HEARING AID SYSTEM COMPRISING A DATABASE OF ACOUSTIC TRANSFER FUNCTIONS

Abstract

A hearing aid system comprises a hearing aid configured to be worn on the head at or in an ear of a user. The hearing aid comprises a microphone system comprising a multitude of M of microphones arranged in said hearing aid and adapted to provide M corresponding electric input signals x.sub.m(n), m=1, . . . , M, n representing time. The environment sound at a given microphone comprises a mixture of a) a target sound signal s.sub.m(n) propagated via an acoustic propagation channel from a direction to or a location (θ) of a target sound source to the m.sup.th microphone of the hearing aid when worn by the user, and b) possible additive noise signals v.sub.m(n) as present at the location of the m.sup.th microphone, wherein the acoustic propagation channel is modeled as x.sub.m(n)=s.sub.m(n)h.sub.m(θ)+v.sub.m(n), and wherein h.sub.m(θ) is an acoustic impulse response for sound for that acoustic propagation channel. The hearing aid system comprises A) a processor connected to said number of microphones, and B) a database Θ comprising a multitude of dictionaries Δ.sub.p, p=1, . . . , P, where p is a person index, of vectors, termed ATF-vectors, whose elements ATF.sub.m, m=1, . . . , M, are frequency dependent acoustic transfer functions representing direction- or location-dependent (θ), and frequency dependent (k) propagation of sound from a direction or location (θ) of a target sound source to each of said M microphones, k being a frequency index, k=1, . . . , K, where K is a number of frequency bands, when said microphone system is mounted on a head at or in an ear of a natural or artificial person (p′), and wherein each of said dictionaries Δ.sub.p comprises ATF-vectors for a given person (p) for a multitude of different directions or locations θ.sub.j, j=1, . . . , J, relative to the microphone system. The processor is configured to, at least in a learning mode of operation, determine personalized ATF-vectors (ATF*) for said user based on said multitude of dictionaries Δ.sub.p of said database Θ, said electric input signals x.sub.m(n), m=1, . . . , M, and said model of the acoustic propagation channels. The invention may e.g. be used in beamforming, own voice estimation, own voice detection, keyword detection, etc.

Claims

1. A hearing aid system comprising a hearing aid configured to be worn on the head at or in an ear of a user, the hearing aid comprising a microphone system comprising a multitude of M of microphones arranged in said hearing aid, where M is larger than or equal to two, the microphone system being adapted for picking up sound from the environment and to provide M corresponding electric input signals x.sub.m(n), m=1, . . . , M, n representing time, the environment sound at a given microphone comprising a mixture of a target sound signal s.sub.m(n) propagated via an acoustic propagation channel from a direction to or a location (θ) of a target sound source to the m.sup.th microphone of the hearing aid when worn by the user, and possible additive noise signals v.sub.m(n) as present at the location of the m.sup.th microphone, wherein the acoustic propagation channel is modeled as x.sub.m(n)=s.sub.m(n)h.sub.m(θ)+v.sub.m(n), and wherein h.sub.m(θ) is an acoustic impulse response for sound for that acoustic propagation channel; the hearing aid system comprising a processor connected to said number of microphones, and a database Θ comprising a multitude of dictionaries Δ.sub.p, p=1, . . . , P, where p is a person index, of vectors, termed ATF-vectors, whose elements ATF.sub.m, m=1, . . . , M, are frequency dependent acoustic transfer functions representing direction- or location-dependent (θ), and frequency dependent (k) propagation of sound from a direction or location (θ) of a target sound source to each of said M microphones, k being a frequency index, k=1, . . . , K, where K is a number of frequency bands, when said microphone system is mounted on a head at or in an ear of a natural or artificial person (p), and wherein each of said dictionaries Δ.sub.p comprises ATF-vectors for a given person (p) for a multitude of different directions or locations θ.sub.j, j=1, . . . , J, relative to the microphone system; and wherein the processor is configured to, at least in a learning mode of operation, determine personalized ATF-vectors (ATF*) for said user based on said multitude of dictionaries Δ.sub.p of said database Θ, said electric input signals x.sub.m(n), m=1, . . . , M, and said model of the acoustic propagation channels.

2. A hearing aid system according to claim 1 wherein said frequency dependent acoustic transfer functions ATF comprise absolute acoustic transfer functions AATF.

3. A hearing aid system according to claim 1 wherein said frequency dependent acoustic transfer functions ATF comprise relative acoustic transfer functions RATF.

4. A hearing aid system according to claim 1 wherein each of said dictionaries Δ.sub.p, p=1, . . . , P, of said database Θ are hearing aid-orientation specific and comprises ATF vectors (ATF.sub.θ,p,φ) for a multitude of different hearing aid-orientations φ.sub.q, q=1, . . . , Q, on the head of the given person (p), for said multitude of different directions or locations θ.sub.j, j=1, . . . , J.

5. A hearing aid system according to claim 1 wherein each of said dictionaries Δ.sub.p of said database Θ comprises a set of person- and hearing aid-orientation-specific AATF-vectors H.sub.θ,p,φ and/or RATF-vectors d.sub.θ,p,φ comprising absolute or relative transfer functions for a given person (p) among said multitude of different persons, p=1, . . . , P, with different heads, and for a multitude of different hearing aid-orientations (φ) on said given head, and for said multitude of different directions or locations θ.sub.j, j=1, . . . , J.

6. A hearing aid system according to claim 1 wherein said personalized AATF or RATF-vector (H*, d*) for said user is determined for different frequency indices (k) using the same AATF or RATF-vectors (H.sub.θ,p, d.sub.θ,p, H.sub.θ,p,φ, d.sub.θ,p,φ) for some or all frequency indices to estimate a given personalized AATF or RATF-vector (H*, d*).

7. A hearing aid system according to claim 1 wherein the personalized AATF or RATF-vector (H* or d*), respectively, for the user is determined by a statistical method or a learning algorithm.

8. A hearing aid system according to claim 7 wherein the personalized AATF or RATF-vector (H* or d*) for the user is determined by minimizing a cost function.

9. A hearing aid system according to claim 1 configured to log the estimated personalized ATF-vectors (ATF*) over time and thereby building a database of personalized acoustic transfer functions for different directions/locations.

10. A hearing aid system according to claim 1 configured to enter said learning mode of operation during or after a power-up of the hearing aid system, or on request from the user, or if one or more sensors indicate a change in a position of the hearing aid, e.g. due to remounting of the hearing aid.

