Feedback cancellation using filter tap coherence values in hearing aid devices

Abstract

A system (20) for hearing assistance comprises one or more microphones (23, 24), a speaker (28) and a processing circuit (26). The one or more microphones are configured to be mounted near a subject's head and output electrical signals in response to acoustic waves incident on the microphones. The speaker is configured for mounting proximate an ear of a subject. The processing circuit is configured to amplify and filter the electrical signal using a digital filter (100) having a plurality of taps having respective tap coefficients selected to suppress feedback from the speaker to the microphone, in order to generate a drive signal for input to the speaker, and the processing circuitry is configured to adaptively calculate the tap coefficients while estimating respective coherence values for the tap coefficients over time, and to weight corresponding to update amounts for the tap coefficients in response to the respective coherence values.

Description

[0001] Field

[0002] The present invention generally relates to hearing aids, and more particularly to devices and methods for acoustic feedback cancellation.

[0003] Background

[0004] Speech understanding in noisy environments is a significant problem for hearing-impaired individuals. In addition to gain loss, hearing impairment is typically accompanied by a reduction in the temporal resolution of the sensory system. These characteristics further reduce the ability of hearing-impaired individuals to filter out the target source from background noise, particularly the ability to understand speech in noisy environments.

[0005] Some newer hearing aids provide a directional listening mode to improve speech clarity in noisy environments. This mode utilizes an array of microphones and applies beamforming techniques to combine multiple microphone inputs into a single directional audio output channel. The output channel has a spatial characteristic that increases the contribution of sound waves from the target direction relative to sound waves from other directions.

[0006] For example, PCT International Publication WO 2017 / 158507, the disclosure of which is incorporated herein by reference, describes a hearing aid device that includes a housing configured to be physically attached to a mobile phone. The microphone array is spaced within the housing and is configured to generate electrical signals in response to acoustic inputs to the microphones. An interface is fixed within the housing together with a processing circuit that is coupled to receive and process the electrical signals from the microphones in order to generate a combined signal for output via the interface.

[0007] As another example, PCT International Publication WO 2021 / 074818, the disclosure of which is incorporated herein by reference, describes a device for hearing assistance that includes an eyeglass frame that includes a front piece and temples, where one or more microphones are mounted at respective first positions on the front piece and are configured to output electrical signals in response to first sound waves incident on the microphones. A speaker mounted at a second position on one of the temples outputs second sound waves. A processing circuit generates a drive signal for the speaker by processing the electrical signals output by the microphones so that the speaker reproduces selected sounds present in the first sound waves with a delay within 20% of the transmission time of the first sound waves from the first position to the second position, thereby causing constructive interference between the first sound waves and the second sound waves.

[0008] Overview

[0009] The embodiments of the present invention described below provide improved devices and methods for hearing assistance.

[0010] The embodiments described herein provide a system for hearing assistance, the system including one or more microphones, a speaker, and processing circuitry. The one or more microphones are configured to be mounted near a subject's head and output an electrical signal in response to sound waves incident on the microphones. The speaker is configured to be mounted near the subject's ear. The processing circuitry is configured to amplify and filter the electrical signal using a digital filter having a plurality of taps to generate a drive signal for input to the speaker, the plurality of taps having respective tap coefficients selected to suppress feedback from the speaker to the microphones, and the processing circuitry is configured to adaptively calculate the tap coefficients while estimating respective coherence values of the tap coefficients over time and to weight an update amount applied to the tap coefficients in response to the respective coherence values.

[0011] In some embodiments, the processing circuitry is configured to adjust the tap coefficients to estimate a transfer function between one or more of the speaker and the microphones. In other embodiments, the processing circuitry is configured to use a gradient descent method having respective convergence factors to adjust the tap coefficients. In still other embodiments, the processing circuitry is configured to calculate the convergence factors based on the coherence values respectively.

