Abstract
A hearing device, e.g. a hearing aid, comprises a) an input unit comprising a multitude of input transducers for providing respective electric input signals representing sound in an environment of the user; b) an output unit comprising an output transducer for providing stimuli perceivable to the user as sound based on said electric input signals or a processed version thereof; c) first and second spatial filters each connected to said input unit and configured to provide respective first and second spatially filtered signals based on said multitude of electric input signals and configurable beamformer weights. The first spatial filter implements at a given time, a feedback cancelling beamformer, or a target maintaining, noise cancelling, beamformer directed at said environment of the user. The second spatial filter implements at a given time, a feedback cancelling beamformer, or an own voice beamformer directed at the mouth of the user.
Claims
1. A hearing device configured to be located at or in an ear of a user, the hearing device comprising: an input unit including a multitude of input transducers for providing respective electric input signals representing sound in an environment of the user; an output unit including an output transducer for providing stimuli perceivable to the user as sound based on said electric input signals or a processed version thereof; and first and second spatial filters each connected to said input unit and configured to provide respective first and second spatially filtered signals based on said multitude of electric input signals and configurable beamformer weights, wherein the first spatial filter, at a given time, implements a feedback cancelling beamformer, or a target maintaining noise cancelling beamformer, directed at said environment of the user, the second spatial filter, at a given time, implements a feedback cancelling beamformer, or an own voice beamformer, directed at the mouth of the user, and the second spatial filter is controlled by an own voice presence control signal, and/or a far-end talker presence control signal, and/or a telephone mode control signal.
2. A hearing device according to claim 1, configured to operate in a number of modes including a communication mode.
3. A hearing device according to claim 2, configured to determine or select the beamformer weights in dependence of a mode of operation of the hearing device.
4. A hearing device according to claim 2, wherein said communication mode comprises a telephone mode.
5. A hearing device according to claim 4, configured to provide than, in said telephone mode of operation, the user's own voice is picked up by the input transducers and spatially filtered by the own voice beamformer providing the second spatially filtered signal, which is fed to a transmitter of the hearing device and transmitted to a telephone.
6. A hearing device according to claim 5, configured to provide that said signal from said telephone is mixed with the first spatially filtered signal from the environment in a combination unit and that the mixed signal is fed to said output transducer for presentation to the user as sound.
7. A hearing device according to claim 4, configured to provide that, in said telephone mode of operation, a signal is received from said telephone by a receiver of the hearing device.
8. A hearing device according to claim 7, further comprising: a separate voice detector coupled to the receiver to decide on whether the signal from the telephone contains speech.
9. A hearing device according to claim 8, configured to control said fading of the second spatial filter in dependence of the separate voice detector.
10. A hearing device according to claim 8, configured to control said fading of the second spatial filter in dependence of the own voice detector and the separate voice detector.
11. A hearing device according to claim 4, configured to provide that, in said telephone mode of operation, the second spatial filter is adapted to fade between A) the own voice beamformer adapted to pick up the user's voice while cancelling noise from the surroundings, when the hearing device user is talking, and B) the feedback cancelling beamformer, when the far-end user is talking.
12. A hearing device according to claim 11, configured to control said fading of the second spatial filter in dependence of the own voice presence control signal.
13. A hearing device according to claim 4, configured to provide that, in said telephone mode of operation, the second spatial filter implements the feedback cancelling beamformer.
14. A hearing device according to claim 4, configured to provide that, in said telephone mode of operation, the second spatial filter is adapted to fade between A) the own voice beamformer adapted to pick up the user's voice while cancelling noise from the surroundings, when the hearing device user is talking, and B) the feedback cancelling beamformer, when the hearing device user is not talking.
15. A hearing device according to claim 4, configured to provide that, in said telephone mode of operation, the second spatial filter is adapted to fade between A) the own voice beamformer adapted to pick up the user's voice while cancelling noise from the surroundings, when the far-end user is not talking, and B) the feedback cancelling beamformer, when the far-end user is talking.
16. A hearing device according to claim 1, further comprising: an own voice detector for estimating whether or not, or with what probability, a given input sound originates from the voice of the user of the system.
17. A hearing device according to claim 1, further comprising: a mode indicator providing a mode control signal indicating whether or not the hearing device is in the telephone mode of operation.
18. A hearing device according to claim 1, containing two input transducers.
19. A hearing device according to claim 1, wherein the output transducer is, or comprises, a loudspeaker or a vibrator of a bone conduction hearing device.
20. A hearing device according to claim 1, wherein said input unit comprises respective filter banks configured to provide said electric input signals in a time-frequency representation (k,m), where k and m are frequency and time indices, respectively.
21. A hearing device according to claim 1, configured to provide that said beamformer weights are frequency dependent.
22. A hearing device according to claim 1, configured to provide that said beamformer weights are adaptively determined.
23. A hearing device according to claim 1, wherein the hearing device is or comprises a hearing aid, a headset, an earphone, an ear protection device or a combination thereof.
