Hearing device with active noise cancellation

Abstract

The present application discloses a hearing device with active noise cancellation, comprising: a microphone system comprising two microphones disposed outside the ear canal of a wearer when the hearing device is worn, where the microphone system is configured to capture audio and to generate an audio signal based on the captured audio; an ambient sound detector configured to receive the audio signals, the ambient sound detector further configured to determine a classification of an ambient sound environment based on the audio signals, where the classification includes determining a coherence metric between at least two audio signals; and an active noise control system configured to receive the audio signal, the active noise control system configured to generate a feed-forward compensation signal based on the audio signal; wherein the hearing device is configured to mix an audio signal with a proportion of the generated feed-forward compensation signal according to a determined classification of the ambient sound environment.

Description

Technical Field

[0001] The present application relates to a hearing device with active noise cancellation. More specifically, the present application relates to a method of controlling active noise cancellation in a device with first and second microphones. Background Art

[0002] US 10,586,523 B1 discloses the use of a wind noise detector to control the feed-forward gain of an ANC system. However, this has several limitations.

[0003] Therefore, there is a need to provide a solution that solves at least some of the above-mentioned problems, or at least provides an alternative to the prior art. Summary of the invention

[0004] According to one aspect, the present invention provides a method for operating a hearing device including an active noise control (ANC) system, the hearing device being configured to be placed at the ear of a wearer, the hearing device comprising a microphone system including two microphones. Advantageously, the two microphones will be arranged outside the ear canal of the wearer when the hearing device is worn. The method is useful for use in a hearing aid. The method may include capturing audio with the microphone system. The microphone system may be configured as a directional microphone system, or configured to provide a signal enabling the establishment of a directional signal. The method may include generating an audio signal based on the captured audio. The audio signal originates from an input system including one or more input transducers. The signal from the aforementioned input transducer is converted into a processable signal, such as a time-frequency unit. This may be or include an analog-to-digital conversion. The method may include providing the audio signal to an ambient sound detector. The ambient sound detector may include one or more level detectors, one or more frequency analyzers, etc. The method may include determining a classification of an ambient sound environment based on the audio signal using the ambient sound detector, wherein the classification may include determining a coherence measure between at least two audio signals. The method may include providing the audio signal to the ANC system, the ANC system generating a feed-forward compensation signal based on the audio signal. The method may include mixing the audio signal with a proportion of the generated feed-forward compensation signal based on the determined classification of the sound environment. Mixing the feed-forward compensation signal based at least in part on the ambience improves noise control in the signal provided to the user.

[0005] Advantageously, the microphone system may be configured such that the first microphone is located at a position outside the ear canal and the second microphone is located at a position inside the ear canal.Such a configuration is expected to improve the performance of the noise control system.

[0006] Advantageously, the method may further comprise that said classification comprises or is a measure of ambientness.Looking at ambientness in comparison to a measure of, for example, wind noise has been shown to improve the performance of noise control systems.

[0007] The inventors have recognised that it is advantageous to control an active noise control system not based on noise sources that are not from the environment of the hearing device user, such noise sources include wind noise and handling noise. By using a coherence measure between (at least) two input transducers facing towards the environment, it is possible to achieve improved active noise control. It is particularly advantageous if the ambience is not based on or does not include any kind of estimate of an error or residual noise signal (e.g. from the ear canal of a user). Advantageously, when the hearing device or hearing aid is placed at the ear of the user, the ambience is determined based on signals from two or more input transducers located outside the ear canal of the user. Advantageously, when the hearing device or hearing aid is placed at the ear of the user, the ambience is determined based only on signals from two or more input transducers located outside the ear canal of the user. Advantageously, when the hearing device or hearing aid is placed at the ear of the user, the ambience is determined based on signals from two or more input transducers located outside the ear canal of the user, e.g. facing in a direction away from the user's head. Advantageously, the ambience is determined based on the ear canal input transducers without affecting the error signal.