[0012] In an embodiment, the processing circuitry is configured to calculate a convergence factor by multiplying a common convergence factor by the respective coherence value. In another embodiment, the processing circuitry is configured to evaluate the coherence value of a given tap based on a plurality of coefficient update amounts calculated for the given tap over a specified time period. In still another embodiment, the system for hearing assistance includes an eyeglass frame, and the microphone and the speaker are mounted at respective positions on the eyeglass frame.

[0013] In some embodiments, the one or more microphones include a plurality of microphones, and the processing circuitry is configured to apply a beamforming function to the electrical signals output by the plurality of microphones to enhance selected sounds originating within a selected angular range while suppressing background sounds originating outside the selected angular range.

[0014] According to an embodiment described herein, there is further provided a method for hearing assistance, the method including mounting an array of microphones near a subject's head, the microphones outputting an electrical signal in response to sound waves incident on the microphones, and mounting a speaker near the subject's ear. Amplifying and filtering the electrical signal using a digital filter having a plurality of taps to generate a drive signal for input to the speaker, the plurality of taps having respective tap coefficients selected to suppress feedback from the speaker to the microphones. Adaptively calculating the tap coefficients while estimating respective coherence values of the tap coefficients over time and weighting an update amount applied to the tap coefficients in response to the respective coherence values.

[0015] According to another embodiment described herein, there is further provided a head-mounted device (HMD) including a frame, one or more microphones, a speaker, and a processing circuit. The frame is configured to be mounted on a subject's head. The one or more microphones are mounted on the frame and are configured to output an electrical signal in response to sound waves incident on the microphones. The speaker is mounted on the frame. The processing circuit is configured to amplify and filter the electrical signal using a digital filter having a plurality of taps to generate a drive signal for input to the speaker, the plurality of taps having respective tap coefficients selected to suppress feedback from the speaker to the microphone, and the processing circuit is configured to adaptively calculate the tap coefficients while estimating respective coherence values of the tap coefficients over time and to weight an update amount applied to the tap coefficients in response to the respective coherence values.

[0016] In some embodiments, the HMD includes a device selected from the list including: an eyewear device, spectacles, a spectacle frame, goggles, a helmet, a face mask, a headset, and a clip-on device. In other embodiments, the one or more microphones are mounted on a front piece of the frame and the speaker is mounted on the frame near the subject's ear.

[0017] The present invention will be more fully understood from the following detailed description of embodiments of the invention in conjunction with the accompanying drawings, in which: Brief Description of the Drawings

[0019] Figure 1 is a schematic diagram showing a hearing assistance device based on a spectacle frame according to an embodiment of the present invention;

[0020] Figure 2 is a block diagram schematically showing details of a hearing assistance device according to an embodiment of the present invention;

[0021] Figure 3 is a block diagram schematically showing details of a feedback canceller applicable to a hearing assistance device according to an embodiment of the present invention; and

[0022] Figure 4A and Figure 4B is a block diagram schematically showing a processing scheme supporting both beamforming and feedback cancellation according to an embodiment of the present invention.

[0023] Detailed Description

[0024] Overview

[0025] Despite the need for directional hearing assistance and the theoretical benefits of microphone arrays in this regard, in practice, the directional performance of hearing aids is far lower than that achieved through natural hearing. Generally, good directional hearing assistance requires a relatively large number of microphones that are well-spaced, unobtrusive in design, and enable the user to easily direct the directional response of the hearing aid towards the point of interest, such as towards a conversation partner in a noisy environment. The processing circuit applies beamforming filters to the signals output by the microphones in response to incident sound waves to generate an audio output that enhances the sounds impinging on the microphone array within an angular range around the direction of interest while suppressing background noise. The audio output should reproduce the natural auditory experience as closely as possible while minimizing disturbing artifacts.