11. A hearing aid system according to claim 1, wherein for given electric input signals, the processor is configured to, at least in said learning mode of operation, evaluate each of the dictionaries Δ.sub.p of AATF or RATF-vectors (H.sub.θ,p, d.sub.θ,p) for different persons p, p=1, . . . , P, that correspond to a candidate direction to or location (θ) for all values of the frequency index k, k=1, . . . , K, and to determine an optimal person (p*) based thereon.

12. A hearing aid system according to claim 11 configured to enter a normal mode of operation after said learning mode of operation, and wherein the hearing aid system is configured to analyse data of the preceding learning mode to identify the person (p**) among said P persons that most frequently has been determined as the optimal person (p*), and to use the dictionary Δ.sub.p** of said person (p**) to determine personalized ATF-vectors (ATF*), e.g. absolute acoustic transfer functions (AATF), for said user in said normal mode of operation.

13. A hearing aid system according to claim 4, wherein for given electric input signals, the processor is configured to, at least in a learning mode of operation, evaluate each of the dictionaries Δ.sub.p of AATF or RATF-vectors (H.sub.θ,φ,p, d.sub.θ,φ,p) for different persons p, p=1, . . . , P, and for the multitude of different hearing aid-orientations φ.sub.q, q=1, . . . , Q, on the head of said person (p), that correspond to a candidate direction to or location (θ) for all values of the frequency index k, k=1, . . . , K, and to determine an optimal person (p*) and an optimal hearing aid-orientation (φ.sub.q*) based thereon.

14. A hearing aid system according to claim 13 configured to enter a normal mode of operation after said learning mode of operation, and wherein the hearing aid system is configured to analyse data of the preceding learning mode to identify the person (p**) among said P persons that most frequently has been determined as the optimal person (p*), and to identify the hearing aid orientation (φ.sub.q**) among said Q orientations in the directory Δ.sub.p** of AATF or RATF-vectors (H.sub.θ,φ,p**, d.sub.θ,φ,p**) that most frequently has been determined as the optimal hearing aid orientation (φ.sub.q*), and to use the dictionary Δ.sub.p** of said person (p**) and said hearing aid orientation (φ.sub.q**) to determine personalized ATF-vectors (ATF*) for said user in said normal mode of operation.

15. A hearing aid system according to claim 11 wherein the processor is configured to select the AATF or RATF vector (H.sub.θ,p, d.sub.θ,p, H.sub.θ,φ,p, d.sub.θ,φ,p) corresponding to a specific person (p), and optionally to a specific hearing aid orientation (φ.sub.q), that is optimal as the personalized AATF or RATF-vector (H* or d*), respectively, for said user in the given acoustic situation.

16. A hearing aid system according to claim 1 wherein the hearing aid comprises said database Θ.

17. A hearing aid system according to claim 1 wherein the hearing aid comprises said processor.

18. A hearing aid according to claim 1 comprising a beamformer filter configured to provide a spatially filtered signal based on said electric input signals and beamformer weights, wherein the beamformer weights are determined using said personalized AATF or RATF-vector (H*, d*) for said user.

19. A hearing aid system according to claim 1 comprising an auxiliary device wherein said database is stored, and wherein said hearing aid and said auxiliary device comprise antenna and transceiver circuitry allowing data to be exchanged between them.

20. A method of operating a hearing aid system comprising a hearing aid configured to be worn on the head at or in an ear of a user is provided, the method comprising providing by a multitude of microphones a corresponding multitude of electric input signals x.sub.m(n), m=1, . . . , M, n representing time, comprising environment sound from the environment of the user, wherein the environment sound of a given one of said multitude of electric input signals comprises a mixture of a target sound signal s.sub.m(n) propagated via an acoustic propagation channel from a direction to or a location (θ) of a target sound source to the m.sup.th microphone of the hearing aid when worn by the user, and possible additive noise signals v.sub.m(n) as present at the location of the m.sup.th microphone, wherein the acoustic propagation channel is modeled as x.sub.m(n)=s.sub.m(n)h.sub.n(θ)+v.sub.m(n), and wherein h.sub.m(θ) is an acoustic impulse response for sound for that acoustic propagation channel providing, or providing access to, a database Θ comprising a dictionary Δ.sub.p of vectors, termed ATF-vectors, whose elements ATF.sub.m(θ,p,k), m=1, . . . , M, are frequency dependent acoustic transfer functions representing direction- or location-dependent (θ), and frequency dependent (k) propagation of sound from a location (θ) of a target sound source to each of said M microphones, k being a frequency index, k=1, . . . , K, where K is a number of frequency bands, when said microphone system is mounted on a head at or in an ear of a natural or artificial person (p), and wherein said dictionary Δ.sub.p comprises ATF-vectors ATF for said person (p) for a multitude of different directions or locations θ.sub.j, j=1, . . . , J relative to the microphone system; providing that the database Θ comprises a multitude P of dictionaries Δ.sub.p, p=1, . . . , P, where p is a person index, said dictionaries comprising ATF-vectors ATF for a corresponding multitude of different natural or artificial persons (p); and processing, at least in a learning mode of operation, said multitude of dictionaries Δ.sub.p of said database Θ, said electric input signals x.sub.m(n), m=1, . . . , M, and said model of the acoustic propagation channels to thereby determine personalized ATF-vectors ATF* for said user.

Description

BRIEF DESCRIPTION OF DRAWINGS

[0121] The aspects of the disclosure may be best understood from the following detailed description taken in conjunction with the accompanying figures. The figures are schematic and simplified for clarity, and they just show details to improve the understanding of the claims, while other details are left out. Throughout, the same reference numerals are used for identical or corresponding parts. The individual features of each aspect may each be combined with any or all features of the other aspects. These and other aspects, features and/or technical effect will be apparent from and elucidated with reference to the illustrations described hereinafter in which:

[0122] FIG. 1A schematically illustrates a multitude Q of examples of a head of a person (p) wearing a hearing aid system comprising left and right hearing aids, wherein the right hearing aid is mounted as intended (with a microphone axis parallel to a horizontal reference direction φ=0) and where in the top drawing the left hearing aid is mounted with its microphone direction tilted out of the horizontal plane with a first relatively small tilt angle (φ=φ.sub.1) compared to the reference direction (φ=0), and in the bottom drawing the left hearing aid is mounted with its microphone direction tilted out of the horizontal plane with a different, Q.sup.th, relatively large tilt angle (φ=φ.sub.Q) compared to the reference direction (φ=0),