[0008] Advantageously, the method may further comprise: the ambient degree is represented as a value in the interval [0:1].

[0009] Advantageously, the method may further comprise: the classification of the ambient sound environment is performed for frequencies below 3 kHz. Excluding higher frequencies may eliminate certain error sources and thus improve the performance of the noise control.

[0010] Advantageously, the method may further comprise that the classification of the ambient sound environment is based on or comprises a coherence measure determined for each of the plurality of frequency bands or a subset thereof.

[0011] Advantageously, the method may further comprise: the classification of the ambient sound environment is also based on one or more of: level, detection of the presence of own voice, coherence measure between at least one frequency band of each of the audio signals. Adding further parameters to the decision is expected to make the decision more accurate or at least more specific.

[0012] Advantageously, the method may further comprise that the coherence measure may be determined for at least two adjacent frequency channels or at least two non-adjacent frequency channels.If the coherence measure is determined across more than one frequency channel, it is expected that the coherence measure may be determined more accurately.

[0013] Advantageously, the method may further comprise that the classification of the ambient sound environment may be performed using a linear and / or non-linear combination of more than one input from the ambient sound detector.

[0014] Advantageously, the method may further comprise: the ambient sound detector may comprise a trained network trained using sounds from a first group classified as ambient sounds and comprising one or more of: background noise, speech, music and / or machine noise, and a second group classified as non-ambient sounds and comprising one or more of: own voice, processing noise, wind noise, low level sound.

[0015] Advantageously, the method may further comprise: the input of the trained network of the ambient sound detector is or comprises a coherence measure. In general, the hearing aid may comprise a neural network. The neural network may comprise a model layer for processing an electrical input signal or a signal based on the electrical input signal to provide a processed signal. The model layer may comprise an input layer, one or more intermediate layers, and an output layer. Each layer may comprise one or more nodes. The input layer of the neural network may be regarded as a layer in which data is fed into the neural network. Each node in the input layer may represent a feature of the input data. The one or more intermediate layers may perform a variety of different calculations and transformations on the input data to extract features and patterns. Each node in the one or more intermediate layers may receive one or more inputs from the previous layer, determine a weighted sum based on one or more inputs from the previous layer, add a bias to the weighted sum to determine a result, and then pass the result through an activation function to introduce nonlinearity. The output layer may be regarded as the last layer that produces a prediction or output of the network. Each node in the output layer may correspond to a possible output or class label.

[0016] The neural network may be a trained neural network. To train the neural network, the neural network may be initialized with initial values ​​of parameters of the neural network, such initialization may be performed by known methods, such as Xavier initialization or He initialization. Alternatively, the neural network may be initialized with parameters determined during previous training of the neural network. The initialized neural network may be provided with training data to produce one or more outputs. The training data may be provided as labeled data or unlabeled data. The training data may be provided in pairs, each pair comprising an input sample to be provided as input to the neural network and a reference true value. One or more outputs of the neural network may be provided to a cost function configured to determine a cost based on a difference between one or more outputs of the neural network and a target. The target may be provided as part of the training data, such as a reference true value or a label associated with the data. The target may be a target value, such as a mean opinion score or a signal-to-noise ratio. To train the neural network, one or more parameters of the neural network are adjusted to minimize the cost of the cost function. One or more parameters of the neural network may be adjusted according to an optimization algorithm, such as stochastic gradient descent, or Levenberg-Marquardt optimization. The training of the neural network may be iterated across multiple rounds. Each round may include passing the entire training data set through the neural network and updating the neural network parameters accordingly.

