Patent Application
20190075403
This application claims the priority, under 35 U.S.C. § 119, of German patent application DE 10 2017 215 825.5, filed Sep. 7, 2017; the prior application is herewith incorporated by reference in its entirety.
The invention relates to a method for detecting a defect in a hearing instrument that has at least a first input transducer and at least one output transducer.
In a hearing device, sound signals from the environment are converted into electrical signals by one or more input transducers, and these signals are further processed by a signal processor or the like. They are then converted back into an output sound signal by an output transducer. The output sound signal is fed to the ear of a user, who usually has a hearing impairment. In this way, the electrical signals in the signal processor are processed, as much as possible, so as to compensate for this impairment through corresponding processing.
For this purpose, as far as possible, error-free functioning of the electroacoustic hardware components, i.e. the input transducer and the output transducer, is particularly necessary. These components in hearing devices typically lose aspects of their performance with increasing operating time, i.e., at comparable sound pressures the input transducers will produce electrical signals of increasingly lower amplitudes, while the output transducer over time generates an increasingly lower sound pressure from a normalized test signal. This loss of performance capacity, which is primarily due to wear of the electroacoustic components, is aggravated by the fact that the components in the hearing device are exposed to the influences of moisture or sebum when worn in the ear. Malfunction of the hearing device is therefore often caused by a corresponding damage or impairment of one of the electroacoustic hardware components.
A total failure of one of these components—i.e. of one of the input transducers or the output transducer—is easy for the user of the hearing device to recognize. A merely gradual decrease in performance, however, as occur for example through attenuation or underperformance in a particular frequency range, is often quite difficult for the user, or a hearing aid acoustician, to recognize without a specific measurement. This results in a long-term operation of the hearing device with a correction for the user's hearing impairment that is not adequate for the user's hearing impairment; this also may affect the user's engagement in the world and ability to concentrate, due to the consequently reduced intelligibility of speech.
Such problems with electroacoustic hardware components may also occur in other hearing instruments such as mobile telephones. Here too, a defect in an input transducer is difficult for the user to recognize, because the user may not even be able to check the input signal generated from the user's own speech, and thus may have to rely on statements from the people the user is speaking to. Also, a wideband attenuation in the output transducer is difficult for the user to recognize, especially because of mobile telephone users' tendency to attribute shortcomings in the output sound signal primarily to inadequate signal transmission through the mobile network. Moreover, even when worn on the body, e.g. in a trouser or jacket pocket, mobile telephones are potentially exposed to influences such as moisture and impacts that may impair the electroacoustic components.
Detecting possible deterioration of the device's proper operation over a longer period of operation is thus a general problem for hearing instruments that have electroacoustic components.
The object of the invention, accordingly, is to provide a method for detecting a defect in a hearing instrument, the method being as simple as possible to carry out with high reliability and requiring no additional conditions of the hearing instrument in order to be carried out; and in particular, requiring no additional devices.
With the above and other objects in view there is provided, in accordance with the invention, a method of detecting a defect in a hearing instrument having at least one first input transducer and at least one output transducer. The novel method comprises:
In other words, the invention provides for a method for detecting a defect in a hearing instrument, wherein a first transfer function of a first acoustic system, comprising the output transducer and the first input transducer, is determined, and at least a first reference function is determined for the first transfer function. The first transfer function of the first acoustic system is compared with the first reference function, and a defect in the hearing instrument is detected based on this comparison.
The term “hearing instrument,” as used herein, generally refers to any device in which a sound signal of the environment is converted by an electroacoustic input transducer to an internal electrical signal, and an output sound signal is generated from an electrical output signal of the device by an electroacoustic output transducer, i.e., in particular a hearing device and a mobile telephone.
Preferably, in this case, the hearing instrument also has a signal processing unit, and during operation the first input transducer generates a first input signal from a sound signal of the environment, this input signal is supplied to the signal processing unit, the signal processing unit emits an output signal, and the output transducer converts this output signal into an output sound signal. The output signal in this case may be based on the input signal, as is the case in a hearing device, or it may be based on a signal received via an antenna, as is the case in a mobile telephone. In the latter case, the signal processing unit may in particular be set up to prepare the input signal for transmission via a transmitting antenna—for example by coding it in a transmission protocol—and to decode a signal received at a receiving antenna and convert it into an output signal.
