EP2070384B1 - Hearing device controlled by a perceptive model and corresponding method - Google Patents

Description

The present invention relates to a hearing device with a signal processing device for processing an input signal to an output signal and a modeling device in which a perceptive model is implemented in order to generate a psychoacoustic value for driving the signal processing device. Moreover, the present invention relates to a corresponding method for operating a hearing device. The term hearing device is understood here in particular to be worn on the ear device, such as a hearing aid, a headset, headphones and the like.

Hearing aids are portable hearing aids that are used to care for the hearing impaired. To meet the numerous individual needs, different types of hearing aids such as behind-the-ear hearing aids (BTE), behind-the-ear hearing aids with external handset (RIC: RECEIVER IN THE CANAL) and in-the-ear hearing aids ( IdO), eg Concha hearing aids or canal hearing aids (ITE, CIC). The hearing aids listed by way of example 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 integrated in a signal processing unit. This basic structure is in FIG. 1 shown using the example of a behind-the-ear hearing aid. 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

The nature of the hearing aid supply provides a pre-setting of the hearing system to be adapted on the basis of hearing loss. The further course of the hearing aid adaptation is characterized by fine adjustment steps based on the reports of the hearing aid wearer. Depending on the situation described, the hearing care professional then attempts to transfer the subjective hearing impressions of the hearing impaired to technical parameters of the hearing system. As a rule, however, it is not possible to configure the parameters of a hearing system in such a way that the hearing system corresponds to the individual hearing wishes of the hearing impaired person in all situations.

The adaptation of hearing devices or hearing systems is done so far with a variety of measures. For example, potentiometers on hearing systems are used, with which the hard of hearing has the opportunity to independently adjust a psychoacoustic dimension (loudness) depending on the situation. In addition, the use of so-called multi-memory devices is known, which give the hearing aid wearer the possibility, depending on the acoustic situation, an alternative configuration of the hearing system to load. However, classifiers that classify the acoustic environment and automatically adjust the parameters of the hearing system based on a logic dictated by the manufacturer can also be implemented in the hearing systems. In addition, learning hearing aids are also known, which automatically adjust their parameterization within given tolerances based on user changes.

Further, the document discloses US 2002/0111745 A1 a portable hearing aid system. In this case, parameters of a hearing response can be obtained by audiometer. A response prediction is used to perform a basic setting of a hearing aid.

Furthermore, the document describes EP 0 661 905 A2 a generic method for adjusting a hearing aid and a corresponding hearing aid. A perceptive model is used to obtain a psychoacoustic variable, in particular loudness, on the one hand for a norm group and on the other hand for a single individual. On the basis of the difference between the two psychoacoustic variables, setting data are determined, with which the signal transmission to a hearing device is designed or adjusted ex situ or conducted in situ.

From the publication DE 103 08 483 A1 is a method for automatic strength adjustment known. In particular, the speech intelligibility is to be improved when supplied with a hearing aid. Therefore, during operation, automatic adjustment of the gain takes place as a function of the detected signal levels and the signal frequency. In this case, the determination of gain parameters taking into account a loudness model and a speech intelligibility model takes place.

In the US 2007/0135862 A1 For example, a method and a system for adapting a hearing device to a wearer of the hearing device as part of an adjustment process are disclosed. in the Method is thereby determined for an input signal a desired perception. This can be done using known perceptive models such as loudness models. Subsequently, one or more stimulus signals, which correspond to the input signal, are given to the wearer of the hearing device, and then the perception at the wearer of the hearing device is measured. The stimuli signals are generated via different stimuli modes of the hearing device, each stimulus signal is generated by means of a weighting. During the fitting session, the weighting of the weighting is adjusted on the basis of the measured evoked perception and the desired perception by the wearer of the hearing device. The perceptive models can be used to determine the desired perception and the rules for changing the stimulus signals.

The object of the present invention is to make the adaptation of a hearing device as simple as possible and to propose a corresponding hearing device and a related method.

According to the invention, this object is achieved by a hearing device having a signal processing device for processing an input signal to an output signal, a modeling device in which a perceptive model is implemented to generate a psychoacoustic value for driving the signal processing device, wherein a hearing loss imaging data can be entered into the modeling device and the perceptive model obtains the psychoacoustic value for driving the signal processing device from the data and the output signal. The data representing the hearing loss may in particular be audiogram data.

Further, according to the present invention, there is provided a method of operating a hearing apparatus by processing an input signal to an output signal in the hearing apparatus, obtaining a psychoacoustic value using a perceptual model, and controlling the input signal based on the psychoacoustic value, the perceptual model resulting in hearing loss imaging data, in particular audiogram data, and the output signal of the hearing aid acquires the psychoacoustic value for controlling or regulating the processing.