[0123] FIG. 1B schematically illustrates a multitude Q of examples of a head of a person (p) wearing a hearing aid system comprising left and right hearing aids, wherein the left hearing aid is mounted as intended (parallel to a horizontal reference direction φ=0) and in the top drawing the right hearing aid is mounted with its microphone direction tilted in the horizontal plane with a first relatively small tilt angle (φ=φ.sub.1) compared to the reference direction (φ=0), and in the bottom drawing the right hearing aid is mounted with its microphone direction tilted in the horizontal plane with a different, Q.sup.th, relatively large tilt angle (φ=φ.sub.Q) compared to the reference direction (φ=0),

[0124] FIG. 1C schematically illustrates a typical geometrical setup of a user wearing a binaural hearing aid system in an environment comprising a (point) source in a front half plane of the user,

[0125] FIG. 1D schematically illustrates a head of a person (p) wearing a hearing aid system comprising left and right hearing aids, wherein the left and right hearing aid are mounted as intended (parallel to a horizontal reference direction φ=0), and where the test sound is positioned at a multitude J of directions (represented by angles θ.sub.j, j=1, . . . , J) in a horizontal plane relative to the centre of the persons head, and

[0126] FIG. 1E schematically illustrates a multitude P of examples of heads of a person (p=1, . . . , P) wearing a hearing aid system comprising left and right hearing aids, wherein the heads of the different persons have different characteristics, here head size a.sub.p,

[0127] FIG. 2A schematically illustrates for a given 1.sup.st test person (p=1), a combination of measurements of acoustic transfer functions ATF for different locations (θ.sub.j, j=1, . . . , J), and for each location for different hearing aid-orientations (φ.sub.q, q=1, . . . , Q), and for each hearing aid orientation for each frequency index (k, k=1, . . . , K), and

[0128] FIG. 2B schematically illustrates the same as FIG. 2A, but for the P.sup.th test person, each person (p, p=1, . . . , P) being assumed to have different acoustic characteristics of the head, e.g. different head sizes (a.sub.p),

[0129] FIG. 3 schematically shows an exemplary block diagram of a hearing aid according to the present disclosure,

[0130] FIG. 4 shows an exemplary block diagram of a hearing aid according to the present disclosure comprising a beamformer with personalized weights,

[0131] FIG. 5A shows a learning mode for determining the optimal head-and-torso characteristics from the set of candidate head-and-torso characteristics (‘person’) in a dictionary, and a subsequent normal mode wherein the optimal person-specific head-and-torso-characteristics (‘person’) determined in the learning mode is fixed; and

[0132] FIG. 5B shows a learning mode for determining an optimal (representation of the current) orientation of the hearing aid on the user's head and a subsequent normal mode wherein the optimal orientation determined in the learning mode is fixed.

[0133] The personalized parameters z*(z=p, θ, φ) may e.g. be stored together with a parameter indicating a quality (e.g. a signal to noise ratio (SNR), or an estimated noise level, or a signal level, etc.) of the electric input signals that were used to determine the parameter value(s) in question. Thereby the logged personalized parameter values may be qualified.

[0134] The figures are schematic and simplified for clarity, and they just show details which are essential to the understanding of the disclosure, while other details are left out. Throughout, the same reference signs are used for identical or corresponding parts.

[0135] Further scope of applicability of the present disclosure will become apparent from the detailed description given hereinafter. However, it should be understood that the detailed description and specific examples, while indicating preferred embodiments of the disclosure, are given by way of illustration only. Other embodiments may become apparent to those skilled in the art from the following detailed description.

DETAILED DESCRIPTION OF EMBODIMENTS

[0136] The detailed description set forth below in connection with the appended drawings is intended as a description of various configurations. The detailed description includes specific details for the purpose of providing a thorough understanding of various concepts. However, it will be apparent to those skilled in the art that these concepts may be practiced without these specific details. Several aspects of the apparatus and methods are described by various blocks, functional units, modules, components, circuits, steps, processes, algorithms, etc. (collectively referred to as “elements”). Depending upon particular application, design constraints or other reasons, these elements may be implemented using electronic hardware, computer program, or any combination thereof.

[0137] The electronic hardware may include micro-electronic-mechanical systems (MEMS), integrated circuits (e.g. application specific), microprocessors, microcontrollers, digital signal processors (DSPs), field programmable gate arrays (FPGAs), programmable logic devices (PLDs), gated logic, discrete hardware circuits, printed circuit boards (PCB) (e.g. flexible PCBs), and other suitable hardware configured to perform the various functionality described throughout this disclosure, e.g. sensors, e.g. for sensing and/or registering physical properties of the environment, the device, the user, etc. Computer program shall be construed broadly to mean instructions, instruction sets, code, code segments, program code, programs, subprograms, software modules, applications, software applications, software packages, routines, subroutines, objects, executables, threads of execution, procedures, functions, etc., whether referred to as software, firmware, middleware, microcode, hardware description language, or otherwise.

[0138] The present application relates to the field of hearing aids, in particular to beamforming/noise reduction.

[0139] The present disclosure relates to a hearing aid (e.g. comprising a microphone array), or a binaural hearing aid system, configured to estimate personalized absolute or relative acoustic transfer functions for a user of the hearing aid (or hearing aid system).

[0140] The present disclosure is based on the assumption that a dictionary of absolute acoustic transfer function (AATFs) and/or relative transfer functions (RATFs), i.e., acoustic transfer functions from a target signal source to any microphones in the hearing aid system relative to a reference microphone, is available. Basically, the proposed scheme aims at finding the AATF or RATF in the dictionary which, with highest likelihood (or other optimization measure) (among the dictionary entries), was “used” in creating the currently observed (noisy) target signal.

[0141] This dictionary element may then be used for beamforming purposes (the absolute or relative acoustic transfer function is an element of most beamformers, e.g. an MVDR beamformer).

[0142] Additionally, since each AATF or RATF dictionary element has a corresponding direction or location attached to it, an estimate of the direction of arrival (DOA) is thereby provided. Likewise, since each AATF or RATF dictionary element may have a corresponding hearing aid-orientation associated with it, an estimate of the hearing aid-orientation (or its deviation from an intended orientation) is thereby provided. Likewise, since each AATF or RATF dictionary element may have a corresponding person (or characteristics of the head) associated with it, an estimate of characteristics of the head of the user can thereby be provided.