The determination of the first reference function may be carried out, in particular, before determining the current first transfer function. In this case, the first reference function may in particular also be “trivial,” in other words, given by a frequency-independent limit value for the first transfer function or for the magnitude of the first transfer function. Preferably, however, the reference function is non-trivial, and thus frequency-dependent.
By determining a transfer function for an acoustic system comprising the first input transducer and the output transducer, advantageous information is provided, in particular for the purpose of detecting defects in these components. As a result of using the transfer function, this information is also available in frequency-resolved form, which simplifies analysis with regard to a defect. The determination of the first transfer function preferably takes place without using an external sound generator to stimulate or inspect the first input transducer or using an additional external microphone to inspect the output transducer. This may be achieved by a suitable selection of the first acoustic system.
In this case, the first reference function should be determined in such a way that it may serve as a reference for the first transfer function when the hearing instrument is fully functional, i.e. free of defects. By comparing the first transfer function with the first reference function, for example, those frequency ranges in which the functionality of the hearing instrument is impaired may be identified. To more precisely localize the defect, the first transfer function and first reference function may now be examined, particularly in the frequency domain and time domain. This provides additional information content and may allow conclusions to be drawn as to exactly which component a defect is present in, i.e. whether the defect is present at the first input transducer or the output transducer. A defect of the output transducer may result in an impulse response of the first transfer function which is considerably weakened compared to the values of the first reference function, while a defect of the input transducer may, among other things, have a impulse response of the first transfer function that is time-shifted relative to the values of the first reference function.
Conveniently, the open loop transfer function is determined as the first transfer function of the first acoustic system, the open signal loop being formed from the output transducer, an acoustic feedback path from the output transducer to the first input transducer, and the first input transducer. The open loop transfer function may be determined in a particularly simple manner, for example by means of a suitable test signal, which is converted by the output transducer into a test sound signal, and by an analysis of the signal component of the test signal in a first input signal generated by the first input transducer, to estimate on this basis the portion of the test sound signal arriving at the first input transducer. Another advantage of using the open signal loop as the first acoustic system, and thus using the open loop transfer function as the first transfer function, is that the first input transducer and the output transducer are completely within that system, so that there is no need for any additional sound generators or any additional measuring apparatus.
In this case, preferably, an additional closed loop transfer function is determined, and from this, the open loop transfer function is determined as the first transfer function, wherein the closed signal loop is formed from the output transducer, an acoustic feedback path from the output transducer to the first input transducer, the first input transducer, and a signal processing path from the first input transducer to the output transducer. The closed signal loop is thus formed by the open signal loop, which is closed from the input transducer to the output transducer by the signal processing path. This is advantageous, particularly in a hearing instrument designed as a hearing device, because a closed loop transfer function is often determined in the context of suppressing acoustic feedback anyway, and thus there is no need for any additional measurements or functionality.
Preferably, the closed loop transfer function is determined by an adaptive filter, wherein the open signal loop is determined based on the closed signal loop, taking into account a signal processing that takes place along the signal processing path. This may be achieved in particular by correcting the closed loop transfer function, which has been determined by the adaptive filter, by a corresponding transfer function of the internal signal processing processes that take place along the signal processing path of the hearing instrument, because these processes are presumed to be completely known.
Advantageously, in this case, the adaptive filter is used in the hearing instrument for suppressing acoustic feedback via the acoustic feedback path running from the output transducer to the first input transducer. This means, in particular, that the adaptive filter is furnished and set up for feedback suppression as needed during normal use of the hearing instrument, and that the adaptive filter may be used in the context of detecting a defect in the hearing instrument by accessing the closed loop transfer function that was determined for the purpose of feedback suppression. Optionally, the adaptive filter may also be operated in a dedicated mode for detecting a hearing instrument defect.