Advantageously, it is thus possible to use a psychoacoustic model of hearing impairment to determine nominal values that control or regulate the signal processing.

The modeling device preferably receives level information and / or classification information relating to the input signal for generating the control signal. Hereby, the perceptive model can be parameterized according to the current hearing situation. In interaction with the psychoacoustic model, the signal processing can thus be parameterized in an advantageous manner.

The output signal of the signal processing device can be transmitted to the modeling device indirectly via a receiver and a probe microphone of the hearing device. In this way, the transfer function of the handset or loudspeaker of the hearing device for the control of the signal processing device can be taken into account. Alternatively, the acoustic output signal of the hearing aid can be suitably modeled and supplied to the psychoacoustic model in digital form.

In particular, the psychoacoustic value may relate to loudness, pleasantriness, sharpness, roughness, or hearing effort. In principle, of course, any other psychoacoustic dimensions can also be used to adapt or control the hearing device.

Particularly advantageous is the constant tracking of one or more parameters of the signal processing device by the modeling. As a result, the hearing device can always be adapted individually to the current hearing situation. In this case, the one or more parameters relate, for example, to the amplification, the compression, the directional microphone characteristic or the noise suppression of the hearing device.

A further particularly noteworthy embodiment is that the modeling device obtains a plurality of psychoacoustic values and respectively compares them with desired values, and then combines the corresponding difference values weighted into an error variable, wherein the signal processing device is controlled or regulated such that the error magnitude is minimized , In this way, a multidimensional space of psychoacoustic variables is used to control the hearing device. In this case, the setpoint values can be changed by the user via a potentiometer of the hearing device or a remote control of the hearing device. alternative the setpoints are specified by the audiologist, possibly also in multi-program devices.

FIG 1

den prinzipiellen Aufbau eines Hörgeräts gemäß dem Stand der Technik und

FIG 2

ein Blockschaltdiagramm eines erfindungsgemäßen Hörgeräts.

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

FIG. 1

the basic structure of a hearing aid according to the prior art and

FIG. 2

a block diagram of a hearing aid according to the invention.

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

According to the example of FIG. 2 a hearing aid according to the invention is equipped with at least one microphone 10, which supplies an input signal for a signal processing unit 11. The output signal of the signal processing unit 11 is supplied to a loudspeaker or earphone 12. The signal processing unit 11 can be parameterized in a known manner, for example with regard to amplification, filtering. Parameterization or programming takes place, for example, as part of an adaptation. In the present example, a modeling device 13 is used for automatic parameterization of the signal processing unit 11. The modeling device 13 has a perceptive model, which is explained in greater detail below. In principle, the perceptive model serves to convert the output signal of the hearing device into a subjective perceptual dimension (eg loudness). This psychoacoustic variable is then used to drive the signal processing unit 11. In the simplest case, therefore, the output signal of the signal processing unit 11 is tapped off and provided to the modeling device 13. However, the transfer function of the handset 12 is disregarded. In a first approximation, the transfer function of the listener and the subsequent acoustic Coupling can be modeled. If one also wants to take this transfer function into account in gaining the psychoacoustic variable, it is necessary, for example, to introduce into the ear canal a probe microphone 14 for the measurement, in order to precisely measure the actual sound situation in front of the eardrum. The alternative tap of the output signal is in FIG. 2 dashed lines.

The probe microphone 14 can also be used to somewhat adjust the simple tap of the output signal of the signal processing unit 11. For this purpose, a single measurement with the probe microphone is sufficient so that the difference between the two taps can be determined and taken into account in the model of the modeling device. It is then possible to continue working with this corrected model without continuing to use the probe microphone 14.

The perceptive model in the modeling device 13 is individualized in that the hearing loss, for. B. described by the audiogram of the hearing impaired, is provided via a programming socket 15 of the modeling device 13. Based on the output signal from the signal processing unit 11 or the probe microphone 14, the modeling device 13 now generates a control signal S for the signal processing unit 11 on the basis of the output signal from the signal processing unit 11, so that it is correspondingly parameterized.

Optionally, it can be provided that the modeling device 13 is supplied by a level meter 16 with a level signal from the input signal of the signal processing unit 11. Alternatively or additionally, a classifier 17 classifies the input signal and provides the modeling device 13 with a corresponding classification signal. With the input level signal and / or the classification signal, a more differentiated control signal S can be obtained by the modeling device 13. Alternatively, the signals of the optional level meter 16 and the optional classifier instead of the modeling device 13 of the signal processing unit 11 supplied.