[0026] One of these artifacts is a loud whistling that can be caused by acoustic feedback from the audio output of a speaker located near the user's ear to the input of the microphone. This whistling occurs when the acoustic feedback gain of the hearing aid at a given frequency is greater than a certain threshold. Feedback cancellation in hearing aid devices is generally more challenging than in applications such as video conferencing and telephone calls, where the echo signal may be delayed by approximately 100 milliseconds, while in hearing aid devices, the feedback signal is typically delayed by less than 20 milliseconds, resulting in a high correlation between the spectra of the system output and input. Conventional solutions for suppressing or eliminating the feedback signal include reducing the gain of the hearing aid and filtering the audio frequency range in which the feedback occurs, but these solutions also reduce the effectiveness of the hearing aid in amplifying weak and high-pitched sounds. The feedback gain can also be mechanically reduced by fitting an ear mold to the user's ear, but many users find this solution uncomfortable and unappealing.

[0027] Embodiments of the invention described herein solve the problem of acoustic feedback by providing methods and systems for novel feedback cancellation by estimating the feedback signal and subtracting the feedback signal from the input signal. In the disclosed embodiments, an array of microphones mounted near the user's head outputs electrical signals in response to incoming sound waves incident on the microphones. A speaker is mounted near the user's ear. The processing circuit amplifies and filters the electrical signals using a digital filter having a plurality of taps to generate a drive signal for input to the speaker, the plurality of taps having respective tap coefficients selected to suppress feedback from the speaker to the microphone, and the processing circuit adaptively calculates the tap coefficients while estimating the respective coherence values of the tap coefficients over time and weights the update amount applied to the tap coefficients in response to the respective coherence values.

[0028] In some embodiments, a microphone and a speaker are mounted on a frame, and the frame is mounted on a user's head. In some embodiments described below, the microphone and the speaker are mounted on an eyeglass frame. Alternatively, the microphone and the speaker can be mounted on other kinds of frames or head-mounted devices (HMDs), such as virtual reality (VR) or augmented reality (AR) head-mounted devices, or in other kinds of mounting arrangements.

[0029] In this context, an HMD includes any kind of frame on which a microphone and a speaker can be mounted. The HMD can be selected from a list including (but not limited to) the following: an ocular wear device, glasses, a spectacle frame, goggles, a helmet, a face mask, a head-mounted device, and a clip-on device. In some embodiments, one or more microphones are mounted on a front piece of the frame, and the speaker is mounted on the frame near the subject's ear.

[0030] In some embodiments, a processing circuit adjusts tap coefficients to estimate a transfer function between one or more of the speaker and the microphone. The processing circuit uses the estimated transfer function to estimate a feedback signal to be subtracted from an input signal. The processing circuit can use a gradient descent method with a corresponding convergence factor to adjust the tap coefficients, and the processing circuit calculates the corresponding convergence factor based on a coherence value. In an embodiment, the processing circuit calculates the convergence factor by multiplying a common convergence factor by the corresponding coherence value.

[0031] In some embodiments, the processing circuit evaluates the coherence value of a given tap based on a plurality of coefficient update amounts calculated for the given tap over a specified time period.

[0032] In some embodiments, the system includes an eyeglass frame, where the microphone and the speaker are mounted at corresponding positions on the eyeglass frame.

[0033] In an embodiment, one or more microphones include a plurality of microphones, and the processing circuit applies a beamforming function to electrical signals output by the plurality of microphones to enhance selected sounds originating within a selected angular range while suppressing background sounds originating outside the selected angular range.

[0034] System description

[0035] Figure 1 is a schematic diagrammatic illustration of a hearing assistance device 20 integrated into an eyeglass frame 22 according to an embodiment of the present invention. An array of microphones 23, 24 is mounted at corresponding positions on the eyeglass frame 22 and outputs electrical signals in response to sound waves incident on the microphones. In the illustrated example, the microphone 23 is mounted on a front piece 30 of the frame 22, while the microphone 24 is mounted on a temple 32 that is connected to a corresponding edge of the front piece 30. Although Figure 1The extensive array of microphones 23 and 24 shown is useful in some applications of the present invention, but the principles of signal processing and hearing assistance described herein can also be alternatively applied using a smaller number of microphones (with necessary modifications). For example, these principles can be applied using an array of microphones 23 on the front piece 30, as well as in devices using other microphone mounting arrangements that are not necessarily based on glasses.