Description
BRIEF DESCRIPTION OF DRAWINGS
(1) 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:
(2) FIG. 1A shows a first embodiment of a hearing device comprising a directional system comprising a multitude of input transducers according to the present disclosure;
(3) FIG. 1B a second embodiment of a hearing device comprising a directional system comprising two microphones according to the present disclosure (partly in the frequency domain);
(4) FIG. 2A shows a third embodiment of a hearing device comprising a directional system with two microphones according to the present disclosure wherein a compressor controls the gain of the system using the input levels from the microphones;
(5) FIG. 2B shows a fourth embodiment of a hearing device comprising a directional system with two microphones according to the present disclosure wherein a compressor controls the gain of the system using the input levels from the microphones (partly in the frequency domain);
(6) FIG. 3 schematically shows a fifth embodiment of a hearing device comprising a directional system with two microphones according to the present disclosure wherein the hearing device further comprises a feedback estimation and cancellation system;
(7) FIG. 4 shows a typical level compression curve characterized by providing relatively high gain at relatively low input levels and lower gain at higher input levels;
(8) FIG. 5 shows an example of a hearing device comprising a compressor for controlling the spatial filter controller and the hearing device gain unit based on the level of the resulting weighted combination of the input signals;
(9) FIG. 6A shows a first embodiment of a hearing device comprising three microphones located in an ITE part adapted for being located at or in an ear canal of the user;
(10) FIG. 6B shows a second embodiment of a hearing device comprising three microphones located in an ITE-part adapted for being located at or in an ear canal of the user;
(11) FIG. 6C shows an embodiment of a hearing device comprising two microphones located in an ITE-part adapted for being located at or in an ear canal of the user;
(12) FIG. 7A shows a first exemplary telephone mode use case of a hearing device according to the present disclosure;
(13) FIG. 7B shows a second exemplary telephone mode use case of a hearing device according to the present disclosure; and
(14) FIG. 8 shows an embodiment of an own voice beamformer, e.g. for the telephone mode illustrated in FIG. 7A, 7B.
(15) 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.
(16) 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
(17) 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.
(18) The electronic hardware may include microprocessors, microcontrollers, digital signal processors (DSPs), field programmable gate arrays (FPGAs), programmable logic devices (PLDs), gated logic, discrete hardware circuits, and other suitable hardware configured to perform the various functionality described throughout this disclosure. 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.
(19) The present application relates to the field of hearing devices, e.g. hearing aids, in particular to feedback management.
(20) In the present application, a spatial feedback system that cancels or attenuates the acoustical feedback from the vent or an acoustical leakage between the ear mould and the ear canal wall is disclosed. The spatial anti feedback is achieved by using the two microphones already present in a conventional directional ITE style HI. The conventional use of the two microphones is to spatially filter the external sounds from the environment in order to separate acoustical noise from wanted acoustical signals usually from the frontal direction. This spatial filtering is in this invention also used to attenuate the feedback from the vent or leakage without attenuating the wanted external acoustical sound signal. This is here termed spatial anti feedback.
(21) FIG. 1A shows an embodiment of a hearing device comprising a directional system according to the present disclosure. The hearing device (HD), e.g. a hearing aid, is configured to be located at or in an ear of a user, e.g. fully or partially in an ear canal of the user. The hearing device comprises an input unit comprising a multitude of input transducers (M1, . . . , MN) for providing respective electric input signals (IN1, IN2, . . . , INN) representing sound in an environment of the user. The hearing device further comprises an output unit comprising an output transducer (SP), here a loudspeaker, for providing stimuli perceivable to the user as sound based on said electric input signals or a processed version thereof. The hearing device further comprises a spatial filter (w1, w2, . . . , wN, CU) connected to the input unit and to the output unit, and configured to provide a spatially filtered signal (OUT) based on the multitude of electric input signals and configurable beamformer weights (w1p, w2p, . . . , wNp, where p is a beamformer weight set index). The spatial filter comprises weighting units (w1, w2, . . . , wN), e.g. multiplication units, each being adapted to apply respective beamformer weights (w1p, w2p, . . . , wNp) to the respective electric input signals (IN1, IN2, . . . , INN) and to provide respective weighted input signals (Y.sub.1, Y.sub.2, . . . , Y.sub.N). The spatial filter further comprises a combination unit (CU), e.g. a summation unit, for combining the weighted input signals to one or more spatially filtered signals, here one (signal OUT), which is fed to the output transducer (SP, possibly further processed before). The hearing device (HD) further comprises a spatial filter controller (SCU) configured to apply (at least) first and/or second different sets (p=1, 2) of beamformer weights (w1p, w2p, . . . , wNp) to said multitude of electric input signals (IN1, IN2, . . . , INN). The first set of beamformer weights (p=1) is applied to provide spatial filtering of sound from the output transducer (SP) (leaking back to the input transducers, cf. dashed arrows indicating feedback paths h1, h2, . . . , hN from the output transducer (SP) to each of the N input transducers (M1, M2, . . . , MN), respectively). The second set of beamformer weights (p=2) is applied to provide spatial filtering of an external sound field (e.g. from a sound source located in the acoustic far-field relative to the hearing device, cf. FIG. 6A, 6B, 6C). The hearing device further comprises a memory (MEM) accessible from the spatial filter controller (SCU). The spatial filter controller is configured to adaptively select an appropriate set of beamformer weights (signal wip) among two or more sets (p=1, 2, . . . ) of beamformer weights stored in the memory (including the first and second sets of beamformer weights). At a given point in time, adaptive selection of an appropriate set beamformer weights may e.g. be dependent of a current input level of one or more of the multitude of input signals or of a currently requested gain from a compressor, and/or of a currently estimated loop gain.