According to the invention, the above-described problem of the simplified adaptation of a hearing device or a hearing device is thus solved by implementing a single perceptive model of hearing impairment in addition to the various algorithms on the chip of a hearing aid. The computational effort for the modeling is thus lower, which is why the in FIG. 2 illustrated circuit can also be implemented in a hearing aid, which in the method according to the cited document EP 0 661 905 A2 hardly possible.

Perceptual models suitable for implementation are already known, for example, under the names "PEMO-Q, PHAQM, MCHI". Necessary quantities for a processing of the models are usually information about the hearing loss (audiometric hearing loss) as well as an "audio stream", d. H. a listener edition.

On the basis of these data, perceptive models provide information about psychoacoustic dimensions such as loudness, pleasantness, sharpness and roughness. In addition, other psychoacoustic variables are conceivable, such. As listening effort, subjective speech intelligibility or transmission quality.

With the perceptual model implemented on the output side, several psychoacoustic characteristics can also be obtained from the audio stream in conjunction with the tonomaudiometric hearing loss. If one of the parameters falls below a previously defined level, the parameters of the hearing system are tracked automatically in order not to fall below the specified minimum value, for example for the loudness. Analogously to this example, the other psychoacoustic parameters mentioned can also be automatically optimized by: Parameters such as gain, compression, directional characteristics, noise removal, etc. are tracked automatically. The number of parameters to be tracked is not necessarily limited.

According to a further developed embodiment, an optimization of a composite of selected characteristic variables is constantly carried out and the parameters are correspondingly continuously adjusted adaptively. For example, the loudness is kept in a predetermined setting range. This is possible, for example, by establishing a common error function from the weighted characteristics according to the following equation: $$\mathrm{error}\left(\mathrm{t}\right)=\mathrm{G}\mathrm{1}*{\left(\mathrm{LH}\left(\mathrm{t}\right)-\mathrm{LH\_opt}\right)}^{\wedge }\mathrm{2}+\mathrm{G}\mathrm{2}*\left(\mathrm{HA}\left(\mathrm{t}\right)-\mathrm{HA\_opt}\right)+\mathrm{G}\mathrm{3}*$$

LH, LH_opt:

Lautheit, bzw. optimale Lautheit (1. Kenngröße)

HA, HA_opt:

Höranstrengung bzw. optimale Höranstrengung (2. Kenngröße)

g1, g2, g3, ...

individuelle Gewichtung des Beitrags dieser Kenngrößen zum Gesamtfehler

(t):

akt. Zeitpunkt

Where:

LH, LH_opt:

Loudness, or optimal loudness (1st characteristic)

HA, HA_opt:

Listening effort or optimal listening effort (2nd parameter)

g1, g2, g3, ...

individual weighting of the contribution of these parameters to the total error

(T):

act. time

The aim now is to continuously adaptively adjust the parameters to be optimized (gain, compression, directional microphone, noise removal, etc.) based on the minimization of this error function. Weighting can take into account the importance of the parameter in the optimization. In the special case that only one parameter is to be optimized, its weight must be set to one and those of the other parameters to zero. The sum of all weights always returns the value 1.

If an analytical description of the parameters exists as a function of the parameters to be optimized, directly known Optimization method (eg LMS, RLS) are used, the derivations of the parameters according to the parameters to be optimized. Otherwise, the error can also be determined directly for each parameter at closely adjacent values, thus determining the direction of the preferred parameter change.

The adaptation according to the invention can also be carried out in multi-memory devices. In this case, different programs of a hearing aid could be designed to maximize the psychoacoustic dimension "Pleasure" in the basic program and to minimize another dimension such as "listening effort" in another program. Thus, it is not necessary to load each other programs, but only to adjust the parameterization of the psychoacoustic control unit. In this example, the user can switch between the different operating modes via a suitable operating element, such as a push button on the hearing system, a remote control, voice control, etc.

The switching of the operating modes can also be done automatically. For this purpose, the hearing aid for a certain period on the switching behavior of the hearing aid wearer in various operating modes such. In addition, the hearing device registers certain characteristics of the input signal (eg level, degree of modulation, pitch, formand, etc.) at the switching times and thus links the switching behavior with the characteristic of the input signal. With the thus trained learning function, the hearing aid can automatically switch the operating modes depending on the input signal and the requirements of the hearing aid wearer after a learning period.

In addition to the change between discrete operating modes, it is also conceivable to adjust threshold values with respect to the psychoacoustic parameters via a potentiometer or a remote control in finer granularity. These thresholds can on the one hand be set directly by the user by means of a suitable input medium or readjusted automatically via a learning algorithm individually.