Alternatively, a test signal is supplied to the output transducer, a test sound signal is generated from the test signal by the output transducer, a first input signal is generated by the first input transducer from an input sound comprising the test sound signal, and the open loop transfer function is determined as a first transfer function from the input signal and the test signal. This means that the open loop transfer function is determined by direct measurement. In particular, in this case the spectral power density of the test signal is constant over the frequency, so the test signal is “white noise”. A direct measurement of the open loop transfer function may thus be realized with particular ease. This also applies to the case in which the hearing instrument is provided via a mobile telephone, because for this purpose the loudspeaker only needs to generate the test sound signal, and only the component of the test sound signal that reaches the microphone needs to be measured there.
In particular, the determination of the first transfer function takes place at predetermined intervals, i.e. either regularly or based on the respective duration of the operating phases. The first transfer function may also be determined via user input. In particular, in this case, the user input may activate the complete method for detecting a defect, for example if the user subjectively perceives that there is a malfunction in the hearing instrument and wants to obtain objective clarity on that point. Also, the complete method for detecting a defect may be performed regularly or based on the respective duration of the operating phases, for example, as part of a maintenance program or the like.
In an advantageous configuration, a cross-correlation is used for comparing the first transfer function with the first reference function. The cross-correlation, in this case, may be taken in particular from the first transfer function and first reference function in the frequency domain and/or from the first transfer function and the first reference function in the time domain, in which the impulse response of the first acoustic system is specified. The cross-correlation is used in particular as an additional criterion for monitoring deviations of the first transfer function with respect to the first reference function. In particular, the corresponding correlation coefficient may be used. This has the advantage that, in the case of a frequency-band-wise deviation between the first transfer function and the first reference function, the degree of deviation is difficult to quantify and in particular is more difficult to put in relation to other scenarios. To this end, the correlation coefficient provides a single value that affords such comparability.
Expediently, the first reference function is determined from a measurement of the first transfer function under normalized conditions. In particular, for a hearing device, this determination may take place at a hearing aid acoustician. Such a measurement is particularly easy to implement as part of a fitting session that is taking place anyway. In the case of a mobile telephone, such a measurement may be taken at the manufacturer or at a qualified distributor.
Alternatively, the first reference function may be determined by time-averaging multiple values of the first transfer function at different times. The values may be determined at multiple times in particular by a routine detection of the values during a predetermined operating interval after initial operation, e.g. in the first days. This is based on the assumption that the hearing instrument is still fully functional at the start of operation, and therefore the initially detected values of the first transfer function are a suitable basis for the first reference function, and that averaging over a plurality of values is advantageous for a true reference, irrespective of the respective conditions at the time at which the respective value has been determined. This procedure is particularly advantageous if the first transfer function cannot be directly measured under normalized conditions—for example, if a fitting session at a hearing aid acoustician is not contemplated when putting a hearing device into operation.
Advantageously, the first transfer function is determined by time-averaging a plurality of values of the open loop transfer function. In this way, it is possible to compensate for short-term fluctuations. In this case, the time averaging preferably comprises those values that reflect the current status of the hearing instrument as accurately as possible, which may be achieved in particular by a significant weighting of the most recent values. The determination of the values of the open loop transfer function, in this case, may take place in the background over a longer period of time, and the determination of the first transfer function from these values may then take place over a decreasing weighting of the values during averaging.
Preferably, a defect of the first input transducer and/or the output transducer is detected. The method described is particularly suitable for detecting defects in these components.
Conveniently, a measure is determined for a correlation between the first transfer function and the first reference function, wherein the defect is detected based on the measure of correlation. A cross-correlation may for example be used as a measure of correlation.
Alternatively or additionally, a first polynomial, which approximates the first transfer function, and a first reference polynomial, which approximates the first reference function, may be determined, the defect being recognized with reference to the first polynomial and the first reference polynomial based on a coefficient comparison. In this case, for example, a threshold value may be predetermined for the deviation of the polynomial coefficients from each other, above which it is concluded that there is a defect in the hearing instrument. The threshold value may be selected differently for each of the respectively different orders of polynomial coefficients. In particular, as a criterion for a defect in the hearing instrument, in addition to the aforementioned coefficient comparison, the aforementioned measure of the correlation of these transfer functions may also be used.
