Method and hearing system for processing an audio signal containing impulse noise

Abstract

The disclosure relates to a method of processing an audio signal in a hearing device, the method comprising - detecting impulse noise (521 - 524, 541 - 544, 561 - 564) in the audio signal (510, 530, 550); and, - applying a gain reduction (661, 671, 681, 691) on the audio signal (510, 530, 550) to reduce a signal level of the impulse noise (521 - 524, 541 - 544, 561 - 564). The disclosure also relates to a hearing system for performing the method. To provide for an effective reduction of the impulse noise, e.g., in a way that the outcome is less dependent on a processing latency, the disclosure proposes - adjusting, upon detecting the impulse noise (521 - 524, 541 - 544, 561 - 564) at one or more preceding times, a delay of the audio signal (510, 530, 550) to improve a timing of the applying of the gain reduction (661, 671, 681, 691) when the impulse noise (521 - 524, 541 - 544, 561 - 564) is detected at a subsequent time.

Description

TECHNICAL FIELD

[0001] The disclosure relates to a method of processing an audio signal, according to the preamble of claim 1. The disclosure further relates to a hearing system configured to perform the method, according to the preamble of claim 15.BACKGROUND

[0002] Hearing devices may be used to improve the hearing capability or communication capability of a user, for instance by compensating a hearing loss of a hearing-impaired user, in which case the hearing device is commonly referred to as a hearing instrument such as a hearing aid, or hearing prosthesis. A hearing device may also be used to output sound based on an audio signal which may be communicated by a wire or wirelessly to the hearing device. A hearing device may also be used to reproduce a sound in a user's ear canal detected by an input transducer such as a microphone or a microphone array. The reproduced sound may be amplified to account for a hearing loss, such as in a hearing instrument, or may be output without accounting for a hearing loss, for instance to provide for a faithful reproduction of detected ambient sound and / or to add audio features of an augmented reality in the reproduced ambient sound, such as in a hearable. A hearing device may also provide for a situational enhancement of an acoustic scene, e.g. beamforming and / or active noise cancelling (ANC), with or without amplification of the reproduced sound. A hearing device may also be implemented as a hearing protection device, such as an earplug, configured to protect the user's hearing. Different types of hearing devices configured to be be worn at an ear include earbuds, earphones, hearables, and hearing instruments such as receiver-in-the-canal (RIC) hearing aids, behind-the-ear (BTE) hearing aids, in-the-ear (ITE) hearing aids, invisible-in-the-canal (IIC) hearing aids, completely-in-the-canal (CIC) hearing aids, cochlear implant systems configured to provide electrical stimulation representative of audio content to a user, a bimodal hearing system configured to provide both amplification and electrical stimulation representative of audio content to a user, or any other suitable hearing prostheses. A hearing system comprising two hearing devices configured to be worn at different ears of the user is sometimes also referred to as a binaural hearing device. A hearing system may also comprise a hearing device, e.g., a single monaural hearing device or a binaural hearing device, and a user device, e.g., a smartphone and / or a smartwatch, communicatively coupled to the hearing device.

[0003] Hearing devices are often employed in conjunction with communication devices, such as smartphones or tablets, for instance when listening to sound data processed by the communication device and / or during a phone conversation operated by the communication device. More recently, communication devices have been integrated with hearing devices such that the hearing devices at least partially comprise the functionality of those communication devices. A hearing system may comprise, for instance, a hearing device and a communication device.

[0004] Advanced audio signal processing techniques have been developed to improve a hearing experience for users in various fields. For example, hearing instruments rely on audio signal processing to enhance auditory perception for individuals with hearing impairments.