[0017] Another aspect of the present invention relates to a hearing device comprising a housing configured to be placed at an ear of a user. The hearing device may comprise a microphone system comprising two microphones. Such a microphone system may be configured to capture audio and generate an audio signal based on the captured audio. The conversion to the audio signal may be performed by an AD converter or the like, which may be included in the microphone system or connected to the microphone system. The hearing device may comprise an ambient sound detector configured to receive the audio signal, the ambient sound detector further configured to determine a classification of the ambient sound environment based on the audio signal. In such a classification, the classification may comprise determining a coherence measure between at least two audio signals or parts thereof or signals derived from the audio signals or signals derived from parts of the audio signals. The hearing device may comprise an active noise control (ANC) system configured to receive the audio signal, the ANC system configured to generate a feedforward compensation signal based on the audio signal. The hearing device may be configured to mix the audio signal or a signal derived therefrom with a proportion of the generated feedforward compensation signal according to the determined classification of the sound environment.

[0018] Furthermore, the hearing device may be configured to create a signal compensated for hearing loss based on the signal from the microphone system or a signal derived therefrom. The signal compensated for hearing loss may be created in a sound processor. The signal compensated for hearing loss may amplify or attenuate signals in one or more frequency bands, frequency shift the signal, transform the signal, reduce noise, directionality, or perform other hearing loss compensation processing to provide a signal that can be perceived as sound to the user / wearer.

[0019] The microphone system may be configured such that the first microphone is located at a position outside the ear canal and the second microphone is located at a position inside the ear canal.

[0020] The ambient sound detector may comprise a trained network trained using sounds from a first group classified as ambient sounds and comprising one or more of: background noise, speech, music and / or machine noise, and a second group classified as non-ambient sounds and comprising one or more of: own voice, processing noise, wind noise, low level sound.

[0021] The trained network may be a neural network with 4 layers.

[0022] One or more of the above advantages may be combined. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] Various aspects of the present invention will be best understood from the detailed description below in conjunction with the accompanying drawings. For clarity, the drawings are schematic and simplified, and only the details necessary for understanding the present invention are given, while other details are omitted. Throughout the specification, the same reference numerals are used for the same or corresponding parts. The various features of each aspect may be combined with any or all features of the other aspects. These and other aspects, features and / or technical effects will be apparent from and illustrated in conjunction with the following figures, in which:

[0024] Figure 1 A hearing aid is schematically shown;

[0025] Figure 2 Schematically shows elements of a hearing aid having two microphones;

[0026] Figure 3 Schematically shows the elements of a hearing aid having three microphones, one microphone being located in the ear canal and two microphones being located behind the pinna;

[0027] Figure 4 A graph showing the coherence value as a function of frequency is schematically shown;

[0028] Figure 5 Schematically shows G FF is a function of the number of environmental degrees;

[0029] Figure 6 The ambient degree is schematically shown as a function of multiple inputs;

[0030] Figure 7 Schematically shows an example structure of a (feed-forward) neural network;

[0031] Figure 8 An example of training data is schematically shown. DETAILED DESCRIPTION

[0032] The detailed description proposed below in conjunction with the accompanying drawings serves as a description of a variety of different configurations. The detailed description includes specific details for providing a thorough understanding of a number of different concepts. However, it is apparent to those skilled in the art that these concepts can be implemented without these specific details. Several aspects of the apparatus and method are described by a number of different blocks, functional units, modules, components, circuits, steps, processes, algorithms, etc. (collectively referred to as "elements"). Depending on the specific application, design limitations or other reasons, these elements can be implemented using electronic hardware, computer programs or any combination thereof.

[0033] The electronic hardware may include microelectromechanical systems (MEMS), (e.g., application specific) integrated circuits, microprocessors, microcontrollers, digital signal processors (DSPs), field programmable gate arrays (FPGAs), programmable logic devices (PLDs), gating logic, discrete hardware circuits, printed circuit boards (PCBs) (e.g., flexible PCBs), and other suitable hardware configured to perform a number of different functions described in this specification, such as sensors for sensing and / or recording physical properties of the environment, device, user, etc. Computer programs shall be broadly construed as instructions, instruction sets, codes, code segments, program codes, programs, subroutines, software modules, applications, software applications, software packages, routines, subroutines, objects, executables, execution threads, programs, functions, etc., whether referred to as software, firmware, middleware, microcode, hardware description language, or otherwise.