(22) FIG. 1B shows an embodiment of a hearing device comprising a directional system according to the present disclosure. The input unit comprises (e.g. contains only) two microphones (M1, M2) for converting sound from the environment to respective electric input signals IN1, IN2. In the embodiment of FIG. 1B, the processing of the forward path of the hearing device (from sound input to sound output) is, at least partly, conducted in the frequency domain. The input unit comprises respective filter banks (FB-A1, FB-A2) configured to provide the electric input signals (IN1, IN2) in a time-frequency representation (k,m), e.g. as digitized frequency sub-band signals (X.sub.1, X.sub.2), where k and m are frequency and time indices, respectively. The frequency sub-band electric input signals (X.sub.1, X.sub.2) are fed to the spatial filter (weighting units (w1, w2)) and to the spatial filter controller (SCU). Depending on the input signals (X.sub.1, X.sub.2), e.g. their level, and/or SNR, an appropriate set of beamformer filtering weights (wip) is selected at a given point in time from the memory (MEM) by the spatial filter controller (SCU) and applied to the respective weighting units (w1, w2), cf. signals w1p, w2p, thereby providing respective weighted input signals Y.sub.1, Y.sub.2. The weighted input signals Y.sub.1, Y.sub.2. are added by the SUM unit (‘+’) to provide spatially filtered (beamformed) signal Y.sub.BF. The hearing device further comprises a synthesis filter bank (FB-S) for converting spatially filtered frequency sub-band signal YBF to spatially filtered time domain signal OUT, which is again fed to the loudspeaker (SP) for conversion to acoustic stimuli.
(23) The Spatial filter controller (SCU), is configured to apply different filter weights, w1p and w2p, to the two microphone channels, in order to either do spatial anti-feedback or to do spatial filtering of the external sound field (e.g. a first set (p=1) of beamformer weights (w11, w21) for spatially filtering the sound field from the loudspeaker (SP)), and a second set (p=2) of beamformer weights (w12, w22) for spatially filtering the external sound field from sound sources in the environment around the user (not originating from the loudspeaker of the hearing device).
(24) The acoustical feedback can be very unpredictable especially if the feedback is dominated by a leakage. It is therefore an advantage to individually calibrate the spatial anti feedback on the user's ear. This can be achieved by making an estimate of the feedback path using a conventional adaptive feedback path estimation (cf. e.g. FIG. 3) and then use the difference in the estimated feedback paths to generate a set of filter weights, w1 and w2, to achieve the spatial anti feedback. Alternatively, the filter weights could also be achieved by making an adaptive system that minimizes the output of the directional unit (output=s1*w1+s2*w2), while playing out a signal that will ensure that the input on the microphones is dominated by a feedback signal. The filter weights may alternatively or additionally be estimated from an on-line feedback path estimate.
(25) One problem with reusing the two microphones is that it is difficult to achieve both a spatial filtering of the external sounds and on the same time do spatial anti feedback (when only two microphones are available). This invention presents two ways of solving this problem. First by making the system adaptive using the input level and second to make the system work in separate frequency bands.
(26) A conventional HI uses dynamic range compression (compressive amplification) in order to use the limited dynamic range of the users' hearing. This means that the gain in the HI is higher at low input levels and lower at higher input levels. By making the spatial anti feedback adaptive using the input level (or a signal derived from the input level (such as e.g. the applied gain)), the system can use the spatial anti-feedback at low input levels where the gain of the instrument is higher and hence the problem with feedback is also higher. In situations with low input level there is usually not a need for spatial filtering of the external sound field.