[0005] Some important audio processing applications address impulse noise. Impulse noise refers to a type of noise which may be characterized by a transient event of a high intensity burst of sound. These bursts can occur suddenly and are usually of very brief duration, often appearing as spikes in the audio signal. Impulse noise can be caused by various factors, including electrical interference, mechanical disturbances, or environmental events such as clicks, pops, or other transient noises. Examples include noise caused by a switching of mechanical or electronic devices, gunshots or firecrackers, applause, clapping, door slams, scratching of cutlery, clinking of glasses, and / or the like. Impulse noise can be harmful for many reasons including an auditory disruption causing discomfort or even pain for the user, a signal distortion compromising a signal integrity and / or masking important audio information, a damaging of audio equipment caused by the sudden and extreme amplitude changes, a communication interference disrupting a clarity and intelligibility of speech, and a data corruption affecting a quality of audio reproduction and transmission.

[0006] Traditional audio processing methods aiming to reduce impulse noise in the audio signal typically involve a combination of multiple operations including a detection of the impulse noise, a calculation of a gain reduction, and an application of the gain reduction. Impulse noise detection involves recognizing the occurrence of an acoustic shock and assessing its intensity. This process often incorporates multiple measures to ensure accurate detection. Gain calculation aims to determine an appropriate strength and temporal slope for a change of gain to be applied to the audio signal. This calculation is crucial for ensuring that the transient event is perceived naturally by the user, without causing discomfort. In many cases, unlike complete suppression, the objective is to modulate the transient event to a level that is both noticeable and tolerable. Finally, the gain application step involves applying the calculated gain change, which may vary over time. In particular, the gain modification may be applied relative to an existing gain function within the signal path. This ensures that the transient event is handled dynamically, maintaining the integrity of the overall auditory experience.

[0007] For instance, according to WO 2010 / 083879 A1, the impulse noise detection mechanism includes measuring a peak difference of a signal upstream of a band split filter bank and comparing the peak difference against a peak difference limit. US 2003 / 0031335 A1 discloses an adaptive setting of a threshold value depending on a level of an input audio signal which can limit the transient noise contained in an output audio signal.

[0008] One of the effects of advanced audio processing techniques can be a latency leading to a delay between a time at which an audio signal is received by a processor and a time at which the processed signal is output, e.g., to an audio output unit for stimulating a user's hearing. Latency can impact the user experience because it may affect a synchronization between what users see and hear and may give rise to a comb filter effect. For instance, latency may limit an understanding of speech, a lip-synchronization of speech contained in the audio signal, an environmental sound awareness, and / or a user's listening comfort. Latency may also contribute to an increased perception of echo and feedback. Thus, to ensure optimal performance and user satisfaction, an effort is often made to minimize the latency so as to allow a faster processing, e.g., in real-time. This may involve a balancing of the need for advanced processing features, such as noise reduction and speech enhancement, with the requirement for low-latency operation.

[0009] One side effect of the processing latency is that it often provides a "look ahead" capability in a sense that a desired modification of an input audio signal can be performed rather spontaneously after the latency period while performing a required signal evaluation and / or preparation beforehand during the latency period. As an example, when performing the audio processing in a frequency domain, the inherent processing latency associated with the frequency domain transformation may significantly contribute to provide such a "look ahead" capability. In many cases, when performing a reduction of impulse noise in the audio signal, this latency also allows sufficient time for the impulse noise detection and gain calculation components to perform their respective functions effectively, ensuring timely and accurate reduction of impulse noise by the gain application component.

[0010] In modem trends, however, which aim to further reduce the processing latency as much as possible, such a timely and accurate application of a target gain for reducing impulse noise can be rather challenging, at least when a prior detection of impulse noise and / or an according determination of an appropriate gain function, which may be customized to the detected impulse noise and / or a currently applied gain function, shall be performed in a reliable manner. Nevertheless, as reduced latency and a reduction of impulse noise are both critical factors influencing the hearing experience for a user, as described above, it would be desirable to optimize the audio signal processing in a way to provide for the combined advantages of both processing schemes as much as possible whereas the inherent disadvantages of each processing scheme are kept at a minimum.SUMMARY