[0034] A hearing device (or hearing instrument, hearing aid) may be or may include a hearing aid adapted to improve or enhance the hearing ability of a user by receiving an acoustic signal from the user's environment, generating a corresponding audio signal, possibly modifying the audio signal, and providing the possibly modified audio signal as an audible signal to at least one ear of the user. "Improving or enhancing the hearing ability of a user" may include compensating for a specific hearing loss of an individual user. "Hearing device" may also refer to a device such as a hearable device, an earphone or a headset adapted to electronically receive an audio signal, possibly modify the audio signal, and provide the possibly modified audio signal as an audible signal to at least one ear of a user. The audible signal may be provided in the form of an acoustic signal radiated into the outer ear of the user, or an acoustic signal transmitted to the inner ear of the user as a mechanical vibration through the bone structure of the user's head and / or through parts of the middle ear, or an electrical signal transmitted directly or indirectly to the cochlear nerve and / or auditory cortex of the user.

[0035] The hearing device may be adapted to be worn in any known manner. This may include: i) arranging the unit of the hearing device behind the ear (with a tube that directs airborne acoustic signals into the ear canal or with a receiver / speaker that is arranged close to or in the ear canal and connected to the unit behind the ear by a wire (or wirelessly), such as a behind-the-ear hearing aid; and / or ii) arranging the hearing device in its entirety or in part in the pinna and / or ear canal of the user, such as an in-the-ear hearing aid or an in-the-canal / deep-in-the-canal hearing aid; or iii) arranging the unit of the hearing device to be connected to a fixing device implanted in the skull, such as a bone-anchored hearing aid; or iv) arranging the unit of the hearing device as a unit that is implanted in its entirety or in part, such as a bone-anchored hearing aid. The hearing device may be implemented in a single unit (housing) or in a plurality of units that are each connected to one another.

[0036] A "hearing system" refers to a system comprising one or two hearing devices, and a "binaural hearing system" refers to a system comprising two hearing devices, wherein the hearing devices are adapted to provide audible signals to both ears of a user in a coordinated manner. A hearing system or binaural hearing system may also include one or more auxiliary devices in communication with at least one hearing device, which auxiliary device affects the operation of the hearing device and / or benefits from the function of the hearing device. A wired or wireless communication link is established between at least one hearing device and the auxiliary device so that information (such as control and status signals, possibly audio signals) can be exchanged therebetween. The auxiliary device may include at least one of the following: a remote control, a remote microphone, an audio gateway device, a wireless communication device such as a mobile phone (such as a smart phone) or a tablet or another device (such as including a graphical interface), a broadcast system, a car audio system, a music player or a combination thereof. The audio gateway device may be adapted to receive a plurality of audio signals, such as from an entertainment device such as a TV or a music player, from a telephone device such as a mobile phone, or from a computer such as a PC. The auxiliary device may also be adapted to (e.g. enable a user to) select and / or combine appropriate signals from the received audio signals (or signal combinations) for transmission to at least one hearing device. The remote control is adapted to control the functions and / or operation of at least one hearing device. The functions of the remote control may be implemented in a smartphone or other (eg portable) electronic device, which may run an application (APP) for controlling the functions of the at least one hearing device.

[0037] Generally, a hearing device comprises i) an input unit such as a microphone for receiving acoustic signals from the user's surroundings and providing a corresponding input audio signal; and / or ii) a receiving unit for electronically receiving the input audio signal. The hearing device also comprises a signal processing unit for processing the input audio signal and an output unit for providing an audible signal to the user based on the processed audio signal.