It is also advantageous if when a second transfer function of a second acoustic system comprising the output transducer and a second input transducer of the hearing instrument is determined, at least a second reference function is determined for the second transfer function, the second transfer function is compared with the second reference function, and a defect in the hearing instrument is detected based on the comparison of the first transfer function with the first reference function and of the second transfer function with the second reference function. This is advantageous for hearing instruments that have a second input transducer, such as for example certain embodiments of hearing devices.
In particular, a comparison of the first transfer function with the second transfer function is additionally used for detecting a defect in the hearing instrument. In addition, this comparison also makes it easier to localize the defect. In rough terms, there are at least three possibilities for a defect in electroacoustic hardware: the two input transducers and the output transducer. The aforementioned comparisons of the transfer function with the corresponding reference function relate respectively either to an input transducer and the output transducer, or to both input transducers, because the contribution of the output transducer may be eliminated when comparing the first and second transfer functions, for example by simple subtraction.
In particular, the first and second transfer functions may be compared with the respectively associated first or second reference function, and also with each other, on the basis of a measure for the correlation of the transfer functions and/or reference functions. Alternatively or additionally, two transfer and/or reference functions to be compared may each respectively be approximated by polynomials, and a comparison of the relevant polynomial coefficients may be used to compare the aforementioned functions.
The second reference function may be determined in particular before determining the current second transfer function. In this case, the second reference function may in particular also be “trivial,” that is to say, it may be given by a frequency-independent limit value for the second transfer function or the magnitude of the second transfer function. Preferably, however, the reference function is non-trivial, and thus frequency-dependent.
Expediently, in this case, a first limit value, a second limit value and a third limit value are predetermined, a first difference being taken from the first transfer function and the first reference function, a second difference being taken from the second transfer function and the second reference function, and a third difference being taken from the first transfer function and the second transfer function. A defect in the first input transducer is detected when the first difference exceeds the first limit value in at least one frequency range but the second difference does not exceed the second limit value, and/or a defect in the output transducer is detected when there are respectively different frequency ranges for the first difference and the second difference, in which these exceed the first limit value or the second limit value but the third difference does not exceed the third limit value. In particular, in this case, the first limit value and the second limit value are identical. This embodiment is particularly easy to implement due to the low complexity of the computational operations used.
The invention also describes a hearing instrument with at least a first input transducer and an output transducer, which is set up to carry out the method described above. The advantages stated for the method and the developments thereof apply analogously to the hearing instrument. Preferably, the hearing instrument for carrying out the method comprises a control unit that has been set up correspondingly. This unit may for example also be implemented in a signal processing unit of the hearing instrument by means of corresponding command blocks.
In a particularly advantageous configuration, the hearing instrument is designed as a hearing device. Especially for the input and output transducers used in hearing devices, and in view of possible environmental influences to which a hearing device and its components are exposed during operation, this method is particularly practical for detecting a defect without the need for a costly measurement at a hearing aid acoustician.
Other features which are considered as characteristic for the invention are set forth in the appended claims.
Although the invention is illustrated and described herein as embodied in a method of detecting a defect in a hearing instrument, it is nevertheless not intended to be limited to the details shown, since various modifications and structural changes may be made therein without departing from the spirit of the invention and within the scope and range of equivalents of the claims.
The construction and method of operation of the invention, however, together with additional objects and advantages thereof will be best understood from the following description of specific embodiments when read in connection with the accompanying drawings.
Corresponding parts and sizes are assigned the same reference numerals in all drawing figures.