[0011] It is an object of the present disclosure to provide for an improvement of current audio processing methods for impulse noise reduction, in particular in a way that the desired processing outcome of reducing an impulse noise in an audio signal is less dependent on an inherent latency caused by the signal processing and / or more compatible with and / or less affected by a signal processing of a rather low and / or varying latency. It is another object to propose a method for audio processing and / or a hearing system and / or a hearing device in which transient events of impulse noise can be handled in an effective and / or adaptive way, e.g., by also optimizing an impact on other processing schemes applied on the audio signal and / or advantages thereof. It is a further object to propose a method and / or system and / or device allowing to identify an increased likelihood of forthcoming transient events of impulse noise in the audio signal and / or to provide for impulse noise reduction in the audio signal when its likelihood is increased. It is a further object to improve audio signal processing in a hearing system and / or hearing device in way that improves a hearing experience for the user.

[0012] At least one of these objects can be achieved by a method of processing an audio signal comprising the features of patent claim 1 and / or a hearing device for performing the method comprising the features of patent claim 15. Advantageous embodiments are defined by the dependent claims and the following description.

[0013] Accordingly, the present disclosure proposes a of processing an audio signal comprising detecting impulse noise in the audio signal; applying a gain reduction on the audio signal to reduce a signal level of the impulse noise; and adjusting, upon detecting the impulse noise at one or more preceding times, a delay of the audio signal to improve a timing of the applying of the gain reduction when the impulse noise is detected at a subsequent time.

[0014] Independently, the present disclosure proposes a non-transitory computer-readable medium storing instructions that, when executed by a processor, which may be included in a hearing device, cause a hearing device to perform the method.

[0015] Independently, the present disclosure proposes a hearing system comprising a hearing device configured to be worn at an ear of a user, the hearing system comprising an audio input unit for obtaining an input audio signal; a processing unit for audio signal processing of the input audio signal to obtain an output audio signal; and an audio output unit for outputting the output audio signal so as to stimulate the user's hearing, wherein the audio input is included in the hearing device and the processor is configured to perform the method based on the input audio signal. E.g., the method may be performed by providing the input audio signal obtained by the audio input unit to the processing unit for the processing of the input audio signal.

[0016] In some implementations, the hearing system may comprise the hearing device and / or a remote device communicatively coupled to the hearing device. E.g., the audio input unit and / or the processor may be included in the hearing device and / or the remote device. In some examples, the hearing system may be provided as the hearing device, wherein the audio input unit and the processor are included in the hearing device. Independently, the present disclosure thus also proposes a hearing device configured to be worn at an ear of a user, the hearing device comprising an audio input unit for obtaining an input audio signal; a processing unit for audio signal processing of the input audio signal to obtain an output audio signal; and an audio output unit for outputting the output audio signal so as to stimulate the user's hearing, wherein the processor is configured to perform the method based on the input audio signal.

[0017] Subsequently, additional features of some implementations of the method and / or the hearing device and / or the computer readable medium are described. Each of those features can be provided solely or in combination with at least another feature. The features can be correspondingly provided in some implementations of the method and / or the hearing device and / or the computer readable medium.

[0018] In some implementations, the delay is adjusted by applying a delay scheme on the audio signal, the delay scheme defining a minimum duration during which the delay adjustment is maintained. E.g., the delay scheme may define a minimum duration during which the delay is kept at an adjusted value differing from a value of the delay before the adjusting of the delay.

[0019] In some implementations, when the delay scheme is applied, the delay scheme is updated upon detecting the impulse noise at the subsequent time, wherein the updating of the delay scheme comprises extending said minimum duration. In some examples, the minimum duration is extended by restarting the delay, e.g., from time zero, for a prolongation period. The prolongation period may correspond to the minimum duration of the delay scheme applied before the updating of the delay scheme and / or may be different from this minimum duration, e.g., it may be larger.

[0020] In some implementations, the delay scheme further defines an optimum value of the delay, the optimum value minimizing a difference between a time at which the impulse noise occurs in the audio signal and a time at which the gain reduction is applied on the audio signal.