[0038] The input unit may include multiple input microphones, for example, for providing direction-dependent audio signal processing. The aforementioned directional microphone system is suitable for (relatively) enhancing a target sound source among multiple sound sources in the user's environment and / or attenuating other sound sources (such as noise). On the one hand, the directional system is suitable for detecting (such as adaptively detecting) from which direction a specific part of the microphone signal originates. This can be achieved using conventionally known methods. The signal processing unit may include an amplifier suitable for applying a frequency-dependent gain to the input audio signal. The signal processing unit may also be suitable for providing other suitable functions such as compression, noise reduction, etc. The output unit may include an output transducer such as a speaker / receiver for providing an air-transmitted acoustic signal to the skull percutaneously or transcutaneously, or a vibrator for providing a structure-transmitted or liquid-transmitted signal. In some hearing devices, the output unit may include one or more output electrodes such as a cochlear implant for providing an electrical signal.

[0039] Figure 1 Schematically shown is a hearing device HD comprising a behind-the-ear housing BTE, which is configured to be located in the area between the pinna and the skull of a user.

[0040] The BTE comprises most of the electronic components of the hearing device, for example: a microphone system comprising a first and a second microphone FM and RM, a processor SPU, a memory MEM, a power supply BAT, a wireless interface I and a wireless interface II, however, one or more of these components may be located in other parts, such as the in-ear housing / part ITE or distributed across other parts. The ITE part comprises an output transducer SP, which is configured to transform the processed electrical signal into a signal perceptible to the user as sound, such as air-borne audio. One or more electronic components may be arranged in conjunction with a substrate SUB or may be arranged on the substrate SUB. The power supply BAT may be a primary battery or a storage battery, i.e. replaceable or rechargeable. Recharging may be contact charging or wireless charging.

[0041] The wireless interface may comprise an induction based system comprising a coil configured to receive and / or transmit low frequency signals such as magnetic induction signals, for example from a hearing device located on the opposite side.

[0042] The wireless interface may include an RF frequency based system including a radio frequency antenna, such as an antenna configured to receive and / or transmit at approximately 2.4 GHz.

[0043] Figure 2 Schematically shows parts of a hearing device HD, wherein two microphones (here denoted M1 and M2) may correspond to Figure 1 The term may refer to the microphones FM and RM in the BTE, but may also refer to a system where one microphone is located at the BTE and the other microphone is located at the ITE.

[0044] Two microphones M1 and M2 capture audio. Here, the two microphones constitute the entire microphone system, however, the microphone system may include additional microphones (which are not shown here). The audio signals captured by the respective microphones M1 and M2 are fed to active noise controllers ANC1 and ANC2, respectively. Each of these ANC elements is configured to provide a feed-forward compensation signal.

[0045] The purpose of an active noise control (ANC) system in a hearing device is to cancel the direct sound originating from the main noise source P that reaches the eardrum.

[0046] Furthermore, the signals from the microphone systems M1 and M2 are forwarded to an ambient sound detector ASD. The ambient sound detector ASD may use one or more of a range of measures to establish a measure of the ambient sound conditions for the hearing device. Examples include detection of own voice, sound level in general or in one or more specific frequency bands, presence or level of wind noise.

[0047] At the corresponding point, the audio signal is mixed with the generated feedforward compensation signal in a given ratio. Here, the ratio is determined at the ASD. The generated feedforward compensation signal depends on the determined sound environment classification. The output of the ambient sound detector ASD is used to control the ratio in which the feedforward compensation signal is mixed.

[0048] It is considered advantageous if the ambient sound detector is configured to determine or use a coherence measure between the microphones of the microphone system. This means that if the microphone system comprises two microphones, such as Figure 1 and Figure 2 As shown in , the coherence measure can be calculated or determined in one or more frequency bands of the two microphone signals. If the microphone system includes more than two microphones, for example three microphones, the coherence can be determined or calculated between a group of microphones. Figure 3 In FIG. 4 , this is illustrated by a hearing device with a microphone system comprising two microphones M1 located outside the ear canal and one microphone M2 located in the ear canal. It is then necessary to choose which of the two outside microphones M1 will be used for the coherence measure relative to the in-ear microphone. The in-ear microphone can be directed towards the eardrum or towards the environment.

[0049] Figure 3 An ANC system implemented as a hybrid, ie using a feed-forward (FF) ANC method and a feedback (FB) ANC method, is schematically shown.

