KR102596749B1 - Magnitude and phase correction of hearing devices - Google Patents

Abstract

A method of correcting magnitude and phase distortion in an open ear hearing device includes determining the effect of insertion of the hearing device 12 substantially on the eardrum 11 when in the ear. The magnitude and phase response of the complex insertion transfer function (ITF) is corrected when the transfer function to the eardrum substantially matches the transfer function without an appropriately positioned hearing device.

Description

Magnitude and phase correction of hearing devices

The present invention generally relates to hearing devices worn by a person to improve the person's ability to hear sounds. In this specification, it will sometimes be referred to as a “hearing aid.” However, such designation is not intended to limit the invention to use by persons with hearing loss. The present invention can also be used by people without hearing impairment.

The invention more particularly relates to a hearing device in which at least a portion of the device occludes the ear canal and creates undesirable insertion effects. The invention is particularly applicable to open-ear hearing devices, but can also be used with closed-ear hearing devices.

Inserting all or part of a hearing aid into the ear distorts both the loudness and phase of the sound reaching the ear drum. Ideally, the hearing device will compensate for this effect so that the arriving sound is not distorted after passing through the hearing device and the ear canal. Many hearing aids compensate for size effects but do not adequately resolve phase distortion. As a result, users often complain that the sound is unnatural and lacks the directional cues that are important to the listening experience. These complaints are especially prevalent among musicians and professionals in the music industry who require hearing aids to compensate for partial loss of hearing, even though their ears are trained to distinguish subtle differences.

One proposed solution to compensate for the effects of hearing aid insertion is described in US Pat. No. 5,325,436 to Sigfrid Soli et al. Soli's patent discloses a method for determining digital filters that compensate for the insertion effects of ear-mounted hearing aids. In Soli, the loudness and phase response at the ear is measured both without a hearing aid and with the hearing aid in place. The required equalization (EQ) is then calculated. This way, Soli makes assumptions about phase components that are invalid in most cases. The method described in Soli is complex, requires EQ calculations, and may not be effective depending on the assumptions made about phase. Soli presupposes an ear piece that completely hides the ear canal and attenuates all external sounds. Additionally, the corrections described in Soli are intended to preserve absolute timing differences as well as ear-to-ear differences in time intervals: because of this, Soli requires a two-piece fitting of the hearing aid.

The present invention provides a device and method for correcting the implantation effects of a hearing device in the ear, which device does not require any assumptions about the phase response, can be used with monaural fitting, and can be used with an open ear. Suitable for inserts. The present invention is particularly effective in correcting phase distortions and anomalies of sound in the eardrum caused by the presence of the hearing device in the external auditory canal. The devices and methods of the present invention can provide the eardrum with amplified sound that is perceived naturally and retains directional cues for an improved listening experience. In other words, the device is perceived as acoustically transparent. Improvements to the listening experience will be realized by most users, but especially by music industry professionals seeking to regain their ability to discern subtle musical differences.

The present invention relates to a method and device for correcting magnitude and phase distortion in a hearing device, at least a portion of which is inserted into the ear when worn by a user. The method includes determining the effectiveness of insertion of the hearing device when in the user's ear. The insertion effect is characterized by a complex insertion transfer function (ITF) with magnitude and phase response and is determined at the eardrum. When the transfer function to the eardrum matches the transfer function without the hearing device at the appropriate location, both the magnitude and phase responses of the ITF are corrected.

Preferably, the insertion effect is corrected by at least one and suitably a plurality of second order minimum phase filters. The second order minimum phase filter is preferably an infinite impulse response (IIR) filter, more preferably a biquad filter.

Correcting for insertion effects in both magnitude and phase involves determining appropriate equalization, which can be approximated by taking the ratio of the composite Head Related Transfer Function (HRTF) and the composite Insertion Transfer Function (ITF). It can be determined as such, but it cannot be determined entirely. The composite HRTF and ITF may be determined by measurements on a manikin with or without a hearing device, or by measurements directly on a hearing device user. The phase response is corrected only when the phase response is at the minimum phase.

Once the magnitude and phase response of the hearing device are known, equalization to correct for ITF can be calculated for all parts of the transfer function that are of minimum phase. However, in most cases this is not possible because there are no analysis methods that can handle non-minimum phase regions.

More practically, the desired equalization may be determined through an iterative process. Different minimum phase filtering can be introduced in the hearing device to correct spectral regions where minimum phase phenomena dominate: in other regions where phase correction is not possible, it is possible to correct the magnitude response. This is performed iteratively until the desired phase correction is achieved.