Referring now to the figures of the drawing in detail and first, particularly, to
To detect a defect at the first input transducer 4, second input transducer 6 or output transducer 8, when the hearing device 2 is operating, the signal processing unit 14 outputs a test signal 18 as the output signal 16, and this signal is converted into a test sound signal 20 by the output transducer 8. In the present case, the test sound signal 20 is substantially white noise; in other words, it has a substantially flat frequency spectrum. But other types of signals are also conceivable here, such as sine tones of different frequencies, chirps, “perfect sweeps” or the like, which allow determinations about as broad a frequency spectrum as possible.
The first input transducer 4 and second input transducer 6 now respectively convert the corresponding sound signals into input signals 10 and 12, and thus also convert the component of the test sound signal 20 arriving at the respective input transducers 4, 6 via the corresponding acoustic feedback path 22 or 24 that runs from the output transducer 8 to the input transducer 4, 6.
With respect to the first input signal 10 and the output signal 8, a first transfer function T1 is determined for a first acoustic system 26 that is formed by the open signal loop from the output transducer 8 via the acoustic feedback path 22 to the first input transducer 4. This may be done by directly measuring the component of the test signal 18 in the first input signal 4, or it may be done via an estimate based on the closed signal loop formed from the first acoustic system 26, i.e. the open signal loop, and the signal processing unit 14. The closed signal loop or the transfer function thereof is often already available in hearing devices because it has already been determined for the purpose of suppressing acoustic feedback via the acoustic feedback path 22.
In addition, a second transfer function T2 is determined based on the second input signal 12 and the output signal 8 for a second acoustic system 28 that is formed by the open signal loop that runs from the output transducer 8 via the acoustic feedback path 24 to the second input transducer 6.
A first reference function and a second reference function are now respectively stored for the first transfer function T1 and the second transfer function T2. This may take place by means of measurements of the first transfer function T1 and the second transfer function T2 under normalized conditions at a hearing aid acoustician, or alternatively by time-averaging the respective values of the first transfer function T1 or T2 during the first days after the device is put into operation, because it may be presumed that at this time, the hardware components to be inspected are still fully functional.
The respectively currently determined first or second transfer function T1, T2 is now compared with the corresponding reference functions in order to be able to conclude from this that there is a possible defect of the hardware components. This will be explained with reference to
In
It may be concluded in this case, that the first acoustic system 26, consisting of the output transducer 8, the corresponding acoustic feedback path 22 and the first input transducer 4, operates largely interference-free; however, a significant defect must be present in the second acoustic system 28, which is formed from the output transducer 8, the acoustic feedback path 24 and the second input transducer 6. The defect is thus attributable to the second input transducer 6.
The first transfer function T1 falling below the negative first limit value −g1 in
In the situation illustrated in
The difference between the first transfer function T1 and the second transfer function T2 essentially reproduces the differences between the two acoustic feedback paths 22, 24 from the output transducer 8 to the first and second input transducers 4 and 6, and the differences between the two input transducers 4, 6. In addition, the differences in the acoustic feedback paths 22, 24 may be neglected, at least with respect to the contributions of the output transducer 8 in the first and second transfer functions, due to the considerable deviation from the respective reference function R1 or R2. This means that, in the present case, it may be concluded from the difference T1-T2 between the two transfer functions, which is relatively small compared to the deviations of the two transfer functions from the respective reference function T1-R1 or T2-R2, that the two input transducers 4, 6 are largely trouble-free, and thus the defect is in the output transducer 8.
Another way to inspect the open loop transfer function from the output transducer 8 via the respective acoustic feedback path 22 and 24 to the corresponding input transducer 4 and 6 with regard to defective hardware uses the cross-correlation of the respective transfer function T1 or T2 with the corresponding reference function R1 or R2 in the frequency domain and in the time domain.
This is illustrated by
The case illustrated in
The invention has been illustrated and described in detail by means of the preferred exemplary embodiment, but this embodiment does not limit the invention. Other variations may be deduced from this embodiment by a person of ordinary skill in the art, without departing from the protected scope of the invention.
The following is a summary list of reference numerals and the corresponding structure used in the above description of the invention:
| Number | Date | Country | Kind |
|---|---|---|---|
| 10 2017 215 825.5 | Sep 2017 | DE | national |