(27) FIG. 2A shows an embodiment of a hearing device comprising a directional system with two microphones according to the present disclosure wherein a compressor controls the gain of the system using the input levels from the microphones. The embodiment of FIG. 2A is equivalent to the embodiment of FIG. 1A apart from the following differences. The embodiment of a hearing device of FIG. 2A comprises only two input transducers (microphones (M1, M2)), but additionally comprises a compressor (COMP) comprising a compressive amplification algorithm for determining an input level dependent (requested) gain in dependence of a user's needs (e.g. hearing impairment) and the current input level. Based thereon, a weight control signal Wctr is fed to the spatial filter controller (SCU), for controlling the currently selected set of beamformer weights wip, i=1, 2, p=1,2 according to a current input level of the electric input signals IN1, IN2 of the requested gain (derived from the compressive amplification algorithm adapted to the user's needs). The hearing device (HD) further comprises a processor (HAG) for further processing the spatially filtered signal Y.sub.BF and provide processed signal (OUT), which is fed to the output transducer (SP). The compressor (COMP) is further configured to feed gain control signal (HAGctr) to the processor (HAG) to allow the processor to apply a relevant gain to the spatially filtered signal Y.sub.BF (in dependence of the input level(s) or the (requested) gain derived therefrom).
(28) FIG. 2B shows an embodiment of a hearing device comprising a directional system with two microphones according to the present disclosure wherein a compressor controls the gain of the system using the input levels from the microphones (partly in the frequency domain). The embodiment of FIG. 2B is equivalent to the embodiment of FIG. 2A apart from the following difference. The embodiment of a hearing device of FIG. 2B comprises appropriate analysis and synthesis filter banks (FB-A1, FB-A2, and FB-S, respectively) to allow processing of the forward path (and analysis part (SCU, COMP, MEM)) to be conducted in the frequency domain (separate processing of individual frequency sub-band signals). In the embodiment of FIG. 2B, the processor (HAG) for further processing the spatially filtered signal Y.sub.BF and provide processed signal Y.sub.G, which is then fed to synthesis filter banks (FB-S) providing processed time domain output signal OUT, which is fed to the loudspeaker (SP).
(29) The input level or the compression level may be used as input to the Spatial filter controller (SCU), in order to switch between spatial anti feedback (first) beamformer weights and conventional (second) directional beamformer weights.
(30) In a situation where the input level from the external sound field is relatively high (e.g. >70 dB SPL) and where the background noise is relatively high, spatial filtering of the external sound field can be activated, and at these high input levels the compression will lower the gain, and the spatial anti-feedback system can be deactivated.
(31) The limit for when the spatial anti-feedback can be deactivated is determined by loop gain. Spatial anti-feedback may be deactivated, when loop gain is low enough for the system to operate without the spatial anti-feedback. Typically, this is when the loop gain (loop magnitude) is lower than 0 dB, but it may depend on how well possible other anti-feedback measures in the HI are working (e.g. feedback cancellation system where an estimate of the feedback path is subtracted from an electric input signal, cf. e.g. FIG. 3).
(32) Estimates of the feedback paths from the output to the input transducers may be provided by several means, e.g. by respective adaptive filters as indicated in FIG. 3. The feedback estimates may be used in the spatial filter controller (SCU) to contribute to the decision of whether to apply the first or second set of beamformer weights at a given point in time (cf. dashed arrows in FIG. 3 feeding feedback estimates EST1, EST2 to the combined spatial filter controller and compressor (SCU-COMP)).
(33) FIG. 3 schematically shows an embodiment of a hearing device comprising a directional system with two microphones according to the present disclosure wherein the hearing device further comprises a feedback estimation and cancellation system. The embodiment of FIG. 3 is equivalent to the embodiment of FIG. 2B apart from the following difference. The hearing device (HD) further comprises respective feedback cancellation systems for estimating and reducing feedback from the output transducer (here loudspeaker (SP)) to first and second input transducers (here microphones (M1, M2)), respectively. The first and second feedback cancellation systems comprises first and second feedback estimators (FBE1 FBE2) and subtraction units (‘+’) inserted in the respective microphone paths so subtract respective estimates (EST1, EST2) of the feedback paths (h1, h2) from the input signals (IN1, IN2). The subtraction units provide respective feedback corrected input signals (ER1, ER2), which are fed to the respective analysis filter banks (FB-A1, FB-A2) and to the respective feedback estimators (FBE1, FBE2). The feedback estimators (FBE1, FBE2) each comprises respective algorithm (ALG1, ALG2) and variable filter parts (FIL1, FIL2) implementing respective adaptive filters (where the algorithm parts (ALG1, ALG2) are configured to determine (and update) filter coefficients of the variable filter parts (FIL1, FIL2) via respective update signals (UP1, UP2). The adaptive filters ((ALG1, FIL1), (ALG2, FIL2)) are e.g. state of the art adaptive filters. The algorithm parts (ALG1, ALG2) may e.g. comprise Least Mean Square (LMS) or Normalized LMS (NLMS) algorithms or similar adaptive algorithms to estimate filter the coefficients (based on reference signal OUT and respective error signals (ER1, ER2)) that when applied to the variable filters for filtering the processed output (reference) signal OUT, thereby providing respective feedback estimates (EST1, EST2), minimizes the respective error signals (ER1, ER2). The feedback estimates (EST1, EST2) may be fed to the spatial filter controller (SCU, here the combined SCU-COMP-unit), for controlling the currently selected set of beamformer weights. Likewise first and second algorithm control signals (A1ctr, A2ctr) may be generated in the combined spatial filter controller and compressor (SCU-COMP) and fed to the respective feedback estimators (FBE1, FBE2), e.g. to control an adaptation rate of the adaptive algorithm, and or an update rate or time of updating the filter coefficients in the variable filter (e.g. including disabling or enabling such update of filter coefficients).