[0143] The database Θ may then—for individual microphones of the microphone system—comprise corresponding values of location of or direction to a sound source (e.g. indicated by horizontal angle θ), and absolute (AATF) or relative transfer functions RATF at different frequencies (AATF(k,θ) or RATF(k,θ), k representing frequency) from the sound source at that location to the microphone in question. The proposed scheme may calculate likelihoods (or other, e.g. cost-function based, measures) for a sub-set of, or all, absolute or relative transfer functions of the database (and thus locations/directions) and microphones and points to the location/direction having largest (e.g. maximum) likelihood (or other measure).

[0144] The microphone system may e.g. constitute or form part of a hearing device, e.g. a hearing aid, adapted to be located in and/or at an ear of a user. In an aspect, a hearing system comprising left and right hearing devices, each comprising a microphone system according to the present disclosure is provided. In an embodiment, the left and right hearing devices (e.g. hearing aids) are configured to be located in and/or at left and right ears, respectively, of a user.

[0145] The method chooses actual AATFs or RATFs from a dictionary of candidate AATFs or RATFs. Using a dictionary of candidate AATFs or RATFs ensures that the resulting AATF or RATF is physically plausible—it is a way of imposing the prior knowledge that the microphones of the hearing assistive device are located at a particular position on the head of the user. According to the present disclosure, the database is populated with AATFs or RATFs from several (potentially many) different heads, and/or AATFs or RATFs for hearing assistive devices in different position on the ear of the user.

[0146] The proposed idea comprises extended dictionaries, where the extension consists of

a) Several AATFs or RATFs from the same person, but measured with different HA positions (e.g., tilts).
b) AATFs or RATFs from the same angles/positions, but for several (potentially many) persons' (heads).

c) Combination of a) and b).

[0147] d) The extended dictionary may contain AATFs or RATFs for each ear individually, or the combined set of microphones for both ears (e.g. for binaural beamforming).

[0148] When trying out dictionary elements (in order to decide the AATFs or RATFs that are active (optimal) for the current situation), we may try out (e.g. evaluate) particular sub-sets of the extended dictionary, for example:

a) For each candidate direction/position, try out (e.g. evaluate) the subset of AATF- or RATF-vectors for all frequencies that correspond to a particular HA-tilt (orientation). Select the subset of AATF or RATF vectors that is best, e.g. in maximum likelihood sense.
b) For each candidate direction/position, try out (e.g. evaluate) the subset of AATF- or RATF-vectors for all frequencies that correspond to a particular head. Select the subset of AATF- or RATF vectors that is best, e.g. in maximum likelihood sense.

c) Combination of a) and b).

[0149] d) Subsets may describe binaural AATFs or RATFs (e.g. using M=3 or M=4 mics), or monaural AATFs or RATFs, typically using M=2 or M=3 dimensional AATF- or RATF vectors.

[0150] ‘Try out’ may e.g. be taken to mean ‘evaluate a likelihood’ of a given candidate among a given subset of transfer functions and to pick out the candidate fulfilling a given optimization criterion (e.g. having a maximum likelihood, ML).

[0151] This procedure has the following advantages:

1) Better performance in general, because we take into account consistent information about HA position (a)), or user head characteristics. For example: all AATF- or RATF-vectors correspond to a particular head.
2) Significantly lower search complexity: we don't do an exhaustive search where all AATF- or RATF vector combinations (for all frequencies, HA orientations (e.g. tilts), persons (e.g. heads)) are tried out (e.g. evaluated) exhaustively. Instead, they are tried out (e.g. evaluated) in physically plausible subsets, i.e., only AATF- or RATF-vectors that “belong together” are tried out.
3) The selected subset provides an estimate of underlying HA-positions or user head types/sizes. This information may be passed on to improve performance of other algorithms, e.g. for compensating for microphone positions, etc.
4) If applied independently to each ear (i.e., monaurally), we can detect if HA location/tilt is different across ears (see FIG. 1A, 1B). If applied jointly to both ears (i.e., in a binaural setup), performance in determining the correct AATF or RATF set is expected to increase because more measurements are used.

[0152] FIGS. 1A and 1B each illustrate a user (U) wearing a binaural hearing system comprising left and right hearing devices HD.sub.L, HD.sub.R), which are differently mounted at left and right ears of a user, in FIG. 1A one hearing device having its microphone axis pointing out of the horizontal plane (φ≠0), and in FIG. 1B one hearing device having its microphone axis not pointing in the look direction of the user (θ≠0).

[0153] FIG. 1A shows a multitude Q of examples of a head of a person (e.g. a test subject p) wearing a hearing aid system comprising left and right hearing aids, wherein the right hearing aid (HD.sub.R) is mounted as intended (with a microphone axis parallel to a horizontal reference direction φ=0 (REF-DIR.sub.R)). In the top drawing, the left hearing aid (HDL) is mounted with its microphone direction (MIC-DIR.sub.L) tilted out of the horizontal plane with a first relatively small tilt angle (φ=φ.sub.1) compared to the reference direction (φ=0) (REF-DIR.sub.L). In the bottom drawing, the left hearing aid (HD.sub.L) is mounted with its microphone direction (MIC-DIR.sub.L) tilted out of the horizontal plane with a different, Q.sup.th, relatively large tilt angle (φ=φ.sub.Q) compared to the reference direction (φ=0). It is indicated (by bold dots •••) that other tilt angles (φ.sub.q) may be present between φ.sub.1 and φ.sub.Q (i.e. symbolizing q=1, . . . , Q).

[0154] FIG. 1B schematically illustrates a multitude Q of examples of a head of a person (e.g. a test subject p) wearing a hearing aid system comprising left and right hearing aids (HD.sub.L, HDR), wherein the left hearing aid (HD.sub.L) is mounted as intended (parallel to a horizontal reference direction φ=0 (REF-DIR.sub.L)) and in the top drawing the right hearing aid (HDR) is mounted with its microphone direction (MIC-DIR.sub.R) tilted in the horizontal plane with a first relatively small tilt angle (φ=φ.sub.1) compared to the reference direction (φ=0) (REF-DIR.sub.L), and in the bottom drawing the right hearing aid (HD.sub.R) is mounted with its microphone direction tilted in the horizontal plane with a different, Q.sup.th, relatively large tilt angle (φ=φ.sub.Q) compared to the reference direction (φ=0). Again, it is indicated (by bold dots •••) that other tilt angles (φ.sub.q) may be present between φ.sub.1 and φ.sub.Q (i.e. symbolizing q=1, . . . , Q).

