Method for training and operating a hearing aid and corresponding hearing aid

The present invention relates to a process for retraining a hearing aid by providing an acoustic Input signal, providing multiple Hörsituationskennungen and assigning the input acoustic signal to a the Hörsituationskennungen by a hearing aid wearer. Moreover, the present invention relates to a corresponding Hearing aid that is nachtrainierbar, and a method for operating a hearing aid according to one Retraining.

Classifiers found in hearing aids for the detection of different Situations use. The preset parameters must however for the appropriate situations for an individual hearing aid wearer does not necessarily be optimal. By retraining as is commonly used in speaker-related speech recognition systems is, the recognition rate can in certain situations with respect to the individual conditions are increased. This is especially the situation of the performance of own voice is of great importance. Similarly, the classifier to certain noise situations, typical of the acoustic environment of the hearing aid are set optimally will.

In this context, from the publication EP 0681411 A1 a programmable hearing aid which is independently the changing environmental situations adapts. The hearing aid parameters are thereby continuously present at the customized ambient noise, with "fuzzy" inputs from Hearing aid in addition to the measured input signals can be used. The objective is to identify the parameters optimize directly with the listening situation and does not tell.

In the document EP 0814634 A1 discloses a method described by the hearing aid wearer, the hearing aid optimally setting up, by a cause lasstes by the hearing aid wearer Retraining is performed. The hearing aid wearer receives a set of predefined parameter sets for that Listening situation, he notifies the hearing, to choose from. Out this limited amount of parameter sets, each one correspond hearing aids default, it selects those he perceives to be optimal for themselves. The relevant hearing aid settings is learned from a set of rules so that in a similar acoustic input signal the same hearing aid settings takes place. This means that the rules an illustration of the acoustic input variables to the optimum Hearing aid parameter set makes. This retraining the listening situation is taken into account only indirectly by appropriate that only the this listening situation Parameter sets for selection are provided. A direct adaptation of the hearing situation to the acoustic Input data is not, however. A disadvantage is, that the hearing aid wearer in the retraining derartigem Sound of the hearing aid, the parameter set used by the is set, must assess. He has to assess, for example, if he get the sound brighter or darker presented want. For certain complex algorithms and dynamic, adaptive parameters, such as. for example, the control of an adaptive Directional microphone, however, the differentiation is between different sets of parameters for the hearing aid wearer difficult if not impossible.

The object of the present invention is thus, the retraining of a hearing aid for the hearing aid wearer to simplify and operation of the hearing device to correspondingly improve.

This object is achieved by a method for retraining a hearing device by providing a acoustic input signal, providing multiple Hörsituationskennungen and assigning the acoustic input signal one of Hörsituationskennungen by a hearing aid wearer and automatic learning of the assignment of the acoustic Input signal to said one of Hörsituationskennungen.

Further, the invention provides a hearing aid with a Receiving means for receiving an acoustic input signal, memory means for storing a plurality of Hörsituationskennungen and an input means for assigning the acoustic input signal to one of the Hörsituationskennungen by a hearing aid wearer and a learning device for automatically learning the assignment of acoustic Input signal to said one of Hörsituationskennungen by the input means.

Moreover, the above object is achieved by a method for operating a hearing device by receiving a acoustic input signal, automatically assigning a a hearing situation characterizing Hörsituationskennung to the acoustic input signal and automatically adjusting the Hearing device in dependence on the Hörsituationskennung.

The invention is based on the knowledge that for the hearing aid wearer Although the differentiation between different is parameter sets difficult, while naming a acoustic situation that exists just such. as the situation "Own voice" or "in-car entertainment", in the most cases very reliable by the hearing aid wearer is possible. These situations go beyond the conventional in Listening situations used hearing aids as "language in peace" and "language in noise" addition. D. h. The more differentiated Listening situations can personalize for signal processing relevant aspects of this "classical" situations relate. The acoustic representations that this novel, diverse situations underlie able to easily retrained individually by specifically naming will. For example, the sound of your own Voice or the specific sound of your car from the hearing aid be learned, for example, by a neural network. The neural network therefore takes in contrast to the mentioned Prior art according to EP 0813634 A1 an image the acoustic input variables that do not result in the Total adjustment (parameter set) of the hearing aid, but on the internal situation representation (Hörsituationskennung) in front. From this is then based on audiologischem Expert knowledge of the hearing aid parameter set to be used derived or the relevant parameter varies and / or supplemented. In particular, the adaptive algorithms this information further use without the hearing aid wearer must assess the result. This simple assignment the acoustic input signal to predetermined auditory situations is due to the adaptability of the algorithms and the resulting associated temporal dynamics far less difficult for the hearing aid wearer than direct sound evaluation of how For example, the frequency response or the compression ratios / -kniepunkte, according to the prior art.

can In a specific embodiment of the present invention one of listening situations the performance of their own voice Hearing aid wearer correspond, so that after the automatic Learning one's own voice can be recognized. This is in many situations, for example, the directional microphone setting of great importance.