(34) FIG. 4 shows a typical level compression curve (gain G [dB] versus input level L [dB SPL]) characterized by providing relatively high gain (HG) at relatively low input levels (L<KP1) and lower gain (LG) at higher input levels (L>KP2). The graph illustrates that at low input levels (e.g. L<L.sub.TH or <KP1) the spatial anti feedback setup of the directional system (first beamformer weights) may advantageously be used (cf. indication ‘Spatial filtering of feedback sound field’), and at higher input levels (e.g. L>L.sub.TH or >KP2) the spatial filtering of the external sounds (second beamformer weights) may advantageously be used (cf. indication ‘Spatial filtering of external sound field’). In the exemplary embodiment of FIG. 4, a threshold level L.sub.TH (KP1<L.sub.TH<KP2) located between the first and second knee points forms the border between using the first and second sets of beamformer weights. The threshold level L.sub.TH may be predetermined, e.g. with a view to the user's hearing profile (e.g. an audiogram, and/or a level sensitivity). The threshold level L.sub.TH may be adaptively determined (cf. dashed double arrow denoted ‘adaptive’ in FIG. 4), e.g. in dependence of a current signal to noise ratio (SNR). The threshold level L.sub.TH may be adaptively determined, e.g. in dependence of a current signal to noise ratio (SNR) and a current requested gain (or input level). The threshold level L.sub.TH may increase with increasing SNR (e.g. within limits minimum and maximum values, L.sub.TH,min and L.sub.TH,max, of the input level). The threshold level L.sub.TH may increase with increasing SNR for relatively low input levels (high gains), for input levels below a predefined threshold level.
(35) The spatial filter controller (SCU) is configured to apply that the first and/or second different sets of beamformer weights to the multitude of electric input signals in dependence of the estimated input level(s) (or the requested gains determined therefrom by a compressive amplification algorithm). In an embodiment, the application of a given set of beamformer weights is further dependent of the current signal to noise ratio (SNR) of the electric input signal(s) or a signal derived therefrom.
(36) If, for example, the electric input signal(s) have a relatively high SNR, and a relatively low gain (high level), there is no need for noise reduction (e.g. provided by the second beamformer weights handling signals from the acoustic far-field), so the first beamformer weights (providing spatial feedback attenuation) can advantageously be applied.
(37) To avoid fluctuations between the two types of directional settings, hysteresis may be built into the decision. In an embodiment, for increasing levels, the switching from the first to the second beamformer weights occur when L becomes larger than KP1+ΔL1 (where ΔL1≤(KP2−KP1)), and so that for decreasing levels, the switching from the second to the first beamformer weights occur when L becomes smaller than KP2−ΔL2 (where ΔL2≤(KP2−KP1)). Alternatively fading between the two sets of beamformer weights may be introduced when input levels are between the two knee points (KP1<L<KP2).
Frequency Bands
(38) The system described above can be designed to work in separate frequency bands, meaning for example that the spatial anti feedback is only active in frequency bands where feedback is a problem (e.g., between 1 kHz and 8 kHz, or between 1 kHz and 4 kHz). Additionally, the adaptive system described above can also be applied separately in frequency bands, meaning that the shift from spatial anti feedback to spatial filtering of the external sound field is only active in the frequency bands where the compression has lowered the gain enough for the system or work without the spatial anti feedback and/or where the spatial filtering of the external sound field is wanted. In an embodiment, only one of the first and second sets of beamformer weights is applied at a given time, in a given frequency band. In an embodiment, the first set of beamformer weights is applied in at least one frequency band, while the second set of beamformer weights is applied in another frequency band at the same time.
(39) FIG. 5 shows an example of a hearing device comprising a compressor (COMP) for controlling the spatial filter controller (SCU) and the hearing device gain unit (HAG) based on a level of the resulting weighted combination of the input signals (beamformed signal Y.sub.BF). The embodiment of a hearing device (HD) of FIG. 5 is equivalent to the embodiment of FIG. 2A apart from the following differences. The embodiment of a hearing device of FIG. 5 comprises signal to noise ratio and level estimators (SNR and LD, respectively) for providing estimates of an SNR and a level of an incoming signal, here the spatially filtered (beamformed) signal Y.sub.BF. Instead of analysing the first and second electric input signals (IN1, IN2) (as in FIG. 2A), the compressor (COMP) of the embodiment of FIG. 5 receives current estimates of the level of the beamformed signal Y.sub.BF. Further, a current SNR (signal snr) of the spatially filtered signal Y.sub.BF is provided to the spatial filter controller (SCU) by the SNR estimator (SNR) together with a requested gain RG provided by the compressor (COMP) and the current estimate of the level IL of the spatially filtered signal Y.sub.BF. The requested gain RG is determined by the compressor (COMP) based on the input level IL of the beamformed signal YBF (as e.g. indicated in FIG. 4, e.g. individually (differently) for a given frequency band). Based thereon, the spatial filter controller (SCU) determines the appropriate set of beamformer weights (wip=w1p, w2p) (as e.g. discussed in connection with FIG. 4) and reads this set out of the memory unit (MEM) using control signal Wctr. The spatial filter controller (SCU) applies appropriate set of beamformer weights (wip=w1p, w2p) to the spatial filter (BFU).