[0036] The processing circuit 26 is fixed within or otherwise connected to the glasses frame 22 and is coupled via wires 27 (such as traces on a flexible printed circuit) to receive the electrical signals output from the microphones 23, 24. Although the processing circuit 26 is shown in Figure 1 , for simplicity, some or all of the processing circuit can alternatively be located within the front piece 30 or in a unit externally connected to the frame 22 at some location within the temple 32. The processing circuit 26 mixes the signals from the microphones so as to generate an audio output having a particular directional response, for example by applying a beamforming function so as to enhance sounds originating within a selected angular range while suppressing background sounds originating outside that range. Generally, although not necessarily, the directional response is aligned with the angular orientation of the frame 22. The processing circuit additionally suppresses acoustic signals picked up by the microphones that originate from the speakers.

[0037] These signal processing functions of the processing circuit 26 will be described in more detail below.

[0038] The processing circuit 26 can transmit the audio output to the user's ears via any suitable kind of interface and speaker. In the illustrated embodiment, the audio output is generated by a drive signal for driving one or more audio speakers 28 that are mounted on the temples 32, typically near the user's ears. Although Figure 1 shows only a single speaker 28 on each temple 32, the device 20 can alternatively include only a single speaker on one of the temples 32, or the device 20 can include two or more speakers mounted on one or both of the temples 32. In the latter case, the processing circuit 26 can apply a beamforming function in the drive signal so as to direct the sound waves from the speakers towards the user's ears. Alternatively, the drive signal can be transmitted to a speaker inserted into the ear, or can be transmitted via a wireless connection (such as a magnetic signal) to a pick-up coil in a hearing aid (not shown) worn by the user of the glasses frame.

[0039] Signal Processing

[0040] Figure 2is a block diagram schematically showing details of a processing circuit 26 in a hearing assistance device 20 according to an embodiment of the present invention. The processing circuit 26 can be implemented in a single integrated circuit chip, or alternatively, the functions of the processing circuit 26 can be distributed among multiple chips, which can be located inside or outside the spectacle frame 22. Although a specific implementation is shown in Figure 2 , the processing circuit 26 can alternatively include any suitable combination of analog and digital hardware circuits, as well as a suitable interface for receiving the electrical signals output by the microphones 23, 24 and outputting drive signals to the speaker 28.

[0041] In this embodiment, the microphones 23, 24 include integrated analog / digital converters that output digital audio signals to the processing circuit 26. Alternatively, the processing circuit 26 can include an analog / digital converter for converting the analog output of the microphone into digital form. The processing circuit 26 generally includes a suitable programmable logic component 40, such as a digital signal processor (DSP) or a gate array, which implements the necessary filtering and mixing functions, as well as a feedback cancellation function, to generate and output a drive signal in digital form for the speaker 28.

[0042] These filtering and mixing functions generally include the application of a beamforming filter 42, the coefficients of which are selected to produce a desired directional response. Specifically, in some embodiments, the coefficients of the beamforming filter 42 are calculated to enhance sounds that impinge on the frame 22 (and thus on the microphones 23, 24) within a selected angular range. Details of filters that can be used for beamforming purposes are further described below.

[0043] Alternatively or additionally, the processing circuit 26 can include a neural network (not shown) that is trained to determine and apply the coefficients to be used in the beamforming filter 42. Further alternatively or additionally, the processing circuit 26 includes a microprocessor that is programmed in software or firmware to perform at least some of the functions described herein.

[0044] In implementing the beamforming filter 42, the processing circuit 26 can apply any suitable beamforming function known in the art in the time domain or the frequency domain. For example, beamforming algorithms that can be used in this context are described in the above-mentioned PCT International Publication WO 2017 / 158507 (especially pages 10 - 11) and US Patent 10,567,888 (especially column 9).