[0050] The FF ANC system uses input from microphone M1 (front and / or rear microphones combined or alone) to predict the direct sound at the eardrum. The predicted signal is inverted to obtain an anti-noise signal to cancel the direct sound. The anti-noise signal S is generated by the FF filter and played in the ear through the receiver.

[0051] In hearing devices also having an in-ear microphone M2, FF ANC is implemented as an adaptive process, where the input from microphone M2 helps the input of microphone M1 to achieve a more accurate prediction of the noise close to the eardrum. Figure 3 , but microphone M2 is not required in some situations. Microphone M2 may be physically present, but selectively or completely not included in the establishment of the ANC signal. However, at present, it is preferred that microphone M2 is present and included in the system.

[0052] The success of the FF ANC system is highly dependent on the coherence measure between the main sound field at microphone M1 and the eardrum location (and thus microphone M2). A higher coherence means a better relationship between M1 and M2, so the input from microphone M1 can be better used to explain the sound field at microphone M2.

[0053] In practice, the benefits of the FF ANC system are best observed for those frequencies that exhibit a coherence equal to or higher than 0.8.

[0054] Figure 4 Examples of two possible coherence scenarios between microphones M1 and M2 across a given frequency range are shown. The example on the left shows that for most frequencies (possibly divided into frequency bands), the coherence is above a threshold of 0.8, meaning that microphone M1 can be largely used to predict the sound field at microphone M2, i.e. in the ear canal, close to the eardrum. Another example is shown on the right, where the coherence for most frequencies / frequency bands is below the aforementioned threshold, resulting in the sound field at microphone M1 being insufficient to be used to predict the sound field at microphone M2 / in the ear canal. This also means that the FF ANC system will largely fail in this example.

[0055] Furthermore, FF systems will likely be more detrimental than beneficial. This is related to environments encountered in the real world, so-called ambient sound-like environments will exhibit high correlation between microphones M1 and M2 at ANC-relevant frequencies, assuming that the position of the reference microphone M1 is well chosen and that the ambient sound is reaching the eardrum. In contrast, more random natural environments such as percussion, processing sounds or wind or even self-voice will largely exhibit no correlation between the mentioned microphones. In the self-voice case, there will be a different correlation and / or relationship between the microphones and the eardrum.

[0056] The FB ANC system uses the input from microphone M2 and feeds it back to the in-ear receiver through a feedback filter. The receiver produces an anti-noise field S that should ideally cancel the sound field in the ear canal at microphone M2. The principle is limited mainly by the gain allowed before the system becomes unstable and starts to oscillate. FB ANC can also be implemented as an adaptive process, where the feedback filter tries to adapt to changes in the S2 path to keep the entire feedback system stable.

[0057] The effects of the FF and FB ANC systems are combined and played from the in-ear receiver, such as Figure 3 as shown in .

[0058] As mentioned above, one problem with FF ANC systems occurs when the reference microphone M1 does not predict the direct sound field at the eardrum very well, meaning that the coherence between microphones M1 and M2 has deteriorated. In these situations, it is necessary to reduce the contribution of the FF ANC system. For a hybrid system, this would mean fading the system to FB ANC. For a system based solely on FF ANC, this would mean reducing / turning off the ANC.

[0059] The present invention provides an ANC system, wherein the FF ANC effect is achieved by means of a gain G FFThe ambient sound detector uses a coherence measure between microphones M1 and M2 to determine how much FF ANC is needed and thus how much gain G is needed. FF If the coherence between microphones M1 and M2 is above a certain threshold (in other words, if the ambientness is high), increase the gain G FF Otherwise, reduce the gain G FF Gain G FF It may also depend on other inputs such as sound input level.

[0060] Figure 5 Shows G FF is a function of the amount of ambient temperature. It can be seen that when the amount of ambient temperature between microphones is low, the gain G FF The ambient sound estimate may be a function of the coherence between the microphones.