(40) In the embodiment of FIG. 5, levels as well as SNR are estimated based on the beamformed signal Y.sub.BF. One or both parameters (level and SNR) can be estimated in various ways, e.g. based on one or more of the electric input signals (IN1, IN2).
(41) In an embodiment, level and SNR are estimated directly from the electric input signals (IN1, IN2). This may be advantageous, because level and SNR may change if the beamformer changes.
(42) FIG. 6A shows an embodiment of a hearing device comprising an ITE part adapted (ITE) for being located at or in an ear canal (Ear canal) of the user. The ITE part may e.g. constitute the hearing device, or it may form part of a hearing device further comprising one or more portable parts, e.g. including a BTE part configured to be worn at or behind the ear (pinna), and operationally connected to the ITE-part via an acoustic or electric or electromagnetic (e.g. optic) connection. The ITE-part comprises a housing (Housing (mould) in FIG. 6A), which may be customized to a particular user's physiognomy (ear, and/or ear canal) or it may be a standard part (‘one-size-fits-all’) intended to be used by a group of customers.
(43) The ITE-part (ITE) comprises a vent channel (or a number of vent channels), in FIG. 6A indicated by a single through-going straight opening (Vent). The vent channel may take on different forms, be it in cross-section of longitudinal extension through the housing of the ITE-part. It may further be distributed on a number of separate venting channels, one or more of which may be formed as through going openings or as indentations in the surface of the housing (forming a channel with a wall (Skin/Tissue) of the ear canal), cf. also Skin-housing leakage channel in FIG. 6A (which may be intentional or un-intentional).
(44) The hearing device (here the ITE-part) comprises three input transducers (here microphones M1, M2, M3, providing respective (e.g. digitized) electrical input signals (possibly as frequency sub-band signals) electrically connected to spatial filter and controller (BF-CNT) providing a spatially filtered (beamformed) signal (e.g. Y.sub.BF in FIG. 5) to a processor (HAG) for applying an appropriate gain according to a user's needs in dependence of the acoustic environment (Environment), as reflected by sound filed S.sub.ENV and electric input signals picked up by the microphones), and providing a processed signal (e.g. Y.sub.G in FIG. 5). The processed signal is fed to an output transducer (here a loudspeaker (SP)) and presented to the user as audible signals (here via sound field S.sub.ED crating vibrations of air in the residual volume (Residual volume) in the ear canal (Ear canal) between the housing of the ITE-part and the ear drum (Ear drum). The spatial filter and controller (BF-CNT) is configured to apply an appropriate set of beamformer weights to the three electric input signals and provide a corresponding spatially filtered signal as proposed by the present disclosure. The set of beamformer weights is selected in dependence of the input level and or requested gain (and thus hearing profile of the user) and possibly other properties of the input signals (e.g. a target signal to noise ratio).
(45) The hearing device may comprise fewer ore more input transducers (e.g. microphones) than three. Some of the microphones may be located in other parts of the hearing device (possibly in concha or elsewhere at or around an ear of the user (e.g. in a BTE part adapted be being arranged at or behind pinna). In an embodiment, one of the microphones is located on or close to the a part of the surface of the ITE part facing the residual volume and ear drum, e.g. to measure or monitor the sound field in the residual volume (e.g. for active noise cancellation, etc.).
(46) The three microphones of the embodiment of FIG. 6A are shown to be located on/or close to a part of the surface of the ITE part facing the environment (opposite the residual volume and ear drum), e.g. mounted on a faceplate of an ear mould. In an embodiment, at least one of the microphones is located along a longitudinal axis of the hearing device in a direction towards the ear drum (to create a microphone axis towards the eardrum). Thereby spatial separation of sound from the outside (environment) and from the inside (residual volume) is facilitated, including spatial filtering of sound from the output transducer (loudspeaker (SP). Such embodiments are shown in FIG. 6B, 6C.