[0045] In one embodiment, the processing circuit 26 applies a minimum variance distortionless response (MVDR) beamforming algorithm when deriving the coefficients of the beamforming filter 42. This algorithm is advantageous in achieving fine spatial resolution and distinguishing sounds originating from the direction of interest from sounds originating from the user's own speech. The MVDR algorithm maximizes the signal-to-noise ratio (SNR) of the audio output by minimizing the average energy (while keeping the target distortion small). The algorithm can be implemented by calculating a vector F(ω) of complex weights of the output signals from each microphone at each frequency, as represented by the following formula:

[0046]

[0047] In this formula, W(ω) is the propagation delay vector between the microphones 23, representing the desired response of the beamforming filter as a function of angle and frequency; and S zz (ω) is the cross-spectral density matrix, representing the covariance of the acoustic signals in the time-frequency domain. To calculate the coefficients of the beamforming filter 42, S zz (ω) is measured or calculated for isotropic far-field noise.

[0048] In an alternative embodiment, the processing circuit 26 applies a linearly constrained minimum variance (LCMV) algorithm when deriving the coefficients of the beamforming filter 42. LCMV beamforming causes the beamforming filter to pass signals from the desired direction with a specified gain and phase delay while minimizing the power of interfering signals and noise from all other directions.

[0049] In some embodiments, the processing circuit 26 includes a feedback canceller 44 that suppresses acoustic feedback from the speaker to the microphone. To this end, the feedback canceller 44 uses a digital filter (not shown) with a plurality of taps having respective tap coefficients that are selected to suppress feedback from the speaker to the microphone, and adaptively calculates the tap coefficients while estimating the respective coherence values of the tap coefficients over time, and weights the update amount applied to the tap coefficients in response to the respective coherence values. The feedback canceller will be described in detail below with reference to Figure 3 Detailed description of the feedback canceller.

[0050] The audio output circuit 46, for example, includes a suitable codec and digital / analog converter that converts the digital drive signal output from the beamforming filter 42 (or from the feedback canceller 44 behind the beamforming filter) into analog form. The analog filter 48 performs further filtering and analog amplification functions to optimize the analog drive signal to the speaker 28.

[0051] The control circuit 50 (such as, an embedded microcontroller) controls the programmable functions and parameters of the processing circuit 26, which may include a feedback canceller 44. The communication interface 52 (e.g., Bluetooth or other wireless interface) enables a user and / or an audiology professional to set and adjust these parameters as needed. The power circuit 54 (such as, a battery inserted into the temple 32) supplies power to the other components of the processing circuit.

[0052] Feedback cancellation processing

[0053] As described above, the sound waves generated by the speaker of the hearing aid device can be picked up by the microphone of the device, which may result in a howl sound. The goal of the feedback canceller is to prevent howling artifacts by reducing the amount of feedback signal within the signal generated by the microphone.

[0054] Next, the principle of feedback cancellation is described. Let Out(t) denote the signal output by the hearing aid device, let p(t) denote the signal received by the microphone solely from the output of the hearing aid device (the version of Out(t) received by the microphone), and let y(t) denote the signal received by the microphone from all audio sources other than the speaker of the hearing aid device, where t represents the time axis. The overall signal x(t) generated by the microphone is given by x(t) = y(t) + p(t).

[0055] The feedback canceller estimates the feedback signal based on the output signal Out(t - Δt) (e.g., a reference signal) generated Δt periods before as follows The feedback canceller estimates the transfer function from the output of the hearing device (speaker) to the microphone, denoted as and applies the estimated transfer function to the signal Out(t - Δt) to produce the estimated feedback signal given by:

[0056]

[0057] The feedback canceller further subtracts the estimated feedback signal from x(t) to produce a signal x′(t) given by:

[0058]

[0059] where the feedback is suppressed. In digital form, the transfer function can be implemented using an adaptive filter including multiple taps, where the tap coefficients are adjusted using any suitable adaptive method. The tap coefficients can be adjusted using any suitable gradient descent method, such as for example the least mean square (LMS) or normalized LMS (NLMS) method. Alternatively, other suitable adaptation methods can also be used. As will be referred to below Figure 4AAnd Figure 4B As described, depending on whether feedback cancellation is performed before or after beamforming, the adaptive filter can model the transfer function between the speaker and multiple microphones or between the speaker and a single microphone.