[0155] FIG. 1C schematically illustrates a typical geometrical setup of a user wearing a binaural hearing system comprising left and right hearing devices (HD.sub.L, HD.sub.R), e.g. hearing aids, on his or her head (HEAD) in an environment comprising a (point) source (S) in a front (left) half plane of the user defined by a distance cis between the sound source (S) and the centre of the user's head (HEAD), e.g. defining a centre of a coordinate system. The user's nose (NOSE) defines a look direction (LOOK-DIR) of the user (or Test subject p), and respective front and rear directions relative to the user are thereby defined (see arrows denoted Front and Rear in the left part of FIG. 1C). The sound source S is located at an angle (−)θ.sub.s to the look direction of the user in a horizontal plane (e.g. through the ears of the user). The left and right hearing devices (HD.sub.L, HD.sub.R) are located—a distance a apart from each other—at left and right ears (Ear.sub.L, Ear.sub.R), respectively, of the user (or Test subject p). Each of the left and right hearing devices (HD.sub.L, HD.sub.R) comprises respective front (M.sub.1x) and rear (M.sub.2x) microphones (x=L (left), R (right)) for picking up sounds from the environment. The front (M.sub.1x) and rear (M.sub.2s) microphones are located on the respective left and right hearing devices a distance ΔL.sub.M (e.g. 10 mm) apart, and the axes formed by the centres of the two sets of microphones (when the hearing devices are correctly mounted at the user's ears) define respective reference directions (REF-DIR.sub.L, REF-DIR.sub.R) of the left and right hearing devices, respectively, of FIGS. 1A, 1B and 1C. The direction to the sound source may define a common direction-of-arrival for sound received at the left and right ears of the user. The real direction-of-arrival of sound from sound source S at the left and right hearing devices will in practice be different (d.sub.sL, d.sub.sR) from the one defined by arrow d.sub.s (the difference being larger, the closer the source is to the user). If considered necessary, the correct angles (θ.sub.L, θ.sub.R) may e.g. be determined from the geometrical setup (including angle θ.sub.s, distance d.sub.s and distance a between the hearing devices).

[0156] As illustrated in FIG. 1A, 1B, the hearing device, e.g. hearing aids, may not necessarily point towards the position corresponding to the ideal position (REF-DIR). The hearing aid(s) may be tilted by a certain (out of the horizontal plane) elevation angle φ (cf. FIG. 1A), and the hearing aids may alternatively or additionally point at a slightly different horizontal direction than anticipated (cf. angle φ in FIG. 1B), or a combination thereof (having e.g. a component out of the horizontal plane as well as in the horizontal plane but deviating from the reference direction). The database Θ (e.g. one for each of the left and right hearing aids (HD.sub.L, HD.sub.R), e.g. each adapted to and stored in the respective hearing aid) is configured to take account of this in that the different dictionaries Δ.sub.p comprise acoustic transfer functions (ATF) (for each microphone m) for a multitude of different hearing aid-orientations φ.sub.q for the different persons (p), and for each direction/location (θ.sub.j) of the sound source (and as a function of frequency (k)).

[0157] As indicated in FIG. 1A, a movement sensor such as an accelerometer or a gyroscope (denoted acc in FIG. 1A) may be used to estimate if the instrument is tilted compared to the horizontal plane (cf. indications of accelerometer acc, and tilt angle φ relative to the direction of the force of gravity (represented by acceleration of gravity g) in FIG. 1A on the left hearing device HD.sub.L). A magnetometer may help determine if the two instruments are not pointing towards the same direction. Such indication may be used to qualify an indication provided by the personalized hearing aid orientation φ* determined according to the present disclosure.

[0158] FIG. 1D schematically illustrates a multitude J of examples of a head of a person (p) wearing a hearing aid system comprising left and right hearing aids (HD.sub.L, HD.sub.R), wherein the left and right hearing aids are mounted with a given hearing aid orientation φ.sub.q (e.g. as intended (parallel to a horizontal reference direction φ=0)), and where the test sound is positioned at a multitude J of directions/locations (here represented by angles θ.sub.j, j=1, . . . , J to sound sources located on a circle around (i.e. a fixed distance from) the test subject p) in a horizontal plane relative to the centre of the persons head. Each angle step is 360°/J, e.g. 30° for J=12, or 15° for J=24. An acoustic transfer function, e.g. an absolute acoustic transfer function AATF.sub.m=i(θ.sub.2, φ.sub.q, p, k) is schematically indicated by the dashed arrow from the sound source at θ.sub.2 to microphone M.sub.i (e.g. defined as a reference microphone) of the right hearing aid (HD.sub.R) for a given person p, a given hearing aid orientation φ.sub.q, and a given frequency k. It is assumed that a dictionary Δ.sub.p of acoustic transfer functions ATF (absolute (AATF) or relative (RATF)) for a given person p comprises values for each microphone (m=1, . . . , M), each direction/location of the sound source (θ.sub.j, j=1, . . . , J), each hearing aid orientation (φ.sub.q, q=1, . . . , Q) and for all frequencies (k=1, . . . , K). The database may comprise a number of (similarly ‘equipped’) dictionaries Δ.sub.p for different artificial or natural persons (p=1, . . . , P).

[0159] FIG. 1E schematically illustrates a multitude P of examples of heads of a person (p=1, . . . , P) wearing a hearing aid system comprising left and right hearing aids (HD.sub.L, HD.sub.R), wherein the heads of the different persons have different characteristics, here head size a.sub.p. Test subject 1 has a relatively smaller head whereas Test subject P has a relatively larger head. As indicated by bold dots •••, acoustic transfer functions (different dictionaries Δ.sub.p) for a number of different persons (Test subjects) may be present in the database Θ (i.e. symbolizing p=1, 2, . . . , P). The number of persons may e.g. be larger than 5, e.g. larger than 10, e.g. smaller than 20. In the given examples of FIG. 1E, acoustic transfer functions ATF for a direction/location θ=0° of the sound source (Test sound) to a microphone M.sub.i of the right hearing aid at frequency k is illustrated (ATF.sub.m(θ=0, k). It is, however, to be understood that acoustic transfer functions for each microphone (m) (possibly for each hearing aid (HD.sub.L, HD.sub.R)), each direction/location (θ.sub.j), (optionally) each hearing aid-orientation (φ.sub.q), each frequency index (k) are recoded and stored in person specific dictionaries (Δ.sub.p) of the database(s). This is illustrated in FIG. 2A, 2B. Hearing aid specific databases (Θ.sub.L, Θ.sub.R) are e.g. stored in each of the left and right hearing aids (HD.sub.L, HD.sub.R), respectively, or in a commonly accessible device or system, e.g. accessible via an APP of a smartphone or other portable device.