(47) FIG. 6B shows an embodiment of a hearing device according to the present disclosure comprising three microphones located in an ITE-part adapted for being located at or in an ear canal of the user. The embodiment of a hearing device (HD) of FIG. 6B comprises three microphones (M1, M2, M3) in an ITE-part. Two of the microphones (M1, M2) face the environment, and one microphone (M3) faces the ear drum (when the hearing device is operationally mounted). The hearing device comprises, or is constituted by, the ITE-part. The ITE-part may comprise a sealing element for providing a tight seal (cf. ‘seal’ in FIG. 6B) towards the walls of the ear canal to acoustically ‘isolate’ the ear drum facing microphone (M3) from the environment sound (S.sub.ITE) impinging on the ear canal (and hearing device), cf. FIG. 6B. In an embodiment, the fitting is more open to allow environment sound to reach the microphone (M3) facing the ear drum. The hearing device (HD) may comprise the same functional elements as the embodiments of FIG. 1A, 1B, 2A, 2B, 3, 5, 6A, 7A.
(48) FIG. 6C shows an embodiment of a hearing device (HD), e.g. a hearing aid, comprising two microphones (M1, M2) located in an ITE-part according to the present disclosure. The ITE-part comprises a housing, wherein the two ITE-microphones are located (e.g. in a longitudinal direction of the housing along an axis of the ear canal (cf. dotted arrow ‘Inward’ in FIG. 6C), when the hearing device (HD) is operationally mounted on or at the user's ear. The ITE-part further comprises a guiding element (‘Guide’ in FIG. 6C) configured to guide the ITE-part in the ear canal during mounting and use of the hearing device (HD) without fully blocking the ear canal (to avoid occlusion, and to allow environment sound (from sound field S.sub.ITE) to reach the microphone (M2) closest to the ear drum.). The ITE-part further comprises a loudspeaker (facing the ear drum) for playing a resulting audio signal to the user, whereby a sound field is generated in the residual volume. A fraction thereof is leaked back towards the ITE-microphones (M1, M2) and the environment. The hearing device (e.g. the ITE-part) may constitute a part customized to the ear or the user, e.g. in form, or alternatively have a standardized form. The hearing device (HD) may comprise the same functional elements as the embodiments of FIG. 1A, 1B, 2A, 2B, 3, 5, 6A, 7A, 7B.
(49) FIGS. 7A and 7B illustrates an exemplary telephone mode of a hearing device (HD) according to the present disclosure. In this application, we may both aim at spatially reducing feedback in the beamformer signal presented locally and the beamformer signal presented to the far-end speaker of a telephone conversation.
(50) FIG. 7A shows an embodiment of a hearing device (HD) comprising two microphones (M1, M2) to provide electric input signals IN1, IN2 representing sound in the environment of a user wearing the hearing device. The hearing device further comprises spatial filters DIR and Own Voice DIR, each providing a spatially filtered signal (ENV and OV respectively) based on the electric input signals. The spatial filter DIR may e.g. implement a first, feedback cancelling, and/or second, target maintaining, noise cancelling, beamformer according to the present disclosure. The spatial filter Own Voice DIR is a spatial filter according to the present disclosure. The spatial filter Own Voice DIR implements a first, feedback cancelling, and/or a second, own voice, beamformer directed at the mouth of the user (its activation being e.g. controlled by an own voice presence control signal, and/or a telephone mode control signal, and/or a far-end talker presence control signal). In a specific telephone mode of operation, the user's own voice is picked up by the microphones M1, M2 and spatially filtered by the own voice beamformer of spatial filter Own Voice DIR providing signal OV, which is fed to transmitter Tx and transmitted (by cable or wireless link to a telephone (cf. dashed arrow denoted ‘To phone’ and telephone symbol). In the specific telephone mode of operation, signal PHIN is received by (wired or wireless) receiver Rx from a telephone (as indicated by telephone symbol and dashed arrow denoted ‘From Phone’). When a far-end talker is active, signal PHIN contains speech from the fare-end talker, e.g. transmitted via a telephone line (e.g. fully or partially wirelessly, but typically at least partially cable-borne). The ‘far-end’ telephone signal PHIN is mixed with the environment signal ENV from the spatial filter DIR in combination unit (here sum unit) ‘+’, and the mixed signal OUT is fed to output transducer SP (e.g. a loudspeaker or a vibrator of a bone conduction hearing device) for presentation to the user as sound.
(51) FIG. 7B is identical to FIG. 7A except that the feedback path during activation of the own voice beamformer during a telephone conversation is indicated in FIG. 7B (in bold dashed line denoted FB.sub.FEOV).
(52) At the own voice beamformer (provided by the Own Voice DIR unit), we do not have feedback (like a closed form loop), but we may have an echo problem as part of the external signal that is picked up by the own voice beamformer, and transmitted back to the far-end talker. This may be the case when the far-end talker is active (cf. encircled digit ‘1’ in FIG. 7B), in which case the voice of the far-end talker is played by the loudspeaker (SP) of the hearing device (HD) (cf. encircled digit ‘2’). Via feedback paths FB1, FB2 (commonly denoted FB in FIG. 7B) the voice of the far-end talker is picked up by the microphones (M1, M2) (cf. encircled digit ‘3’). The two electric input signals are combined in the Own Voice DIR unit (in a normal own voice mode of operation) to own voice signal OV (cf. encircled digit ‘4’). The ‘own voice signal’ OV may not contain the hearing device user's voice, because he or she will probably be silent, when the far-end talker is active. The ‘own voice signal’ OV may, on the other hand, contain a certain fraction of the far-end talker's voice. If the latter is the case, the far-end talker's voice eventually reaches the far-end talker (again) after transmission (by transmitter Tx, e.g. via a local telephone and a PSTN) to ‘the other end’ (cf. encircled digit ‘5’) as an un-desired echo. In that case too, it would be desirable to fade between an own voice beamformer adapted to cancel noise from the surroundings (when the hearing device user is talking) and a feedback cancelling beamformer (when the far-end user is talking) (far-end echo illustrated by the dashed bold line denoted FB.sub.FEOV and encircled digits 1-5).