[0060] Figure 3 FIG. is a block diagram schematically showing details of a feedback canceller 44 applicable to a hearing aid device 20 according to an embodiment of the present invention. Alternatively, the principle of this feedback canceller can be applied to other devices and systems having a suitable microphone array, speaker, and signal processing capabilities.

[0061] Figure 3 The feedback canceller 44 implements the above-described feedback cancellation principle in digital form, where each signal is sampled on a digital time axis, denoted as "n". In Figure 3 the example, the feedback canceller 44 receives an input signal x(n), which is received by microphones 23, 24 and includes a feedback signal from the speaker 28. Using a subtractor 104, the feedback canceller subtracts the estimated feedback signal from x(n) to produce a signal x'(n), where the feedback is suppressed or eliminated.

[0062] The feedback canceller 44 includes an adaptive filter 100, which includes N taps having respective tap coefficients, where N is an integer greater than 1. The feedback canceller generates an estimated feedback signal by filtering an output signal Out(n) using the current values of the tap coefficients of the adaptive filter 100. In some embodiments, the output signal Out(n) includes a drive signal in digital form to the speaker. The adaptive filter can include any suitable number N of taps. In an example embodiment, the number of taps is approximately 100 or more taps, such as 120 taps. The main reasons for selecting so many taps are: (i) performing feedback cancellation on a tight beamformer, and (ii) the frequency response of the speaker of the underlying hearing eyewear being significantly different from a flat frequency response. Due to these reasons, the processed signal is smeared over a relatively long time, which requires a relatively long filter.

[0063] The tap adapter 108 uses any suitable gradient descent method (such as, for example, the LMS or NLMS method) to update the tap coefficients of the adaptive filter 100. Let Δh(n) denote a vector of coefficient update amounts corresponding to the taps of the adaptive filter 100, respectively. The vector Δh(n) has the same length N as the adaptive filter 100. In this example, the tap adapter performs the sequential update steps given below:

[0064]

[0065] Among them, the update vector Δh(n) is given by the following formula:

[0066] Δh(n) = μ · Out(n) · X(n)

[0067] Among them, μ is the scalar convergence factor of the basic gradient descent method, and the vector X(n) is given by the following formula:

[0068] X(n) = [x(n - N + 1)... x(n)]

[0069] Next, an embodiment is described in which the tap adapter 108 adjusts the tap coefficients based on multiple convergence factors rather than a single scalar convergence factor. In such an embodiment, for each tap, the common convergence factor μ is weighted by the corresponding weight value. In some embodiments, the tap adapter calculates the weight value by calculating the coherence value of the corresponding tap as described herein. This method provides a time-based weighting mechanism for modifying the update amount Δh(n) of the tap coefficients applied to the adaptive filter. The inventors have found that, for example, in open-ear hearing eye wearables, weighting the tap coefficient update amount by the corresponding time coherence value of the tap can significantly improve the feedback cancellation performance.

[0070] The performance of the feedback cancellation method can be determined, for example, by measuring the maximum acoustic output gain at which the underlying system remains stable without whistling. The inventors have found that the gain applicable using the disclosed coherence-based feedback cancellation method is significantly higher than the gain achievable when omitting the coherence value.

[0071] Generally, using the coherence value involves evaluating the update amount adaptively applied to each tap of the adaptive filter within a short time period (e.g., within a 16-millisecond time period (or any other suitable time period)), and weighting the update amount of the tap coefficients based on the coherence value respectively. In some embodiments, the coherence value C for weighting the coefficient update amount of the i-th tap i is given, for example, by the following formula:

[0072]

[0073] Among them, n represents the digital time index, represents the coefficient update amount applied to the i-th tap at time n, and W represents the number of samples used to calculate the coherence value. The coherence value falls within the range between 0 and 1, and the maximum value 1 is obtained when all the coefficient values used to calculate the coherence value are equal to each other. Although not mandatory, the coefficient values can be calculated based on the sequence of the W consecutive tap update amounts most recently applied to the relevant tap. Alternatively, other recent W tap update amounts can also be used. The gradient factor of the i-th tap coefficient weighted by the i-th coherence value is given by:

[0074]

[0075] Coherence value C i indicates the corresponding reliability level associated with the coefficient update amount. When a tap is associated with a large coherence value (the update amount is considered highly reliable), the gradient factor μ is weighted high, and when the tap is associated with a smaller coherence value (in this case, the update amount is considered less reliable), the gradient factor μ is weighted low.