[0021] In some implementations, the delay scheme further defines a duration and / or rate in which the delay is decreased over time from the optimum value to return to a value of the delay before the adjusting of the delay; and / or a duration and / or rate in which the delay is increased over time from a value of the delay before the adjusting of the delay to the optimum value; and / or a duration during which the delay is kept constant at the optimum value. In some examples, the minimum duration during which the delay adjustment is maintained comprises the duration during which the delay is kept constant and / or the duration in which the delay is decreased and / or increased. E.g., the duration in which the delay is decreased and / or increased may be defined by the rate in which the delay is decreased and / or increased to the optimum value.

[0022] In some implementations, when updating the delay scheme, the delay scheme may be applied again with equal properties, e.g., with an equal duration during which the delay is kept constant and / or an equal duration in which the delay is decreased and / or increased, so as to extend the minimum duration by the duration during which the delay is kept constant and / or the duration in which the delay is decreased and / or increased. In some examples, when updating the delay scheme, one or more properties of the delay scheme may altered before applying the delay scheme. The minimum duration may then be extended by an altered duration during which the delay is kept constant and / or an altered duration in which the delay is decreased and / or increased.

[0023] In some implementations, the method further comprises determining the delay scheme depending on a latency caused by the audio signal processing; and / or a current value of the delay; and / or a number of the preceding times at which impulse noise has been detected; and / or an intermediate time interval between impulse noise detected at two or more of the preceding times; and / or a time elapsed since the impulse noise has been detected at one or more of the preceding times. In some examples, when a latency of the audio signal processing is larger, the optimum value of the delay may be determined to be smaller and / or the duration and / or rate in which the delay is increased over time may be determined to be smaller. In some examples, when the number of preceding times at which impulse noise has been detected is larger and / or the intermediate time interval is smaller and / or the time elapsed is smaller, the minimum duration of maintaining the delay adjustment may be determined to be larger.

[0024] In some implementations, the audio signal is processed in a time domain. In some examples, the audio signal processing may be performed exclusively in the time domain. In some implementations, the audio signal may be transformed into another domain, e.g., a frequency domain, wherein at least part of the audio signal processing is performed in the other domain.

[0025] In some implementations, when detecting the impulse noise at one or more preceding times, the applying of the gain reduction on the audio signal is omitted. In some examples, the applying of the gain reduction on the audio signal may be omitted when initially detecting the impulse noise, e.g., at the temporally first preceding time which may also be denoted as the initial time. In some examples, the applying of the gain reduction on the audio signal may be omitted when detecting the impulse noise at two or more of the preceding times, e.g., the initial time and the preceding time following the initial time.

[0026] In some implementations, the processing of the audio signal comprises applying a gain function on the audio signal, wherein the gain reduction is applied in addition to the gain function. In some examples, the gain function may provide for a frequency dependent amplification of the audio signal, e.g., to compensate for a hearing loss of the user. For instance, the gain function may be fitted to the hearing loss of an individual user. In some examples, the gain function may provide for a positive gain of the audio signal and / or a negative gain of the audio signal.

[0027] In some implementations, the method further comprises determining the gain reduction, wherein the gain reduction is determined so as to meet a target gain representative of an intended signal level of the impulse noise in the audio signal. In some examples, when the processing of the audio signal comprises applying a gain function on the audio signal, the target gain may correspond to a gain which results in applying the gain reduction in addition to the gain function, e.g., to a sum of the gain function and the gain reduction.

[0028] In some implementations, the determining the gain reduction comprises determining a maximum value of the gain reduction; and / or a duration in which the gain reduction is maintained at the maximum value; and / or a rate and / or duration in which the gain reduction is increased over time, e.g., before the gain reduction reaches the maximum value; and / or a rate and / or duration in which the gain reduction is decreased over time, e.g., after the gain reduction reaches the maximum value.