[0160] FIG. 2A shows schematically illustrates for a given 1.sup.st test person (p=1), a combination of measurements of acoustic transfer functions ATF.sub.m(θ.sub.j, φ.sub.q, p=1, k) for different microphones (m=1, . . . , M) of the right hearing aid (HD.sub.R) and/or for a binaural hearing system comprising left and right hearing aids (HD.sub.L, HD.sub.R), and for each microphone for different directions/locations (θ.sub.j, j=1, . . . , J), and for each direction/location for different hearing aid-orientations (φ.sub.q, q=1, . . . , Q), and for each hearing aid orientation for each frequency index (k, k=1, . . . , K). ATF in FIG. 2A (and 2B) refers to a vector comprising elements ATF.sub.m, m=1, . . . , M. The geometrical measurement setup for different directions/locations is as in FIG. 1D (FIG. 1D illustrating a part of the measurements illustrated by FIG. 2A). The geometrical measurement setup for different hearing aid orientations (φ.sub.q, q=1, . . . , Q) is illustrated by the inserts in the lower right part of FIG. 2A (and 2B). It is intended that the measurements may be performed individually on microphones of the right hearing aid (HD.sub.R) and the left hearing aid (HD.sub.R). The results of the measurements may be stored in respective left and right hearing aids (databases Θ.sub.L and Θ.sub.R) or in a common database Θ.sub.C stored in one of or in each of the left and right hearing aids, or in another device or system in communication with the left and/or right hearing aids. The data of FIG. 2A are e.g. organized in dictionary Δ.sub.p=1.

[0161] FIG. 2B schematically illustrates the same as FIG. 2A, but for the P.sup.th test person, each person (p, p=1, . . . , P) being assumed to have different acoustic characteristics of the head, e.g. different head sizes (a.sub.r). The data of FIG. 2A are e.g. organized in dictionary Δ.sub.p=P.

[0162] It is assumed that the same acoustic transfer functions ATF.sub.m(θ.sub.j, φ.sub.q, p=p′, k) for possible further persons p′ ‘between’ person 1 (FIG. 2A) and person P (FIG. 2B) are measured and stored in the respective databases (Θ.sub.L, Θ.sub.R, Θ.sub.C. The direction to or location of the sound source relative to the hearing aid (microphone system or microphone) is symbolically indicated by symbol θ.sub.j and shown in FIGS. 2A, 2B (and 1C, 1D) as an angle in a horizontal plane, e.g. a horizontal plane through the ears of the person or user (when the person or user is in an upright position). It may however also indicate a location, e.g. in a horizontal plane, e.g. (θ.sub.s, d.sub.s) (as in FIG. 1C) or out of a horizontal plane (e.g. x, y, z). The acoustic transfer functions ATF stored in the database(s) may be or represent absolute acoustic transfer functions AATF or relative acoustic transfer functions RATF.

[0163] FIG. 3 shows an exemplary block diagram of a hearing aid (HD) according to the present disclosure. The hearing aid (HD) may e.g. be configured to be worn on the head at or in an ear of a user (or be partly implanted in the head at an ear of the user). The hearing aid comprises a microphone system comprising a multitude of M of microphones (here two M.sub.1, M.sub.2), e.g. arranged in a predefined geometric configuration, in the housing of the hearing aid. The microphone system is adapted to pick up sound from the environment and to provide corresponding electric input signals x.sub.m(n), m=1, 2, where n represents time. The environment sound at a given microphone may comprise a mixture (in various amounts) of a) a target sound signal s.sub.m(n) propagated via an acoustic propagation channel from a direction to or a location (θ) of a target sound source to the m.sup.th microphone of the hearing aid when worn by the user, and b) additive noise signals v.sub.m(n) as present at the location of the m.sup.th microphone. The acoustic propagation channel is modeled as x.sub.m(n)=s.sub.m(n)h.sub.m(θ)+v.sub.m(n), wherein h.sub.m(θ) is an acoustic impulse response for sound for that acoustic propagation channel. The hearing aid comprises a processor (PRO) connected to the microphones (M.sub.1, M.sub.2) receiving electric signals (X.sub.1, X.sub.2). The electric signals (X.sub.1, X.sub.2) are here provided in a time frequency representation (k, l) as frequency sub-band signals by respective analysis filter banks (FB-A1, FB-A2). One or more (here both) of the electric signals (X.sub.1, X.sub.2) are further fed to a voice activity detector (VAD) for estimating a presence or absence of human voice (e.g. speech). The voice activity detector provides a voice activity control signal (V-NV) indicative of whether or not (or with what probability) an input signal comprises a voice signal (at a given point in time). The voice activity control signal (V-NV) is fed to the processor (PRO) for possible use in the estimation of a current acoustic transfer function (ATF). The hearing aid further comprises a database Θ (MEM [DB]) comprising a dictionary Δ.sub.p of vectors, termed ATF-vectors, whose elements ATF.sub.m(θ,p,k), m=1, . . . , M, are frequency dependent acoustic transfer functions representing direction- or location-dependent (θ), and frequency dependent (k) propagation of sound from a location (θ) of a target sound source to each of said M (here M=2) microphones, k is the frequency index, k=1, . . . , K, where K is a number of frequency bands. The acoustic transfer functions are determined when the microphone system (e.g. a hearing aid device) is mounted on a head at or in an ear of a natural or artificial person (p). The microphone system is preferably mounted on the person in a configuration identical to, or as close as possible to, the configuration of the hearing aid (e.g. a style identical to the style of the hearing aid worn by the user). The dictionary Δ.sub.p comprises ATF-vectors ATF.sub.θ,p,φ, for a person (p), for a multitude of different directions or locations θ.sub.j, j=1, . . . , J relative to the microphone system, and (optionally) for a multitude of different hearing aid-orientations φ.sub.q, q=1, . . . , Q, on the head of said person (p), for the multitude of different directions or locations θ.sub.j, j=1, . . . , J. In FIG. 3, dictionaries Δ.sub.p for a multitude P of persons are stored in the database. The processor (PRO) is connected to the database (MEM [DB]) and configured to determine personalized ATF-vectors ATF*.sub.θ for the user based on the database Θ, the electric input signals x.sub.m(n), m=1, . . . , M, (here m=1, 2) and the model of the acoustic propagation channels. The personalized ATF-vectors ATF* for the user may be determined by a number of different methods available in the art, e.g. maximum likelihood estimate (MLE) methods, cf. e.g. EP3413589A1. Other statistical methods may e.g. include Mean Squared Error (MSE), regression analysis (e.g. Least Squares (LS)), e.g. probabilistic methods (e.g. MLE), e.g. supervised learning (e.g. neural network algorithms). The personalized ATF-vector ATF* for the user may e.g. be determined by minimizing a cost function. The processor (PRO) may be configured—at a given time with given electric input signals—to determine a personalized ATF-vector ATF* for the user as an ATF-vector ATF* (ATF*.sub.m(θ*,φ*,p*,k), m=1, . . . , M, k=1, . . . , K), i.e. an acoustic transfer function (relative or absolute) for each microphone, for each frequency (k). The personalized ATF-vectors ATF*.sub.θ are determined from the dictionary Δ.sub.p and the chosen vector is associated with a specific person p=p*, a specific direction/location θ.sub.j=θ* to/of the sound source, and a specific hearing aid-orientation φ*, and may thus provide information about the user's head characteristics, an estimated direction/location φ* to the target sound source, and an estimate of the current orientation φ.sub.q=φ* of the hearing aid in question. The processor (PRO) may be configured to present this information to other parts of the hearing aid, e.g. as in FIG. 4 to a signal processor (SP) for applying processing algorithms to one or more signals of the forward path, e.g. a beamforming algorithm. In the embodiment of FIG. 3, the personalized ATF-vector ATF* for the user as well as corresponding values of the specific person p=p*, the specific direction/location θ.sub.j=θ* to/of the sound source, and the specific hearing aid-orientation φ* associated with the (current) personalized ATF-vector are fed to the signal processor (SP). The hearing aid may e.g. be configured to log one or more of said personalized parameters (e.g. the person p*). This may be used to get an indication of the head characteristics of the user (if the person p* corresponds to the same p-value when logged over time). The personalized parameters z* (z=p, θ, φ) may e.g. be stored together with a parameter indicating a quality (e.g. a signal to noise ratio (SNR), or an estimated noise level, or a signal level, etc.) of the electric input signals that were used to determine the parameter value(s) in question (see e.g. FIG. 4).