[0061] Surprisingly, it has been found that in some circumstances it is advantageous to limit the ambient sound detector to measure or determine the ambient level by including coherence at frequencies below 3 kHz.

[0062] Additionally, or alternatively, the ambient sound detector may be based on frequency weighted coherence across different sub-bands.

[0063] like Figure 6 As shown in , the ambientness can be determined based on a series of inputs such as level, own voice and coherence across different frequency bands 1 ... K. The combination of different inputs into the ambientness metric can be a linear or non-linear combination. The combination unit can include a neural network based on the labeled input data (ambient / non-ambient).

[0064] The use of a trained neural network has been shown to be advantageous for determining ambientness.For such a trained network, the input features may include the microphone signal and / or a coherence measure estimated from the microphone signal.

[0065] Figure 7 Schematically shows an example structure of a (feed-forward) neural network with M=4 layers. The input signal passes through a number of non-linear layers of type a[l]=f(Wa[l-1]+b). The final output layer is a decision layer that classifies the input sound as ambient or non-ambient. Based on a set of training data, the weights W and biases b are trained to maximize the distinction between ambient and non-ambient sounds.

[0066] Figure 8Examples of training data labeled as ambient sound or non-ambient sound are schematically shown. The audio examples are labeled based on how well an active noise cancellation system is expected to process the sound. For this reason, it is preferred that the hearing device comprises at least two microphones, since a coherence or correlation measure between the microphones may be an important feature of an ambient sound detector. The input signal is fed into a feature extraction unit, which provides features to a classifier unit (e.g. a trained neural network). The feature extraction unit may comprise a filter bank. Features may also be obtained by adding or subtracting the input signals. The obtained features may be correlation or coherence measures or level estimates between the microphone signals.

[0067] The ambient sound processing module may be a part of the hearing device processing module, i.e., the hearing loss compensation processing, which runs in parallel with the ANC processing module. This module runs at a slower rate than the ANC processing module and may be responsible for performing noise reduction, hearing loss compensation, sound environment reproduction, beamforming, speech detection, different control measures, etc.

[0068] To be able to distinguish own voice from ambient sounds, the hearing device or the ambient sound detector comprises an own voice detector or is configured to receive signals from such a detector.

[0069] The ambient sound detector may be configured to evaluate the ambient level of each microphone (M1 front, M1 back, ...) individually or in combination. As mentioned above, the ambient level may be based on the coherence between M1 and M2, i.e. between the external microphone and the microphone in the ear / in the ear canal. In the absence of M2, the (external) microphone with the highest ambient level may be selected based on a comparison of the instantaneous levels between each microphone, e.g. the microphone with the lowest level. Depending on the microphone selected, different ANC filters may be selected.

[0070] The structural features of the apparatus described above, described in detail in the “Detailed Description of the Invention” and defined in the claims may be combined with the steps of the method of the present invention when appropriately replaced by corresponding processes.

[0071] Unless expressly stated, the singular forms "one", "the" used herein include the plural form (i.e., have the meaning of "at least one"). It should be further understood that the terms "having", "including" and / or "comprising" used in the specification indicate the presence of the described features, integers, steps, operations, elements and / or parts, but do not exclude the presence or addition of one or more other features, integers, steps, operations, elements, parts and / or combinations thereof. It should be understood that, unless expressly stated, when an element is referred to as being "connected" or "coupled" to another element, it may be directly connected or coupled to the other element, or there may be an intermediate intervening element. As used herein, the term "and / or" includes any and all combinations of one or more listed related items. Unless expressly stated, the steps of any method disclosed herein do not have to be performed in the exact order disclosed.

[0072] It should be appreciated that reference in this specification to "an embodiment" or "embodiment" or "aspect" or features that "may" include means that the specific features, structures or characteristics described in conjunction with the embodiment are included in at least one embodiment of the present invention. In addition, the specific features, structures or characteristics may be appropriately combined in one or more embodiments of the present invention. The foregoing description is provided to enable those skilled in the art to implement the various aspects described herein. Various modifications will be apparent to those skilled in the art, and the general principles defined herein may be applied to other aspects. Unless expressly stated, an element referred to in the singular does not mean "one and only one", but rather "one or more". Unless expressly stated, the term "some" refers to one or more.