[0076] The method for calculating the coherence value as described above is given by way of example, and other types of coherence values can also be used. For example, the symbol coherence value with reduced complexity is given by:

[0077]

[0078] As another example, each coherence value C i can be multiplied by a factor C given by: g to generalize the coherence value:

[0079]

[0080] where the sum in the last equation is taken over a number W’ > 1 of taps.

[0081] In embodiments where feedback cancellation is performed in the frequency domain, a phase coherence factor can be applied. For example, such example formulas can be found in the paper "Phase Coherence Imaging: Principles, applications and current developments" of Signal Processing in Acoustics: PSP(2 / 3) Presentation 1 in Bruges, Belgium.

[0082] Beamforming and feedback cancellation schemes

[0083] Figure 4A and Figure 4B are block diagrams schematically showing a processing scheme that supports both beamforming and feedback cancellation according to an embodiment of the present invention.

[0084] Figure 4A and Figure 4B The schemes in are different from each other in the order of performing beamforming and feedback cancellation.

[0085] In Figure 4AIn the solution, input signals from multiple microphones are processed by a beamforming filter (such as, for example, the beamforming filter 42 above Figure 2 . Then, for example, a feedback canceller 44 using Figure 2 and Figure 3 is used to perform feedback cancellation on the signal output by the beamforming filter. In such an embodiment, Figure 3 the adaptive filter models the transfer function from the speaker (e.g., Figure 2 28) to the combined (virtual) microphone including multiple microphones. The interface 120 provides the signal output by the feedback canceller to the speaker. The interface 120 may include, for example, a codec / DAC (e.g., Figure 2 46), followed by an analog filter (e.g., Figure 2 48).

[0086] In Figure 4B the solution, input signals from multiple microphones are processed by dedicated respective feedback cancellers 44. The output of the feedback canceller is input to the beamforming filter 42, and the output of the beamforming filter is provided to the speaker via the interface 120. In such an embodiment, Figure 3 the adaptive filter models the transfer function from the speaker (e.g., Figure 2 28) to individual microphones.

[0087] Figure 4A The solution of Figure 4B is not as complex as that of Figure 4A because it has only one feedback canceller instead of multiple feedback cancellers. In addition, Figure 4B the beamforming performance in the solution of Figure 4B may be better than that of the solution of

[0088] because applying individual feedback cancellation to individual microphones (as in Figure 4B ) may reduce the correlation between microphones, which is required for the correct operation of the beamforming filter. Figure 4A On the other hand, Figure 4B the solution in

[0089] Although the embodiments described herein primarily address feedback cancellation in hearing assistive devices, the methods and systems described herein may also be used in other applications, such as feedback cancellation in other HMD devices, and for use in noise canceling headphones.

[0090] It should be understood that the above embodiments are cited by way of example, and the appended claims are not limited to what has been specifically shown and described above. On the contrary, the scope includes combinations and sub-combinations of the various features described above and variations and modifications of these features that will occur to a person skilled in the art after reading the above description and are not disclosed in the prior art. Documents incorporated by reference into this patent application are considered to be an integral part of this application, except that any term is defined in these incorporated documents to some extent in a manner that conflicts with the definition explicitly or implicitly made in this specification, and only the definition in this specification should be considered.

Claims

1. A system for hearing assistance, comprising: One or more microphones configured to be mounted near a subject's head and output an electrical signal in response to sound waves incident on the microphones; A speaker configured to be mounted near the subject's ear; And A processing circuit configured to amplify and filter the electrical signal using a digital filter having a plurality of taps to generate a drive signal for input to the speaker, the plurality of taps having respective tap coefficients selected to suppress feedback from the speaker to the microphone, and the processing circuit being configured to adaptively calculate the tap coefficients while estimating respective coherence values of the tap coefficients over time and weighting an update amount applied to the tap coefficients in response to the respective coherence values.