[0164] The hearing aid (HD) of FIG. 3 comprises a forward (audio signal) path configured to process the electric input signals (IN1, IN2) and to provide enhanced (processed) output signal for being presented to the user. The forward path comprises the input transducers (here microphones (M.sub.1, M.sub.2)) respective analysis filter banks (FB-A1, FB-A2), a signal processor (SP), a synthesis filter bank (FBS), and an output transducer (here a loudspeaker SPK) operationally connected to each other.

[0165] The processor (PRO) and the signal processor (SP) may form part of the same digital signal processor (or be independent units). The analysis filter banks (FB-A1, FB-A2), the processor (PRO), the signal processor (SP), the synthesis filter bank (FBS), and the voice activity detector (VAD) may form part of the same digital signal processor (or be independent units).

[0166] The synthesis filter bank (FBS) is configured to convert a number of frequency sub-band signals to one time-domain signal. The signal processor (SP) is configured to apply one or more processing algorithms to the electric input signals (e.g. beamforming and compressive amplification) and to provide a processed output signal (OUT) for presentation to the user via an output transducer. The output transducer (here a loudspeaker SPK) is configured to convert a signal representing sound to stimuli perceivable by the user as sound (e.g. in the form of vibrations in air, or vibrations in bone, or as electric stimuli of the cochlear nerve).

[0167] The hearing aid may comprise a transceiver allowing an exchange of data with another device, e.g. a smartphone or any other portable or stationary device or system. The database Θ may be located in the other device. Likewise, the processor PRO may be located in the other device.

[0168] FIG. 4 shows an exemplary block diagram of a hearing aid (HD) according to the present disclosure comprising a beamformer with personalized weights. The embodiment of IG. 4 is similar to the embodiment of FIG. 3 but additionally comprises an SNR-estimator (SNRE), a datalogger (D-LOG), and a beamformer filter BF. The SNR-estimator (SNRE) is configured to estimate a current signal-to-noise-ratio (SNR) (or an equivalent estimate of a quality of the current electric input signals (IN1, IN2) or of a signal (e.g. a beamformed signal, (Y.sub.BF)) or signals originating therefrom). The SNR estimate (SNR) is fed to the processor (PRO) (and possibly to other parts of the hearing aid, e.g. to the beamformer and or to the hearing aid gain controller (HA-G). The datalogger (D-LOG) may store personalized parameters PMT* (e.g. the specific person p=p*, the specific direction/location θ.sub.j=θ* to/of the sound source, and the specific hearing aid-orientation go* associated with the (current) personalized ATF-vector ATF* (e.g. absolute or relative acoustic transfer functions (H*.sub.θ or d*.sub.θ) together with a parameter indicating a quality (e.g. a signal to noise ratio (SNR), or an estimated noise level, or a signal level, etc.) of the electric input signals that were used to determine the parameter value(s) in question. The datalogger (D-LOG) receives the personalized parameters PMT* from the processor (PRO). The personalized parameters PMT* may be qualified using the SNR estimate, so that only personalized parameters determined at an SNR above a threshold value are logged (e.g. stored) in the data logger (D-LOG). The processor (PRO) feeds the personalized acoustic transfer function (here the relative acoustic transfer function (d*.sub.θ) and optionally the current direction/location (θ.sub.j=θ*) associated therewith) determined for the current acoustic situation (as reflected by the electric input signals (IN1, IN2), and optionally by the voice activity detector (VAD)) to the signal processor (SP), specifically to the beamformer filter (BF). The beamformer filter (BF) also receives the electric input signals (IN1, IN2) in a time-frequency representation (k,l), where k and l are frequency and time (−frame) indices, respectively. The beamformer filter (BF) provides a beamformed signal Y.sub.BF in dependence of the electric input signals (X.sub.1(k,l), X.sub.2(k,l)) and the personalized relative acoustic transfer function (d*.sub.θ) as e.g. discussed in EP3253075A1. The beamformed signal Y.sub.BF is fed to a hearing aid gain controller (HA-G) for applying a frequency dependent gain (e.g. provided by a compression algorithm) to the beamformed signal according to the user's needs, e.g. to compensate for a hearing impairment of the user. The hearing aid gain controller (HA-G) provides a processed signal (OUT) as a frequency sub-band signal that is fed to the synthesis filter bank (FBS) for conversion to the time-domain signal out (as in FIG. 3). The beamformer filter and the hearing aid gain controller (HA-G) form part of the signal processor (SP), as indicated by dotted rectangle denoted SP in FIG. 4. All fully digital functional components of FIG. 4 (FB-A1, FB-A2, VAD, PRO, SNRE, D-LOG, BF, HA-G, FBS and optionally memory (MEM)) may form part of a customized or standard digital signal processor (adapted for audio processing). The memory (MEM) may e.g. be implemented as a separate chip.

