Patent Application
20090238387
This application claims priority of German application No. 10 2008 015 264.1 DE filed Mar. 20, 2008, which is incorporated by reference herein in its entirety.
The present invention relates to a method for actively reducing occlusion in a hearing apparatus. In this case a sound in an auditory canal is picked up by a microphone while outputting a corresponding microphone signal and the picked-up microphone signal is filtered using an adaptive filter. The filtered microphone signal is fed back to an input of a receiver which is used to emit the sound in the auditory canal. At least part of a transducer transmission function, which is defined for the transmission path from the input of the receiver via the auditory canal to the output of the microphone, is measured and the adaptive filter is adjusted as a function thereof. Here a hearing apparatus is taken to mean any sound-emitting apparatus that can be worn in or on the ear, for example a hearing aid, a headset, headphones and the like.
Hearing aids are portable hearing apparatuses which are used to cater for the hard of hearing. Different hearing aid designs such as behind-the-ear hearing aids (BTE), hearing aid with external receiver (RIC: receiver in the canal) and in-the-ear hearing aids (ITE), for example also concha hearing aids or canal hearing aids (ITE, CIC), are provided in order to accommodate the numerous individual needs. The hearing aids listed by way of example are worn on the outer ear or in the auditory canal. However there are also bone conduction hearing instruments, implantable or vibrotactile hearing aids available on the market. The damaged hearing is stimulated either mechanically or electrically in this case.
As fundamental components hearing aids basically comprise an input transducer, an amplifier and an output transducer. The input transducer is usually a sound pick-up, for example a microphone, and/or an electromagnetic receiver, for example an induction coil. The output transducer is usually implemented as an electroacoustic transducer, for example miniature loudspeaker, or as an electromechanical transducer, for example bone conduction receiver. The amplifier is conventionally integrated in a signal processing unit.
An unpleasant effect when wearing a hearing aid is that the wearer's own voice sounds unnatural. This is due to the fact that the wearer's own voice is conducted into the auditory canal via bone conduction and causes a certain sound pressure there, in particular at lower frequencies. If the auditory canal is open the corresponding compression waves can be conducted to the outside. If the auditory canal is closed by the hearing aid however, then a higher sound pressure builds up here which is called the occlusion effect and since it is unnatural is deemed unpleasant.
A generic method for actively reducing occlusion in hearing aids is known from document WO 2004/021740 A1 and document WO 2006/037156 A1. The first document describes the transducer transmission function from the input of the receiver via the auditory canal to the output of the auditory canal microphone in more detail. It can be determined very accurately in situ using the hearing aid as a measuring instrument. The transducer transmission function is complex, i.e. a function of absolute value and phase over frequency.
In an incompletely adaptive implementation of actively reducing occlusion the measured transducer transmission function is used to determine in a computer the optimum configuration of digital signal processing for actively reducing occlusion. This optimization process could basically also take place completely automatically. However there is the problem here of the algorithm being irreversibly incorrectly changed in certain situations, or of a great deal of computing time being required. Manual intervention is necessary or helpful in these situations.
The object of the present invention therefore consists in further automating an adaptive implementation of actively reducing occlusion.
According to the invention this object is achieved by a method for actively reducing occlusion in a hearing apparatus by picking up a sound in an auditory canal via a microphone while outputting a corresponding microphone signal, filtering the microphone signal using an adjustable filter, feeding back the filtered microphone signal to an input of a receiver which is used to emit sound in the auditory canal, measuring at least part of a transducer transmission function which is defined for the transmission path from the input of the receiver via the auditory canal to the output of the microphone, and adjusting the adjustable filter subject to the transducer transmission function, the transducer transmission function being subjected to an automatic plausibility check, and the adjustable filter only being altered if the transducer transmission function is plausible according to a predefined criterion. The term “adjustable” does not exclude part of the filter from being adaptive, i.e. being capable of automatic adjustment by way of an adaptation rule.
Also provided according to the invention is a hearing apparatus with active occlusion reduction, comprising a receiver for emitting sound in an auditory canal, a microphone for picking up a sound in the auditory canal and for outputting a corresponding microphone signal, an adjustable filter for filtering the microphone signal, the filtered microphone signal being fed back to the input of the receiver, a measuring device for measuring at least part of a transducer transmission function which is defined for the transmission path from the input of the receiver via the auditory canal to the output of the microphone, and an adjusting mechanism for adjusting the adjustable filter subject to the transducer transmission function, and a checking device for automatically checking the plausibility of the transducer transmission function, wherein the adjustable filter can be altered by the adjusting mechanism if the transducer transmission function is plausible according to a predefined criterion.
Advantageously, according to the present invention a check is made before adaptation of the occlusion reduction algorithm as to whether the measured transducer transmission function is plausible. It can thus be ensured that the algorithm, which determines optimum configuration, does not enter an exceptional state from which it does not emerge again, or only emerges with difficulty. This prevents an excessive solution space for the algorithm or an excessive amount of unnecessary computing time.