Alternatively, the desired equalization for correcting ITF magnitude and phase response may be determined subjectively by a user experienced in describing sounds. Users compare their perception of the sounds they hear regardless of whether they have a hearing device in their ear canal. The desired equalization is achieved when the user indicates that there is no perceived difference between the two conditions.

According to the best mode of the invention, the hearing device is configured such that the latency of sound amplified by the hearing device corresponds to less than about 120 degrees of phase at all frequencies amplified by the hearing device. That is, the latency of the hearing device is preferably less than about one third of the period of the highest frequency produced by the hearing device. For example, if the device amplifies sound up to 10 kHz, the desired latency would be less than 30 μs.

Figure 1 is a schematic representation of an open-ear hearing aid worn on the ear, creating an insertion effect and showing two sound paths to the eardrum.
Figure 2 is a graph showing the insertion effect of an open ear hearing device where the head-related transfer function (HRTF) and insertion transfer function (ITF) were measured on an acoustic manikin (magnitude response is shown in the top graph). The phase response is shown in the lower graph.)
Figure 3 is a graph showing how ITF can be compensated with a second order minimum phase filter according to the present invention. The HRTF is the same as in Figure 2, but the auxiliary transfer function (ATF) is the direct sound and the result of the sound amplified and equalized by the hearing device. (The magnitude response is shown on the top plot, and the phase response is shown on the bottom.) HRTF is the head-related transfer function and ATF is the auxiliary transfer function.
Figures 4A and 4B are graphs that, similar to Figure 3, mathematically illustrate how a minimum phase filter can fully compensate for attenuation (magnitude response is shown above; phase response is shown below). The filters are shown individually in Figure 4A and together in Figure 4B.
Figures 5A and 5B are graphs that mathematically illustrate how a 1.5 ms delay makes it impossible for a bandpass filter to compensate for attenuation in either magnitude or phase. Again, the filters are shown individually in Figure 5A and summed together in Figure 5B.
Figure 6 is a generalized flow diagram showing two basic steps according to the invention for correcting the effects of insertion of a hearing device in the external auditory canal.
Figure 7 is a more detailed flow diagram illustrating the steps for compensating for the effects of insertion of a hearing device in the external auditory canal using an acoustic manikin.

When an auditory device is present within the ear canal, the transfer function to the eardrum changes. This change consists of two components: the active response of the device itself and its passive acoustic effects. Once passive effects are compensated, the hearing device becomes truly transparent, sounding natural to the user at all sound levels.

In the case of open hearing aids, incident sound is not completely attenuated by the presence of the receiver in the ear canal: this is true because a direct path around the receiver (or speaker) is provided by a hole in the rubber insert tip that holds the receiver in the proper position. . These devices provide little attenuation of low frequencies (below 500 Hz), but tend to attenuate high frequencies in a variable manner depending on the geometry of the hearing aid, ear tip, and user's ear canal.

These open assistive devices have two advantages for the user: firstly, for people with high-frequency hearing loss (the most common type), the hearing aid does not need to amplify low-frequency sounds at all, and there are no physical constraints on the small speakers used; Less than this. Second, when the entrance to the ear canal is blocked, there is no occlusion effect, which is a change in the perception of one's own voice.

For closed-back hearing aids, incident sound is attenuated at all frequencies and can generally be ignored. This means that the sound produced by the hearing aid is the only significant sound that reaches the eardrum. However, interpolation effects persist in both amplitude and phase and need to be corrected in the same way as described here.

For hearing devices without microphones, such as in-ear monitors, the input signal is now an electrical signal. The effect of insertion of such a device is the same as in the previous case and can be judged from the case where the sound is played through a speaker in front of the wearer.

The method of the present invention is first described for the case of open-ear hearing aids, where an acoustic manikin is used to make the measurements necessary to determine the equalization necessary to effectively correct for the implantation effects of the hearing aid. Instead of using a dummy, we explain how to later rely on a living person instead of using a dummy. The other two cases mentioned above, closed hearing aids and in-ear monitors, are virtually identical and can be calibrated using the same methods described here.