[0029] In some implementations, the gain reduction is determined depending on one or more characteristics of the impulse noise; and / or a latency caused by the audio signal processing; and / or a current value of the delay. In some examples, one or more characteristics of impulse noise may comprise a signal level and / or a duration of the impulse noise and / or a frequency of sound contained in the impulse noise and / or a frequency in which the impulse noise occurs over time in the audio signal. In some examples, when the one or more characteristics of the impulse noise comprise a signal level of the impulse noise and the signal level is larger, the maximum value of the gain reduction is determined to be larger. In some implementations, when the latency caused by the audio signal processing and / or the current value of the delay is smaller, the rate in which the gain reduction is increased over time is determined to be larger.

[0030] In some implementations, a time consumption caused by the detecting of the impulse noise is adjusted depending on a time elapsed since the impulse noise has been detected at one or more of the preceding times; and / or an intermediate time interval between impulse noise detected at two or more of the preceding times; and / or a number of the preceding times at which impulse noise has been detected; and / or one or more characteristics of impulse noise detected at one or more of the preceding times. In some examples, the time consumption caused by the detecting of the impulse noise may be reduced upon detecting the impulse noise at one or more preceding times, e.g., at the initial time. To illustrate, after detecting the impulse noise at the one or more preceding times, an increased likelihood of future occurrences of impulse noise may be employed to reduce the time consumption caused by the detecting of the impulse noise.

[0031] In some implementations, a time consumption caused by the determining of the gain reduction is adjusted depending on a time elapsed since the impulse noise has been detected at one or more of the preceding times; and / or an intermediate time interval between impulse noise detected at two or more of the preceding times; and / or a number of the preceding times at which impulse noise has been detected; and / or one or more characteristics of impulse noise detected at one or more of the preceding times. In some examples, the time consumption caused by the determining of the gain reduction may be reduced upon detecting the impulse noise at one or more preceding times, e.g., at the initial time. To illustrate, after detecting the impulse noise at the one or more preceding times, an increased likelihood of future occurrences of impulse noise with similar characteristics, which may require corresponding properties of the gain reduction, may be employed to reduce the time consumption caused by the determining of the gain reduction.

[0032] In some implementations, when adjusting the time consumption caused by the detecting of the impulse noise and / or by the determining of the gain reduction, the delay of the audio signal is adjusted to account for the adjusted time consumption.

[0033] In some implementations, the processing of the audio signal is performed at a first signal path at which the delay of the audio signal is adjusted and at which the gain reduction is applied on the audio signal, and the impulse noise in the audio signal is detected at a second signal path bypassing the first signal path. In some examples, the determining the delay scheme and / or the determining the gain reduction is also performed at the second signal path.

[0034] In some implementations, the delay may be adjusted upon detecting the impulse noise at the initial time. In some implementations, the delay may be adjusted upon detecting the impulse noise at two or more preceding times, e.g., at the initial time and the following preceding time.

[0035] In some implementations, the hearing device is configured to be worn at an ear of the user. In some examples, the hearing device may be configured to be at least partially inserted into an ear canal of the user.

[0036] In some implementations, by the adjusting of the delay, a latency of the audio processing prior to the adjusted delay may be arbitrary, e.g., rather low and / or vary over time. In this way, advantages of low latency and / or varying latency audio processing may be preserved and / or exploited. Moreover, transient events of impulse noise, which may occur in the audio signal in a random and / or unpredictable manner, can be handled effectively and / or in an adaptive manner. In this way, an impact on other processing schemes applied on the audio signal and / or advantages thereof, e.g., with regard to a desired latency, may be reduced. In particular, an impact may be restricted to the events at which impulse noise is detected at the one or more preceding times and / or at which impulse noise occurs with the increased likelihood. This may contribute to an improved listening experience for the user.BRIEF DESCRIPTION OF THE DRAWINGS