2. The system according to claim 1, wherein The processing circuit is configured to adjust the tap coefficients to estimate a transfer function between the speaker and one or more of the microphones.

3. The system according to claim 1, wherein The processing circuit is configured to adjust the tap coefficients using a gradient descent method having respective convergence factors.

4. The system according to claim 3, wherein, The processing circuit is configured to calculate the convergence factors based on the coherence values respectively.

5. The system according to claim 3, wherein, The processing circuit is configured to calculate the convergence factors by multiplying a common convergence factor by the respective coherence values.

6. The system according to claim 3, wherein The processing circuit is configured to evaluate a coherence value of a given tap based on a plurality of coefficient update amounts calculated for the given tap over a specified time period.

7. The system according to any one of claims 1 - 6, and comprising an eyeglass frame, wherein the microphone and the speaker are mounted at respective positions on the eyeglass frame.

8. The system according to any one of claims 1-6, wherein The one or more microphones include a plurality of microphones, and wherein the processing circuit is configured to apply a beamforming function to the electrical signals output by the plurality of microphones to enhance selected sounds originating within a selected angular range while suppressing background sounds originating outside the selected angular range.

9. A method for hearing assistance, comprising: Mounting an array of microphones near a subject's head, the microphones outputting an electrical signal in response to sound waves incident on the microphones; Mounting a speaker near the subject's ear; And Amplifying and filtering the electrical signal using a digital filter having a plurality of taps to generate a drive signal for input to the speaker, the plurality of taps having respective tap coefficients selected to suppress feedback from the speaker to the microphone, and adaptively calculating the tap coefficients while estimating respective coherence values of the tap coefficients over time and weighting an update amount applied to the tap coefficients in response to the respective coherence values.

10. The method according to claim 9, wherein, Calculating the tap coefficients includes adjusting the tap coefficients to estimate a transfer function between the speaker and one or more of the microphones.

11. The method according to claim 9, wherein, Calculating the tap coefficients includes using a gradient descent method having respective convergence factors to adjust the tap coefficients.

12. The method according to claim 11, and comprising calculating the convergence factors based on the coherence values respectively.

13. The method according to claim 11, wherein, Calculating the convergence factor includes multiplying a common convergence factor by the respective coherence value.

14. The method according to claim 11, and including evaluating the coherence value of a given tap based on a plurality of coefficient update amounts calculated for the given tap over a specified time period.

15. The method according to any one of claims 9-14, wherein, The microphone and the speaker are mounted at respective positions on the spectacle frame.

16. The method according to any one of claims 9-14, wherein, The one or more microphones include a plurality of microphones, and the method includes applying a beamforming function to the electrical signals output by the plurality of microphones so as to enhance selected sounds originating within a selected angular range while suppressing background sounds originating outside the selected angular range.

17. A head-mounted device (HMD) comprising: a frame configured to be mounted on a subject's head; one or more microphones mounted on the frame and configured to output electrical signals in response to sound waves incident on the microphones; a speaker mounted on the frame; and a processing circuit configured to amplify and filter the electrical signals using a digital filter having a plurality of taps to generate a drive signal for input to the speaker, the plurality of taps having respective tap coefficients selected to suppress feedback from the speaker to the microphones, and the processing circuit being configured to adaptively calculate the tap coefficients while estimating respective coherence values of the tap coefficients over time and to weight an update amount applied to the tap coefficients in response to the respective coherence values.

18. The HMD according to claim 17, wherein, The HMD includes a device selected from the list including: an ocular wear device, glasses, a spectacle frame, goggles, a helmet, a face mask, a head-mounted device, and a clip-on device.

19. The HMD according to claim 17 or 18, wherein, The one or more microphones are mounted on a front piece of the frame, and wherein the speaker is mounted on the frame adjacent to the subject's ear.

Images