The acoustic manikin includes microphones in prosthetic ears designed and adjusted to emulate the average human head. The embedded microphone makes it easy to measure sound pressure at the eardrum location of the human body model. These measurements can be used to determine the complex transfer function, which describes how sound travels through the ear to the eardrum with or without the hearing device in the appropriate position. Without hearing devices, the ears are unoccluded and the complex transfer function is commonly referred to as the head-related transfer function (HRTF). When the hearing device is placed in the appropriate position and turned off, the ears are closed and the complex transfer function can be referred to as the insertion transfer function (ITF). The insertion effect is the difference between HRTF and ITF. This is sometimes called "insertion loss" because of the magnitude attenuation associated with it, but phase is also affected because any resonance or filter that changes the magnitude response will necessarily change the phase as well.

The magnitude and phase difference between HRTF and ITF must be corrected for transparent recognition. The effect of insertion of the device into the ear canal is static and passive. Therefore, their resonance can be described in terms of minimum phase. Minimum phase systems have a number of useful properties: their effects are spectrally localized; They have stable inversions; For a given magnitude response, the minimum phase response is unique.

All these properties mean that the insertion effect can be eliminated by adding a second-order minimum phase filter complementary to the processing in the hearing aid. This way, both magnitude and phase responses are corrected. If a non-minimum phase filter is used, either the magnitude or phase response can be corrected, but not both at once. The transfer function that compensates for the insertion effect is called ATF (Aided Transfer Function), and is the same as HRTF when there is no hearing device.

ATF is a combination of direct sound (described as ITF) and amplified sound, at the eardrum. For this sum to work properly, the time delay between sounds must be minimized so that the phase delay at all frequencies amplified by the hearing aid corresponds to less than 120 degrees in phase. Phase delay can be adjusted by moving the microphone closer to the hearing aid's receiver and designing the hearing aid accordingly. These changes tend to be essential to the design. In contrast, the compensation filter for the ATF can be changed if the hearing aid is digital, for example, by reprogramming the digital signal processor chip. (It will be appreciated that the invention is not limited to digital implementations.)

To apply this method to the human ear, the in-ear response is measured with a probe microphone. The probe microphone is placed in the ear canal, and HRTF, ITF, and ATF are measured exactly the same with an acoustic manikin.

An alternative human application would be to take the subjective route: using source material at a level where the subject could hear it without difficulty, the examiner would be asked whether the source perception without the assistive device (HRTF) matched the ATF. If the subject can provide detailed guidance on the exact spectral difference between HRTF and ATF, it may be possible to find a filter equivalent to the measurement method. This approach is best suited for trained listeners, such as musicians or recording engineers.

Figure 1 schematically shows an example of an open-ear hearing aid 12 consisting of a microphone 13, a processor 15 and a speaker 17, where the incident sound, denoted by the number 10, passes through two sound paths, denoted A and B. It reaches the eardrum (11) through. Direct path A travels around the ear piece (not shown) and is characterized by an insertion transfer function (ITF). The amplified path B passes through microphone 13, processor 15 (which provides corrective equalization), and speaker 17. The perceived sound, indicated by arrow P, is the sum of the sounds reaching the eardrum through these two pathways.

An example of the implantation effect from an open ear hearing aid is shown in Figure 2, which shows transfer function measurements from an acoustic manikin. The insertion effect is the difference between HRTF and ITF; As shown in the top graph, the magnitude is different from and above 500 Hz (“insertion loss”); As shown in the bottom graph, the phase differs by more than 500 Hz.

The interpolation effect is shown corrected using a second order minimum phase filter in Figure 3. Note that the difference between ATF and HRTF is quite small in the 1-8 kHz range in magnitude and phase. The small dip at 950Hz is not the minimum phase resonance.

This concept is shown mathematically in the case of the minimum phase filter in Figures 4a and 4b. This is also generally true for any casual filter with stable inversion. In this embodiment, the direct sound attenuated by the hearing aid (ITF) is modeled with a bell-shaped attenuation filter (“attenuation”), which is minimal at the center frequency and approaches unity away from the center. Mathematically, the second-order minimum phase filter is given by the fourth-order equation:

where s is the Laplace variable, W is the angular frequency (=2πF, where F is the center frequency), Q is the quality factor, and G is the gain, which in this case is limited to be greater than 1. The transfer function of this filter is shown as a dashed line in Figure 4a.

The response (“boost”) of a hearing aid is modeled as a bandpass filter with maximum magnitude at the center frequency and gain approaching zero at the edges:

Their summation in the tympanic membrane corresponds to the ATF. Given a fixed attenuation filter the following parameters

A boost with can be shown analytically to result in a unity magnitude and zero phase response as shown in Figure 4b. The filter parameters in FIGS. 4A and 4B were chosen according to that relationship. Such a system is completely transparent.

