high-quality signal processing. In addition, a corresponding signal processing method will be presented.

According to the invention, this object is achieved by a hearing system with a hearing device that is portable at one ear and a second device that has a separate housing relative to the hearing device, wherein the hearing device and the second device each have a signal processor and by a data transmission device with each other for data transmission are coupled, and wherein a signal processing process is divided into two signal processors, so that two sub-processes of the signal processing process on the two signal processors can run parallel in time, and an input signal for the signal processing process from the hearing to the second device or vice versa, the transferable Input signal there in the respective subprocess processed and the resulting output signal for further processing to the other device is retransmitted. It does not have to be a single input signal, but can be provided on both sides of several input signals.

In addition, the invention provides a method for carrying out a signal processing process for a hearing system, the at least one hearing device and a further device with separate housings and one each Signal processor, by transmitting an input signal from the hearing device to the other device or vice versa, by processing the input signal in a subprocess of the signal processing process, which runs in that of the two devices that had received the input signal by retransmitting one of the input signal Output signal to the device from which the input signal comes, and by further processing the output signal of those of the two devices, which had received the output signal from the other device.

Advantageously, it is thus possible to distribute very complex computation algorithms to a plurality of signal processors of the hearing system. This makes it possible to achieve a saving of computing capacity in each individual device in favor of the required power consumption.

Preferably, the second device is also a hearing device. This makes it possible to divide the computational effort arising in the hearing system between the two hearing devices or hearing aids, from which both hearing devices profit.

The second device may also be a remote control with which the hearing device can be operated. One

Part of the computing capacity of the system is thus outsourced to the remote control, in which the power consumption plays a much smaller role than in a device worn on the ear. The remote control can also serve to operate a binaural hearing system, so that the computing capacity can be divided into three or more devices.

According to a special embodiment of the hearing system according to the invention with two hearing devices, both receive an input signal, exchange it via the data transmission devices and calculate a part of the filter parameters for blind-source separation filtering. In each of the two hearing devices, a Fii- tion with the self-calculated filter parameters feasible, and the filter results are merge in one of the two hearing for further processing. This distributes the complex creation of filters for blind source separation to both hearing aids.

According to another embodiment of the hearing system with two hearing devices, the hearing devices also each receive an input signal and exchange it via the data transmission devices. In one of the two hearing devices, all filter parameters are calculated for blind source separation filtering, and in the other hearing device the corresponding complete filtering is carried out after transmission of the filter parameters. In this case, the entire package of filter calculation and filtering is distributed to both hearing devices.

According to a further embodiment, it is provided that in one of the hearing device and the second device of the hearing system according to the invention a blind

Source separation algorithm and in the other at the same time a classification algorithm is executed. As a result, two equally compute-intensive algorithms are distributed to the individual processors in the hearing system. In principle, one or more arbitrary algorithms can be distributed among the processors of the hearing system.

Conveniently, the data is wirelessly transmitted between the components of the hearing system. Thus, the available processors of the hearing system can be used flexibly for complex calculations also spontaneously.

The present invention will be explained in more detail with reference to the accompanying drawings, in which:

1 shows the basic structure of a hearing aid and 2 shows a block diagram of a hearing aid system according to the invention.

The embodiments described in more detail below represent preferred embodiments of the present invention.

As already stated above, the solution according to the invention to the problem of power consumption reduction is to distribute the computation of complex algorithms, such as BSS, over several signal processors. As signal processors in this case, both those of two hearing aids and another external signal processor, which is housed for example in a remote control conceivable. Thus, the required computing power can be distributed to the processors, so that the entire computing capacity of the system can be reduced or used more meaningfully.

According to the first exemplary embodiment shown in FIG. 2, in a hearing aid system with two hearing aids, the calculation of filter values and the filtering itself are split between the two hearing aids 10, 20. In Figure 2, the two hearing aids 10, 20 separated by a dashed line. Even though they are physically separate, there is data communication between them.

In the concrete example, this means that the first hearing device 10 has at least one microphone 11 and likewise the second hearing device 20 has at least one microphone 21. The two hearing aids first exchange the recorded signals, i. the microphone or input signals to each other. For the sake of clarity, the entire signal processing of each hearing aid 10, 20 is symbolized only by the filtering that is essential here. Thus, the input signals in the hearing device 10 are fed to a filter unit 12, which is an individual BSS

Filtering. For this purpose, the filter unit 12 in the present example has four filters Fl, F2, F3 and F4, which are to be fed with individual parameters. The filtering on However, it is very compute-consuming. The calculation of the filter parameters themselves is likewise computation-intensive. Therefore, the calculation of the filter parameters is shifted here into the second hearing device 20. There, an adaptation unit 22 calculates the filter parameters and delivers them via cable connection or wireless to the first hearing device 10 or its filter unit 12. After filtering, the BSS output signals are fed to a selection unit 13, with which the desired source is automatically selected, for example, by classification can be. The selected signal will now be for the binaural

Play also forwarded to the second hearing aid 20. Therefore, in addition to the first data connection 30, 31 for transmitting the input signals and the second data connection 32 for transmitting the filter parameters, a third data connection 33 for transmitting the selected signal from the first hearing aid 10 to the second hearing aid 20. For binaural reproduction takes place in the respective Hearing a corresponding filtering or signal processing, symbolized by the transfer functions Gl and G2, in processing units 14, 24 instead. The two transfer functions G1 and G2 can be calculated, for example, from the filter coefficients F1, F2, F3 and F4 according to the document DE 10 1004 053 790 A1. Therefore, a control line 15 from the filter unit 12 to the processing unit 14 is provided in the first hearing device 10. Likewise, a control line 25 is provided between the adaptation unit 22 and the processing unit 24 in the second hearing device 20. The output signals of the processing units 14, 24 are supplied to a respective loudspeaker 16, 26 in the respective hearing device 10, 20.