(53) Switching (fading) between the first (feedback cancelling) beamformer and the second (own voice, environment noise reducing) beamformer (of the Own Voice DIR) may e.g. be controlled by a voice detector capable of detecting the own voice of the user of the hearing device together with a mode control signal indicating whether or not the hearing device is in a telephone mode of operation. If this is the case, a switching (or fading) of the Own Voice DIR unit between the (second) own voice beamformer and the (first) feedback cancelling beamformer may be made in dependence of whether or not the own voice detector detects the own voice of the user of the hearing device (assuming that the user and the far-end talker are not (generally) talking at the same time). In an embodiment, the hearing device comprises a separate voice detector coupled to the receiver (Rx) to decide on whether the signal from the far-end contains speech (or any other detector indicating voice activity of a far-end talker). This speech detector may then (alternatively) be used to switch between the two beamformers of the Own Voice DIR unit (under the same assumption of non-simultaneous speaking). The hearing device may contain an own voice detector (e.g. connected to one of the electric input signals (IN1, IN2), or the own voice signal OV) as well as a speech detector (e.g. connected to the receiver Rx or the combination unit ‘+’ based on the output signal (OUT)) to detect far-end speech, and let the combined result of the two detectors control the switching between the two beamformers.
(54) FIG. 8 shows an embodiment of an own voice beamformer, e.g. for the telephone mode illustrated in FIG. 7A, 7B, implemented using the configuration comprising two microphones. FIG. 8 shows an own voice beamformer according to the present disclosure illustrating how the own voice-enhancing post filter (OV-PF) gains (G.sub.OV,1(k) and G.sub.OV,2(k) of FIG. 8B) may be estimated. The own voice gains are determined on the basis of a current noise estimate, here provided by a combination of an own voice cancelling beamformer (C.sub.2(k)), defined by (frequency dependent, cf. frequency index k) complex beamformer weights (w.sub.ov_cncl_1(k), w.sub.ov_cncl_2(k)) and another beamformer (C.sub.1(k), here an omni-directional beamformer), defined by complex beamformer weights (w.sub.ov1(k), w.sub.ov2(k)) containing the own voice signal. In an embodiment, the own voice enhancing beamformer is adaptive. A direction from the user's mouth, when the hearing device is operationally mounted is schematically indicated (cf. solid arrow denoted ‘Own Voice’ in FIG. 8). Correspondingly, a direction from an external sound source is schematically indicated in FIG. 8 shows a (possibly adaptive) beamformer configuration, wherein post filter gains (PF gain), G.sub.OV,1(k) and G.sub.OV,2(k), are determined (cf. output of OV-PF-block) and applied to respective input signals X.sub.1(k) and X.sub.2(k) in respective multiplication units (‘X’). The resulting signals (G.sub.OV,1(k) X.sub.1(k) and G.sub.OV,2(k) X.sub.2(k), respectively) are added in sum unit (‘+’) to provide the own voice estimate Y.sub.OV(k). The own voice estimate (Y.sub.BF, OV in FIG. 7A, 7B) may (e.g. an own-voice mode of operation, e.g. when a connection to a telephone or other remote device is established (cf. e.g. FIG. 7A, 7B)) be transmitted to a remote device via a transmitter (cf. e.g. Tx in FIG. 7A, 7B), (e.g. to a far-end listener of a telephone, cf. FIG. 7A, 7B). In the ‘own voice mode’, noise from external sound sources may be reduced by the beamformer.
(55) A binaural hearing system comprising first and second hearing devices (e.g. hearing aids) as described above may be provided. The first and second hearing devices may be configured to allow the exchange of data, e.g. audio data, and with another device, e.g. a telephone, or a speakerphone, a computer (e.g. a PC or a tablet). Own voice estimation may be provided based on signals from microphones in the first and second hearing devices. Own voice detection may be provided in both hearing devices. A final own voice detection decision may be based on own voice detection values from both hearing devices or based on signals from microphones in the first and second hearing devices.
(56) 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.
(57) 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.
(58) 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.
(59) The claims are not intended to be limited to the aspects shown herein, but is 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.
(60) Accordingly, the scope should be judged in terms of the claims that follow.
REFERENCES
(61) EP3267697A1 (Oticon) Oct. 1, 2018