Note that this embodiment corresponds to a filter summing in parallel. When two filters are placed in series, one acts on the output of the other and sums to unity under a much simpler condition: when the filters are opposite each other. The mathematical argument outlined above is a specific case and can be expressed as holding many other filter combinations, including two bell-shaped filters (two biquads), a high-pass and a low-pass.

The above example does not assume a time delay between direct and amplified sound. Therefore, there is no phase change at the peak frequency and negligible phase change at the surrounding frequencies, resulting in coherent summation at the eardrum. These conditions are met if there is no latency in the hearing aid and there is no significant distance (or propagation time) between the microphone and the hearing aid.

If the amplified sound is sufficiently delayed, there will be frequencies at which the phase is shifted by 180° relative to the direct sound. When added together in the eardrum, these sounds destructively cancel each other out. The relative size of the amplified sound to the direct sound at a given frequency determines whether the cancellation is complete (equal magnitude) or partial (unequal magnitude).

Most hearing aids have a latency of at least 1.5 ms, and if it does not last longer, this will result in significant offset and prevent proper compensation of the ITF. This case is modeled by adding a pure delay to the bandpass filter; The delay has a linear phase response as shown in Figures 5A and 5B. For a delay of 1.5 ms, there are two notable effects: 1) the magnitude response at the center frequency is smaller than that of the amplified sound alone, and 2) there is extensive combing around the center frequency. Comb filtering involves multiple notches with gains below -10dB that significantly distort the input signal.

Microphone delay can be reduced by shortening the separation distance between the microphone and receiver; It can be increased by adding delay in the processing circuitry (this may, but not necessarily be, a digital processor) or by moving the microphone further away from the receiver.

The block diagram in Figure 6 shows the basic steps described above for correcting the interjection effect of the hearing device according to the invention. As a first step, the effectiveness of insertion of the hearing device in the ear canal must be determined (block 102). This can be achieved as detailed above by performing measurements with the device present and removed from the external auditory canal. (This effect can also be achieved subjectively from input from the wearer as described above.) Once the effect of insertion of the hearing device in the ear canal is determined, it can then be corrected for magnitude and phase (block 103 ).

Figure 7 shows these steps in more detail where correction is determined using the acoustic manikin. The acoustic manikin provides a microphone embedded behind the outer ear designed to simulate the average frequency response at the eardrum (block 104). Once the hearing device is removed from the dummy's ear so that the external auditory canal is not occluded, the complex head-related transfer function (HRTF) is measured (block 105). Next, by positioning the hearing device in the manikin's external auditory canal (block 106), the complex insertion transfer function (ITF) can be measured with the hearing device turned off (block 107). Using the measured HRTF and ITF, the equalization needed to correct for the effects of implantation of the hearing device in the ear canal can be determined (block 108). As mentioned above, the calibration equalization will be the ratio of the measured HRTF to the measured ITF. This correction may be applied to the hearing device (block 109). The resulting auxiliary transfer function (ATF) can then be measured and compared to the HRTF.

The same steps shown in Figure 7 can be used using real people to correct for interpolation effects with an acoustic manikin. In this case, measurements are performed with a probe microphone at the eardrum.

It can be seen that the above-described steps can be repeated in an iterative manner to fine-tune the correction to reach the optimal ATF.

Although the invention has been described in considerable detail in the foregoing specification, it will be understood that the invention is not intended to be limited to such details except as required by the following claims.

Claims (27)