Automatic learning of at least one Hörgeräteeinstellparameters for the associated hearing situation on the Based on the automatic analysis can during (online) or after (offline) of the presentation of the acoustic input signal done. When online retraining needs the acoustic Input signal can not be completely saved, but requires the hearing aid more processing power to the retraining perform. When offline retraining deleted This additional computational requirements in the hearing aid, however, is a Storage device for the acoustic input signal required. The online evaluation avoids the time-consuming reading, Processing and reprogramming of the data or the Hearing aid.

The input device to associate the acoustic input signal to a hearing situation can also start and stop the Nachtrainings be used. Thereby, the Handling of the hearing aid and the implementation of Nachtrainings easier for the hearing aid wearer.

In addition, the input means of a in the Hearing integrated receiver and an external controller consist. The remote control can be used for wired or wireless communication be configured with the hearing aid. It is conceivable, furthermore, that the remote control only is used for retraining the hearing aid. alternative , the remote control as a multifunction device, for example, as a mobile phone or portable computer with wireless interface Fast, be configured.

The input device may also include a programmable computing unit, in particular comprise a PC, so that the operator takes place via a corresponding programming software.

Finally, the input means may, in a particular Embodiment verbally and in particular by means of one or several key words be operable. This operation the hearing aid for the hearing aid wearer more comfortable designed.

Further, the input acoustic signal may be a manually or automatically processed speech signal comprise. Thus, it is possible to very specifically to train the classifier.

In the operation of the hearing aid, ie after the retraining may a currently existing parameter set by the automatic Mapping the current hearing situation to Hörsituationskennung to be influenced. In particular, a parameter the parameter set varies by automatically associating be supplemented and / or. Thereby, it is possible that the acoustic input signal a complex signal processing undergoes based on expert knowledge, when the neural Network a trained hearing situation, z. B. the own Voice recognize. It is the currently used in the hearing aid modified parameter set accordingly or corresponding Filtering performed.

The present invention will now reference to the accompanying explained drawings, in which:<dl tsize="5"><dt>1 shows</dt><dd>a block diagram of the method according to the prior Technology;</dd><dt>FIG 2</dt><dd>a block diagram of the inventive method;</dd><dt>FIG 3</dt><dd>a schematic representation of a hearing aid with Remote control to enter a listening situation in a first step; and</dd><dt>FIG 4</dt><dd>the situation of the hearing aid shown in FIG 3 during Training phase.</dd></dl>

The embodiment described in more detail below represents a preferred embodiment of the present invention For a better understanding of the invention, however, is initially represent. the method for retraining of the prior Technology described again in detail with reference of Fig. 1

The hearing aid wearer or user 1 is located as shown in FIG 1 shown in a specific acoustical situation, in of the hearing aid, an acoustic input signal 2 available stands. Since the hearing aid for the hearing aid wearer 1 subjectively is not set correctly, he takes a retraining in front. For this he classified the sound and tells the hearing aid the corresponding very general hearing situation or Hörsituationskennung, z. B. "language in noise", with. Each these listening situations 3 is each a number of parameter sets assigned. 4 Due to the selected listening situation 3 has the hearing aid wearer 1, for example, seven parameter sets to select. He can now those parameters set 4 Select, with the hearing aid is set to that in this situation the acoustic subjectively best generated sound.

A neural network 5 learns the desired parameter set 4 for the applied acoustic input signal 2, so that it this parameter set 4 acoustic for a similar situation will choose after the training phase.

The subjective assessment of the sounds caused by the different sets of parameters for the hearing aid settings, is for the hearing aid wearer 1, however, very difficult, as this much detailed knowledge about the effects of hearing aid parameters presupposes.

According to the present invention is therefore intended by the hearing device not the use of specific parameter sets, but only trained to recognize the instantaneous situation will. This is done according to the method of Fig. 2 Again receives the hearing aid wearer or user 1, the acoustic Input signal 2 to retrain the neural network 5 in the hearing of the hearing aid wearer must 1 the acoustic situation, in which it is currently located, only one of a plurality of predetermined, specific listening situations 3 'associate. The number of specific listening situations 3 ' in the case of the present invention usually greater as the number of general listening situations 3 according to FIG 1, as they should be differentiated from the outset. Of the The reason for this is that the general listening situation "Language in noise", for example, the specific listening situation "Own voice" includes.

The neural network 5 does not learn thus assigning a Parameter set to the acoustic input signal 2, but assigning a differentiated listening situation or a Hörsituationskennung 3 'to the acoustic input signal 2 (See FIG. Arrows of solid lines in FIG 2). This means, that the neural network in contrast to the prior art learning at a higher level. On the example of the hearing situation "Own voice in your own car" This is explained in more detail. According to the prior art is this complex situation, a fixed set of parameters for example starting from the parameter set group associated with "language in noise". Since only a number of parameter sets to choose from for the Wearers of such situations "language in noise" makes sense, is surely one of the available standing parameter sets for one's own voice and in addition optimized for your own car.