[0073] Therefore, the protection scope of the present invention should be judged according to the claims.

Claims

1. A method for operating a hearing device comprising an active noise control system, the hearing device being configured to be placed at an ear of a wearer, the hearing device comprising a microphone system comprising two microphones arranged outside the ear canal of the wearer when the hearing device is worn, the method comprising: capturing audio with the microphone system; generating an audio signal based on the captured audio; providing the audio signal to an ambient sound detector; determining, using an ambient sound detector, a classification of an ambient sound environment based on audio signals from two microphones, wherein the classification comprises determining a coherence measure between at least two audio signals; providing an audio signal to the active noise control system, the active noise control system generating a feed-forward compensation signal based on the audio signal; An audio signal is mixed with a proportion of the generated feed-forward compensation signal in accordance with the determined classification of the ambient sound environment.

2. The method according to claim 1, wherein: The microphone system is configured such that a first microphone is located at a position outside a wearer's ear canal and a second microphone is located at a position inside the wearer's ear canal.

3. The method according to claim 1, wherein: The classification includes or is a measure of environmental degree.

4. The method according to claim 3, wherein: The ambient degree is expressed as a normalized value in the interval [0:1].

5. The method according to claim 1, wherein: The classification of the ambient sound environment is performed for frequencies below 3 kHz.

6. The method according to claim 1, wherein: The classification of the ambient sound environment is based on or includes a coherence metric determined for each of a plurality of frequency bands or a subset thereof.

7. The method according to claim 1, wherein: The classification of the ambient sound environment is further based on one or more of: level, detection of the presence of own voice, a coherence measure between at least one frequency band of each of the audio signals.

8. The method according to claim 6, wherein: The coherence measure is determined for at least two adjacent frequency channels or at least two non-adjacent frequency channels.

9. The method according to claim 1, wherein: The classification of the ambient sound environment is performed using linear and / or non-linear combinations of more than one input from the ambient sound detector.

10. The method according to claim 1, wherein: The ambient sound detector comprises a trained network trained using sounds from a first group classified as ambient sounds and comprising one or more of: background noise, speech, music and / or machine noise, and a second group classified as non-ambient sounds and comprising one or more of: own voice, processing noise, wind noise, low level sound.

11. The method according to claim 10, wherein: The input to the trained network of the ambient sound detector is or comprises a coherence measure.

12. A hearing device comprising a housing configured to be placed at an ear of a wearer, the hearing device comprising: a microphone system comprising two microphones arranged outside the ear canal of the wearer when the hearing device is worn, wherein the microphone system is configured to capture audio and to generate an audio signal based on the captured audio; an ambient sound detector configured to receive the audio signal, the ambient sound detector further configured to determine a classification of the ambient sound environment based on the audio signal, wherein the classification comprises determining a coherence measure between at least two audio signals; an active noise control system configured to receive an audio signal, the active noise control system configured to generate a feed-forward compensation signal based on the audio signal; Therein, the hearing device is configured to mix the audio signal with a proportion of the generated feed-forward compensation signal depending on the determined classification of the ambient sound environment.

13. The hearing device according to claim 12, wherein: The microphone system is configured such that a first microphone is located at a location outside the ear canal and a second microphone is located at a location inside the ear canal.

14. The hearing device according to claim 12, wherein: The ambient sound detector comprises a trained network trained using sounds from a first group and a second group, the first group being classified as ambient sounds and comprising one or more of: background noise, speech, music and / or machine noise, the second group being classified as non-ambient sounds and comprising one or more of: own voice, processing noise, wind noise, low level sound.

15. The hearing device according to claim 14, wherein The trained network is a neural network with 4 layers.

Images