In the example of FIG. 2, it is shown that the entire BSS processing need not be carried out individually in each hearing device 10, 20. Rather, a part of the BSS algorithm is executed or calculated in the one hearing device and the remaining part in the other hearing device. The resulting filter signal is then made available to both hearing aids. Due to the fact that the BSS processing for the two hearing aids has to be carried out only once, computing capacity can be saved, whereby the power consumption of both hearing aids decreases. However, the distribution of computationally intensive, complex algorithms on a plurality of signal processors is not limited to the example of FIG. Rather, the division into the signal processors can also be carried out in such a way that, in the case of BSS processing, both the adaptation, ie the filtering parameter calculation and the filtering itself, each depend on the two hearing aids or the hearing aid and the remote control or the hearing aid and a separate external signal processor is divided. For this purpose, the hearing aids or the devices involved in the division initially exchange the recorded signals with each other. In a further step, these signals are supplied to the individual BSS filters, wherein each device calculates only half of the filters and the necessary parameters to the other device for

Will be provided. For example, based on the example of FIG. 2, the filter values Fl and F2 in the first hearing device and the filter values F3 and F4 in the second hearing device can be calculated. Each hearing device carries out the corresponding filtering and one of the two hearing aids then transmits its filtered signals to the other hearing device for further processing, so that in the latter hearing device, for example, the summation can be carried out, which is indicated in FIG. 2 in the filter unit 12. The further signal processing can then take place according to the example of FIG.

According to a further embodiment, it is provided that the BSS processing and a classification are divided into two hearing aids or devices. The hearing aids initially exchange the recorded signals with one another, as in the other exemplary embodiments. The first hearing aid processes these signals through the classification unit, while the second hearing aid calculates the BSS algorithm. During these calculations, control signals, for. B. for source selection, to be exchanged between two hearing aids. Finally, the exchange of the final processed signals takes place again and classification data for further hearing aid signal processing.

The above examples merely indicate how complex algorithms can be distributed among several devices and in particular hearing aids. In this case, it should not be restricted to BSS algorithms, but instead should be shown that, in principle, all arithmetic operations of a hearing system can be divided among the processors or devices involved.

Claims

claims 1st hearing system with - A hearing device (10) which is portable to an ear. - A second device (20) which relative to the hearing device (10) has a separate housing. - The hearing device (10) and the second device (20) each having a signal processor and by a data transmission device (30 to 33) coupled to each other for data transmission, characterized in that a signal processing process is divided between the two signal processors, so that two subprocesses (12, 22) of the signal processing process on the two signal processors can run in parallel in time, and an input signal for the signal processing process from the hearing device (10) to the second device (20) or conversely transferable, the input signal in the respective sub-process can be processed and the resulting output signal is retransmitted to the other device for further processing. 2. Hearing aid according to claim 1, wherein the second device (20) is also a hearing device. 3. The hearing aid according to claim 1, wherein the second device (20) is a remote control. 4. Hearing aid according to claim 2, wherein the two hearing devices each receive an input signal, via the data transmission means (30 to 33) exchange and a blind source separation filtering each calculate a part of the filter parameters, and in each of the two hearing devices each with a filtering Even calculated filter parameters can be carried out, and the filter results in one of the two hearing devices for further processing are merge. 5. Hearing aid according to claim 2, wherein the two hearing devices each receive an input signal, via the data transmission device (30 to 33) exchange, in one of the two hearing devices for a blind Source separation filtering filter parameters calculated and in the other hearing the corresponding filtering after transmission of the filter parameters is feasible. 6. Hearing aid according to one of claims 1 to 3, wherein in one of the hearing device (10) and the second device, a blind source separation algorithm and in the other at the same time a classification algorithm is executable. 7. Hearing aid according to one of the preceding claims, wherein the data transmission device (30 to 33) wireless data transmission is possible. 8. A method for executing a signal processing process for a hearing system, comprising at least one hearing device and another device with separate housings and each having a signal processor, characterized by, - transmitting an input signal from the hearing device (10) to the further device (20) or vice versa . Processing the input signal in a subprocess of the signal processing process running in that of the two devices which had received the input signal, Retransmitting an output signal resulting from the input signal to the device from which the input signal originates, and - processing the output signal of those of the two devices which had received the output signal from the other device. 9. The method of claim 8, wherein in one of the hearing device (10) and the further device (20) run a blind source separation algorithm and in the other at the same time a classification algorithm. 10. The method of claim 8, wherein the further device (20) is also a hearing device, the two hearing devices each receive and exchange an input signal, each calculate a part of the filter parameters for blind source separation filtering and perform filtering with the self-calculated filter parameters, and merged the filter results in one of the two hearing aids and further processed. 11. The method of claim 8, wherein the second device (20) is also a hearing device, the two hearing devices each receive and exchange an input signal, all filter parameters are calculated in one of the two hearing devices for blind-source separation filtering and the corresponding complete filtering is carried out in the other hearing device after transmission of the filter parameters becomes.