A method of correcting size and phase distortion of a hearing device in which at least a portion of the hearing device is inserted into the ear when worn by a user, comprising:
Determining the insertion effect of the hearing device when the hearing device is inserted into the user's ear and the hearing device is turned off, wherein the insertion effect is a complex insertion transfer function (ITF) having a magnitude and phase response. The determining step characterized by; and
Correcting the insertion effect by correcting both the magnitude and phase response of the ITF, wherein the phase response is corrected only when the phase response is at a minimum phase;
A correction method comprising: 2. The method of claim 1, wherein the insertion effect is corrected by at least one second order minimum phase filter. The method of claim 2, wherein the second order minimum phase filter is an infinite impulse response (IIR) filter. The method of claim 2, wherein the second order minimum phase filter is a biquad filter. The method of claim 1, wherein the insertion effect is corrected by a plurality of second-order minimum phase filters. 6. The method of claim 5, wherein the plurality of second order minimum phase filters are infinite impulse response (IIR) filters. The method of claim 5, wherein the plurality of second order minimum phase filters are biquad filters. 2. The method of claim 1, wherein the passage of sound through the ear to the eardrum without a hearing device in the ear is characterized by a composite head-related transfer function (HRTF), and wherein correcting for the insertion effect comprises: (HRTF) and a complex insertion transfer function (ITF), determining a desired equalization for correcting the ITF magnitude and phase response, wherein the magnitude and phase of the desired equalization are the ratio of the HRTF to the ITF. A correction method characterized by: The method of claim 8, wherein the composite HRTF is determined by measuring the composite HRTF of a human body model. 9. The method of claim 8, wherein the composite HRTF is determined by measuring the composite HRTF of a user of the hearing device. 9. The method of claim 8, wherein the composite ITF is determined by measuring the composite ITF of the hearing device on the manikin's ear. The method of claim 8, wherein the composite ITF is determined by measuring the composite ITF of the hearing device when worn by a user. 2. The method of claim 1, wherein correcting for insertion effects comprises determining a desired equalization for correcting ITF magnitude and phase response, wherein the desired equalization is subjectively determined by a user experienced in describing sounds. Compensating, wherein the sound heard by the user when wearing the hearing device is compared with the sound heard without the hearing device and the sound heard is substantially the same, and the desired equalization is achieved when there is no perceived difference between the two. method. 9. The method of claim 8, wherein if the magnitude and phase response of the hearing device are known, the equalization for correcting the magnitude and phase responses is calculated for all parts of the phase response that are of minimum phase. method. 9. Method according to claim 8, wherein different phase filtering is repeatedly introduced to the hearing device, wherein the phase response is minimal phase, until the desired phase correction is achieved. 2. The hearing device of claim 1, wherein the hearing device amplifies sound within the audio frequency spectrum, and the hearing device is configured such that the latency of the amplified sound corresponds to being less than about 120 degrees phase of the highest frequency produced by the hearing device. A method of correction, characterized in that it consists of: 2. The method of claim 1, wherein the latency of the hearing device is less than about one third of the period of the highest frequency produced by the hearing device. A method of correcting size and phase distortion of a hearing device in which at least a portion of the hearing device is inserted into the ear when worn by a user, wherein, when worn, the hearing device has the magnitude and phase response determined when the hearing device is turned off. The method for calculating the insertion effect characterized by a complex insertion transfer function (ITF) with:
configuring the hearing device so that the latency of the hearing device when worn is less than about one third of the period of the highest frequency amplified by the hearing device;
Correcting the magnitude response of the complex ITF; and
correcting the phase response of the composite ITF where the phase response is at a minimum phase;
A correction method comprising: 19. The method of claim 18, wherein the magnitude and phase response of the complex ITF are corrected using at least one minimum phase second order filter. 20. The method of claim 19, wherein the minimum phase second order filter is an infinite impulse response (IIR) filter. The method of claim 19, wherein the minimum phase secondary filter is a biquad filter. 19. The method of claim 18, wherein the latency of the hearing device creates a frequency-dependent phase delay of sound passing through the device, and where the ITF phase response and frequency-dependent phase delays are known, the phase correction is performed at a minimum. A correction method characterized in that calculation is performed on all parts of the phase response. 19. Method according to claim 18, wherein different minimum phase filtering is repeatedly introduced to the hearing device wherein the phase response is the minimum phase until the desired phase correction is achieved. A hearing device for producing amplified sound in one or more selected frequency bands wherein at least a portion of the hearing device is inserted into the ear when worn by a user, comprising:
mike;
a speaker insertable into the ear, wherein the distance between the microphone and the speaker is selected such that the latency of the hearing device when worn is less than about one third of a period of the highest frequency amplified by the hearing device; and
a processor between the microphone and the speaker;
Including,
At least a speaker of the hearing device produces an insertion effect when inserted into the ear, the insertion effect being characterized by a complex insertion transfer function (ITF) with a magnitude and phase response determined when the hearing device is turned off, and ,
and the processor is configured to correct for the insertion effect by correcting both the magnitude and phase response of the ITF, wherein the phase response is corrected only if the phase response is at a minimum phase. 25. The hearing device of claim 24, wherein the processor includes at least one minimum phase second order filter, the minimum phase second order filter being used to correct the magnitude and phase response of the complex ITF. 26. The hearing device of claim 25, wherein the minimum phase second order filter is an infinite impulse response (IIR) filter. 26. The hearing device of claim 25, wherein the minimum phase second order filter is a biquad filter.