According to the invention on the other hand is the situation "own voice" and the other situation "in your own car" learned separately. This listening situations take each specific impact on the complex signal processing. For example, in the situation "own voice" a specific gain, possibly coupled with a special setting Directivity of the hearing aid, and in the situation "in- Auto "a turn very specific noise reduction caused in the hearing aid.

It is especially advantageous that one's voice from the hearing aid can be learned. This is effected in that the acoustic Input signal with your own voice a special Processing is subjected to and the corresponding parameters for the hearing set specific and the hearing situation "own "Be assigned to vote. The same is true for learning, for example, the hearing situation "own car", whereby a very specific noise reduction can be achieved can. So here, learning not only the input signal assigned a hearing situation, but it is also a very specific determination of parameters, such as Filter or gain parameter instead.

When using the hearing aid after retraining, the neural network 5, an acoustic input signal 2 or a 'Assign more specific Hörsituationskennungen 3, so that the currently valid parameter set 4 '(including Filter parameters) will be affected accordingly. A complex Signal processing unit 6 z. B. with adaptive directional microphone the signal processing on the basis of the influenced Parameter set 4 'perform. If the neural network according to the above example, then the input signal "own Voice track in your own car ", it assigns both the Hörsituationskennung "own voice" and the Hörsituationskennung "In-car", so that the current parameter set For example, as regards the specific amplification for their own voice and with respect to the specific Filtering to suppress the noise of your car is varied or supplemented.

Below are two concrete examples of the present Invention shown:

Example 1: An adaptive directional microphone is directed to the Direction, for example from the maximum desired sound, z. Voice signal is incident. located Maintains the wearers with a herlaufenden next to him person to should the Setting the directional microphone on the discussions partners, ie a maximum gain at an angle of about 90 °. when However, says the hearing aid wearer himself, which comes Nutzschallsignal from our own mouth, ie from an angle 0 °. The own language thus pulls the directional microphone characteristics away from the actual interlocutor namely usually with a certain time lag. If that Hearing aid, however, is trained on one's own voice, and Thus, the adaptive control microphone is known which acoustic properties to include your own voice, can Signals, which are classified as "own voice", for tracking the directivity disregarded stay. In contrast would be the adjustment of the hearing aid according to the prior art of EP 0814634 A1 shown in FIG 1, after which the hearing aid wearer assess multiple parameter sets would, because of the dynamism and adaptability of operations unlikely to be successful. In particular, could own Voice is not recognized.

Example 2: A Störgeräuschunterdrückungsverfahren can specifically trained on a complex, time-variable noise will. This noise is suppressed optimally, although it may have similar spectral components or Modulation spectrum such as language, which further processes as a useful signal to be having. Through individual training this acoustic situation, z. B. the aforementioned Situation "in the car" which can Störgeräuschunterdrückungsverfahren are automatically optimally adjusted, for example by specific weighting factors for individual spectral Belts are adjusted or the dynamic behavior is optimally adapted to the Störgeräuschcharakteristik. Also in this case the differences between the settings dynamic noise canceling only evaluate hard right, the situation however, very reliable.

In certain acoustic situations it may be advantageous, if additional to the retraining of the invention Retraining according to the prior art under assessment different parameter sets is carried out by the hearing aid wearer.

The retraining, as it appears for the hearing aid wearer, is now explained in more detail with reference to FIGS 3 and 4. FIG. Of the Hearing aid wearer wants, for example, the situation "own Voice "one retrain in his hearing aid tenth purpose he concludes via a line 11, a remote control 12 to the hearing aid 10 at. The remote control has a button as a control element 13 on.

In the classifier more listening situations are stored. The hearing aid wearer knows that the hearing situation "own Voice "for example, the situation corresponds to 3. Therefore, he presses the button 13 three times, to notify the classifier, that situation 3 is to be retrained.

In a subsequent step is an acoustic signal, Here's own voice, to accommodate for the hearing aid 10 according to 4 shows presented. The hearing aid wearer has the hearing aid now 10 the beginning and end of the training phase report. This is done by holding down the button 13, while Himself speaks. This means that for both he steps training use only one control element 13 have to be. If many Hörsituationskennungen present, can another design to be user-friendly, z. B. with a Display and a controller (slider, track ball, etc.), with the corresponding situation can be selected quickly can.

The actual retraining of the hearing aid 10, during the performance of the acoustic signal 14 carried. alternative the acoustic signal 14 is recorded in the hearing aid and evaluated according to the recording and the selected listening situation based on characteristic acoustic properties assigned. In the case of online Nachtrainings is a permanent or temporary storage of the acoustic signal 14 is not absolutely necessary.

Since the hearing aid 10, only the information about the current situation must be notified, is in contrast to the state art according to EP 0814634 A1 an external control unit not absolutely necessary. However, it may in accordance with the FIG 3 and 4 are used for example for reasons of comfort. However, there is also a record button on the hearing aid itself to be appropriate.

By retraining can the detection rate of Classifier for certain situations significantly over increase the default so that the hearing aid more reliable is adjusted to this situation. By self Start and stop the Nachtrainingsphase by the Wearers may encounter situations further reliably retrained be because the hearing aid wearer decides, when the signal of the situation can be assigned.