The measured part of the transducer transmission function is preferably smoothed for the plausibility check. Certain measurement uncertainties can be compensated thereby.
It may also be advantageous if the transducer transmission function is measured in a first frequency range and is extrapolated with the aid of a model in a second frequency range on the basis of the measured data. A reliable measurable range for example can therefore be used in order to be able to estimate a range that is less reliably measurable for the transducer function or the plausibility check.
According to one variant the transducer transmission function can be deemed implausible if its absolute value in a predefined frequency range is lower than a predefined threshold. Clogging of the hearing apparatus with cerumen can be detected thereby for example.
The transducer transmission function can also be deemed implausible if its phase in a predefined frequency range is below a predefined minimum phase. It can thereby also be checked for example whether one of the components involved is defective or whether the measuring signal was too quiet.
The transducer transmission function can also be deemed implausible if its value, including absolute value and phase, is outside of a predefined tolerance enclosure in the space which is defined by the coordinates absolute value, phase and frequency. Using this kind of tolerance enclosure it can be detected whether the hearing apparatus is working correctly in a certain context. However, the tolerance enclosure can also be used to keep the computing time for changing the algorithm within certain limits. If, for example, the transducer transmission function is not in a very narrow tolerance enclosure, then a small change in the location of the hearing apparatus in the ear can quickly bring about the desired outcome and a longer computing time can be avoided by a change in the algorithm.
The present invention will be described in more detail with reference to the accompanying drawings, in which:
The following exemplary embodiments described in more detail are preferred embodiments of the present invention.
The ITE hearing aid 10 has an outwardly directed microphone 14 to pick up ambient sound (cf. microphone 2 in
However, the case can occur where the system per se is not in order or cannot work properly. This is the case for example if the receiver 16 or auditory canal microphone 17 has failed or the sound output of the receiver 16 and/or the sound output of the auditory canal microphone 17 is blocked by cerumen. In such cases the transducer transmission function is not plausible. These cases can be detected using a plausibility check. It proceeds according to the principle of the block diagram in
Extrapolation of the transducer transmission function can be described using the example of a first order high-pass filter according to
If the transmission path 19 from the input of the receiver 16 to the output of the auditory canal microphone 17 is blocked, the absolute value of the transmission function RVM is low for a certain frequency range. In view of the plausibility it is therefore necessary to extract the amplitude of the transmission function or evaluate its absolute value according to step S3. In this way cases can be detected in which the receiver 16 or the auditory canal microphone 17 are defective. However, the acoustic transmission path from receiver to microphone may also be blocked, for example as a result of clogging with cerumen.
Furthermore, the phase of the transducer transmission function can be extracted according to step S4. With regard to the plausibility check it is known that at low frequencies in the range of 100 Hz the phase cannot assume any desired values. In particular there is a minimum phase at low frequencies due to a series of high-pass filters (receiver, microphone, analog microphone amplifier). The typical value of the minimum phase can be given as a function of the transducer. If a lower measured phase is the minimum phase, the measuring result does not itself have to be in order. For example the measuring signal could have been too quiet if the S/N ratio was temporarily too low. In this case the measurement must be repeated with a louder measuring signal to obtain a valid measuring result.
Sometimes the measured transducer transmission function can also be evaluated directly according to step S1 or a transmission function processed according to step S2, and this is indicated by the arrow S5 in
The data obtained from steps S3 to S6 may for example accordingly be compared with certain thresholds according to step S7 or evaluated using specific criteria. Therefore, as mentioned, the phase can for example be compared with a minimum phase. The absolute value of the transmission function should not lie below a minimum absolute value for a higher frequency range either. The standardized measured or extrapolated transmission function, which represents a spatial curve in the amplitude-phase frequency space, can for example be compared with a tolerance enclosure around this curve. If the tolerance enclosure is never left the measured transmission function is accepted as valid or plausible. A decision is therefore made in step S8 on the basis of the comparison in step S7 as to whether the transmission function is valid or invalid or plausible or implausible. Only after plausibility has been decided is the reduction in occlusion optimized by adapting the loop filter S.
The advantage of this procedure is that only expedient transducer transmission functions are used for determining the optimum configuration of signal processing (in particular of the loop filter). The optimization algorithm is therefore protected from converging into a disadvantageous state as a result of an inexpedient transducer transmission function. Overall this limits the solution space for the algorithm and reduces the computing time thereby. An indication of the cause of the fault can also be given in the case of certain properties of the measured transmission function. Thus for example a leak can be indicated if the cut-off frequency of the high-pass-type transmission function is relatively high.
| Number | Date | Country | Kind |
|---|---|---|---|
| 10 2008 015 264.1 | Mar 2008 | DE | national |