[0169] The hearing aid (HD), e.g. the processor (PRO), may e.g. be configured to log the estimated personalized ATF-vectors ATF* (e.g. d*.sub.θ) over time and thereby building a database of personalized acoustic transfer functions for different directions/locations. The hearing aid, e.g. the processor (PRO), may e.g. be configured to only log personalized ATF-vectors ATF* that are associated with a quality (e.g. SNR) of the electric input signals is above a certain threshold value. In case the logged (possibly qualified by a signal quality parameter) personalized parameter p* is consistently equal to a specific value p.sub.u of p, the dictionary Δ.sub.pu of ATF-vectors associated with that person (p.sub.u) may be used by the hearing aid instead of the proposed estimation scheme. The hearing aid, e.g. the processor (PRO), may be configured to perform the transition itself in dependence of the logged data and a transition criterion (e.g. regarding the number of stored directions/locations for which personalized acoustic transfer functions are stored, and/or regarding a minimum time over which the personalized ATF-vectors ATF* have been logged and/or regarding the quality of the estimated ATF-vectors).

[0170] FIG. 5A shows a learning mode for determining an optimal person (p*) (e.g. ‘head and torso characteristics’) and a subsequent normal mode wherein the optimal person (p**) determined in the duration of the learning mode is fixed. FIG. 5B shows a learning mode for determining an optimal (representation of the current) orientation of the hearing aid on the user's head and a subsequent normal mode wherein the optimal orientation determined in the learning mode is fixed.

[0171] The left part of FIGS. 5A and 5B illustrates the learning mode from time t=t.sub.0 to time t=t.sub.1, i.e. of a duration Δt.sub.LM=t.sub.1−t.sub.0, e.g. ≤1 day, such as ≤1 hour, e.g. ≤10 minutes. During the learning mode, a relevant acoustic transfer functions for given electric input signals (reflecting a specific acoustic situation) may be determined using all dictionaries Δ.sub.p, p=1, . . . , P, of the database Θ, as disclosed in the present application. This may e.g. be as briefly described in the following.

[0172] The following procedure may be followed: For given electric input signals, for each directory Δ.sub.p, p=1, . . . , P, of the database Θ (or physically plausible subset thereof), find the optimal location (θ.sub.j*p) for the given directory (corresponding to a person, p) by determining a cost function for of the locations (θ.sub.j, j=1, . . . , J) (or a subset thereof), and then finally choose the optimum location (θ.sub.j*) among the P directories (or a subset thereof) as the location (θ.sub.j*) exhibiting the lowest cost function (e.g. maximum likelihood). Thereby an optimal person (p*) (and optionally the hearing aid orientation (φ.sub.q*)) can be automatically estimated (as the person (p) (and optionally the hearing aid orientation (φ.sub.q*)) associated with the directory Δ.sub.p, from which the location (θ.sub.j*) having the lowest cost function originates).

[0173] The above procedure may be used to determine each of the data points in the learning mode of FIGS. 5A and 5B. The learning mode of operation may e.g. be entered during or after each power-up of the hearing aid system (or on request, e.g. from a user interface).

[0174] After the learning mode has been finalized, the person (p**) (FIG. 5A) and optionally the hearing aid orientation (φ.sub.q**) (FIG. 5B)) associated with the directory Δ.sub.p most frequently used for estimating the current location (θ*) of the target signal source may be held fixed, at least for a certain amount of time, thereby simplifying the procedure of estimating a current location (or rather the acoustic transfer functions corresponding to the current electric input signals) to a single directory Δ.sub.p** (see ‘Normal mode’ for t>t1 in the right parts of FIG. 5A, 5B).

[0175] It is intended that the structural features of the devices described above, either in the detailed description and/or in the claims, may be combined with steps of the method, when appropriately substituted by a corresponding process.

[0176] As used, the singular forms “a,” “an,” and “the” are intended to include the plural forms as well (i.e. to have the meaning “at least one”), unless expressly stated otherwise. It will be further understood that the terms “includes,” “comprises,” “including,” and/or “comprising,” when used in this specification, specify the presence of stated features, integers, steps, operations, elements, and/or components, but do not preclude the presence or addition of one or more other features, integers, steps, operations, elements, components, and/or groups thereof. It will also be understood that when an element is referred to as being “connected” or “coupled” to another element, it can be directly connected or coupled to the other element but an intervening element may also be present, unless expressly stated otherwise. Furthermore, “connected” or “coupled” as used herein may include wirelessly connected or coupled. As used herein, the term “and/or” includes any and all combinations of one or more of the associated listed items. The steps of any disclosed method is not limited to the exact order stated herein, unless expressly stated otherwise.

[0177] It should be appreciated that reference throughout this specification to “one embodiment” or “an embodiment” or “an aspect” or features included as “may” means that a particular feature, structure or characteristic described in connection with the embodiment is included in at least one embodiment of the disclosure. Furthermore, the particular features, structures or characteristics may be combined as suitable in one or more embodiments of the disclosure. The previous description is provided to enable any person skilled in the art to practice the various aspects described herein. Various modifications to these aspects will be readily apparent to those skilled in the art, and the generic principles defined herein may be applied to other aspects.

[0178] The claims are not intended to be limited to the aspects shown herein but are to be accorded the full scope consistent with the language of the claims, wherein reference to an element in the singular is not intended to mean “one and only one” unless specifically so stated, but rather “one or more.” Unless specifically stated otherwise, the term “some” refers to one or more.

REFERENCES

[0179] [Farina, 2000]: Farina, Angelo. “Simultaneous measurement of impulse response and distortion with a swept-sine technique.” Audio Engineering Society Convention 108. Paper 5093, Audio Engineering Society, Feb. 1, 2000. [0180] EP3413589A1 (Oticon) Dec. 12, 2018 [0181] EP3253075A1 (Oticon) Jun. 12, 2017 [0182] EP2928214A1 (Oticon) Jul. 10, 2015