According to a further embodiment, it is possible to use further parameters for tracking the parameterization of the hearing system, as described above in connection with FIG FIG. 2 already indicated. For example, the currently classified acoustic situation can be used for tracking. In particular, in a recognized situation "speech in noise" the weighting can be drawn more towards minimum listening effort, whereas in the listening situation "music" the optimization with regard to maximum sound quality is in the foreground.

Furthermore, a history of the acoustic situations from a datalogging can also be used for updating the parameterization of the hearing system.

In the above example, the method according to the invention is implemented in a hearing aid. However, it is also conceivable to implement the method on a further device with which the necessary data is exchanged. In addition to a wired solution, the data exchange may be wireless.

With the present invention, an automatic control of a hearing system by psychoacoustic characteristics and not by statistical, pre-optimized settings of a situation detection unit is thus possible. This results in several advantages. On the one hand, a basic setting of the hearing systems is dispensed with using a prescriptive fitting formula, since the hearing system or the hearing device adaptively tracks all parameters in order to optimize the result of the perceptual model. In addition, the hearing system has the objective of optimally providing for individual hearing loss, whereas previous pre-optimized approaches satisfy the individual hearing impaired only on average. In principle, that would be Use of the present invention also conceivable in normal hearing, so that they could benefit, for example, from an adaptive hearing protection in noisy or acoustically difficult environments.

Claims (17)

1. Hearing aid having

   - a signal processing device (11) for processing an input signal to produce an output signal,

   - a modelling device (13) in which a perceptive model (MOD) is implemented in order to produce a psychoacoustic value for actuation of the signal processing device (11) by means of a control signal (S),
   characterized in that

   - data mapping a hearing loss can be input into the modelling device (13) and

   - the perceptive model (MDD) takes the data and the output signal and extracts the psychoacoustic value for actuation of the signal processing device (11).
2. Hearing aid according to Claim 1, wherein the data mapping the hearing loss comprise audiogram data.
3. Hearing aid according to Claim 1 or 2, wherein the modelling device (13) receives a level signal and/or a classification signal for the input signal in order to produce the control signal (S).
4. Hearing aid according to Claim 1, 2 or 3, wherein the output signal from the signal processing device (11) is transmitted to the modelling device (13) indirectly via an earpiece (12) and a probe microphone (14) of the hearing aid.
5. Hearing aid according to one of the preceding claims, wherein the psychoacoustic value relates to loudness, pleasantness, shrillness, harshness or hearing effort.
6. Hearing aid according to one of the preceding claims, wherein one or more parameters of the signal processing device (11) are constantly tracked by the modelling device (13).
7. Hearing aid according to Claim 6, wherein the or one of the plurality of parameter(s) relates to the gain, the compression, the directional microphone characteristic or the noise reduction of the hearing aid.
8. Hearing aid according to one of the preceding claims, wherein the modelling device (13) is used to extract a plurality of psychoacoustic values and to compare each of them with setpoint values, and then the corresponding difference values are combined in weighted fashion to produce an error variable, wherein the signal processing device (11) is controlled or regulated such that the error variable is minimized.
9. Hearing aid according to Claim 8, wherein the setpoint values can be altered by means of a potentiometer of the hearing aid or a remote control of the hearing aid.
10. Method for operating a hearing aid by

    - processing an input signal to produce an output signal in the hearing aid,

    - extracting a psychoacoustic value using a perceptive model (MOD) and

    - controlling or regulating the processing of the input signal on the basis of the psychoacoustic value,
    characterized in that

    - the perceptive model (MOD) takes data mapping a hearing loss and the output signal and extracts the psychoacoustic value for control or regulation of the processing.
11. Method according to Claim 10, wherein the data mapping the hearing loss comprise audiogram data.
12. Method according to Claim 10 or 11, wherein a level signal and/or a classification signal for the input signal is additionally used for controlling or regulating the processing.
13. Method according to one of Claims 10 to 12, wherein the output signal is made available to the perceptive model (MOD) indirectly via an earpiece (12) and a probe microphone (14).
14. Method according to one of Claims 10 to 13, wherein the psychoacoustic value relates to loudness, pleasantness, shrillness, harshness or hearing effort.
15. Method according to one of Claims 10 to 14, wherein one or more parameter(s) for the processing are tracked using the psychoacoustic value.
16. Method according to Claim 15, wherein the or one of the plurality of parameter(s) relates to the gain, the compression, the directional microphone characteristic or the noise reduction of the hearing aid.
17. Method according to one of Claims 10 to 16, wherein the perceptive model (MOD) is used to extract a plurality of psychoacoustic values and to compare each of them with setpoint values, and then the corresponding difference values are combined in weighted fashion to produce an error variable, wherein the processing is controlled or regulated such that the error variable is minimized.

Images