[0037] Reference will now be made in detail to embodiments, examples of which are illustrated in the accompanying drawings. The drawings illustrate various embodiments and are a part of the specification. The illustrated embodiments are merely examples and do not limit the scope of the disclosure. Throughout the drawings, identical or similar reference numbers designate identical or similar elements. In the drawings: Fig. 1schematically illustrates an exemplary impulse noise reduction system; Fig. 2schematically illustrates an exemplary hearing device implementing the impulse noise reduction system illustrated in Fig. 1; Fig. 3schematically illustrates an embodiment of the hearing device illustrated in Fig. 2 as a RIC hearing aid; Figs. 4-8schematically illustrate block diagrams of exemplary signal processing algorithms for impulse noise reduction; Figs. 9A - Cillustrate exemplary audio signals which are processed according to principles described herein; Figs. 10A - Cillustrate, in Figs. 10A, C, exemplary audio signals, and, in Fig. 10B, a delay scheme for adjusting a delay of the audio signal illustrated in Fig. 10A to obtain the audio signal illustrated in Fig. 10C; Figs. 11A - Cillustrate, in Figs. 11A, C, exemplary audio signals, and, in Fig. 10B, a sequence of gain reductions applied on the audio signal illustrated in Fig. 11A to obtain the audio signal illustrated in Fig. 11C; and Fig. 12schematically illustrates an exemplary method of processing an audio signal according to principles described herein. DETAILED DESCRIPTION OF THE DRAWINGS

[0038] Methods, systems, and devices of processing an audio signal in a hearing device for reducing impulse noise in the audio signal are described herein. As will be described in more detail below, an exemplary method may comprise detecting impulse noise in the audio signal; applying a gain reduction on the audio signal to reduce a signal level of the impulse noise; and adjusting, upon detecting the impulse noise at one or more preceding times, a delay of the audio signal to improve a timing of the applying of the gain reduction when the impulse noise is detected at a subsequent time.

[0039] By providing methods, systems, and devices such as those described herein, it may be possible to provide for a compatibility of the impulse noise reduction with various audio processing algorithms. Examples may include algorithms operating with a rather low latency, wherein the low latency may be maintained at least under normal conditions and / or at the absence of impulse noise in the audio signal. Examples may also include algorithms causing variable processing delays, for instance when activating and deactivating optional processing features. Examples may also include algorithms for audio processing in a time domain, algorithms for audio processing in a frequency domain, algorithms providing for hearing loss compensation, beamforming, noise cancellation, speech enhancement, audio reproduction and / or augmentation, and / or the like.

[0040] In some implementations, methods, systems, and devices such as those described herein may be employed in audio processing with arbitrary low latency, wherein the latency lowness may only be limited in situations in which a presence of impulse noise in the audio signal is more likely than in other situations and / or wherein the latency can be selected not to exceed a value which is minimally required for an efficient reduction of the impulse noise.

[0041] To illustrate, the latency may be reduced by turning off time-consuming processing operations depending on specific circumstances, e.g., a current acoustic environment and / or listening intention of the user, in which their presumed benefit is rather low. In some examples, latency may be reduced by performing the processing without converting the audio signal to another domain, such as a frequency domain, but processing the audio signal in its original temporal form, which may also be referred to as a processing in a time domain.

[0042] In some implementations, methods, systems, and devices such as those described herein may be employed to determine situations in which impulse noise is more likely to be contained in the audio signal than in other situations and / or to modify impulse noise when its likelihood is increased. This may result in an improved hearing experience for a user, e.g., by optimizing an impact of impulse noise reduction in a processing scheme optimized for the user in various other aspects.

[0043] To illustrate, by adjusting the delay of the audio signal upon detecting the impulse noise at the one or more preceding times, an impact of the delay adjustment on the audio signal processing can be automatically adapted, e.g., reserved, to situations of an increased likelihood in which impulse noise may occur. An assumption of an increased likelihood of impulse noise after a preceding detection may be based on the observation, that, in real life situations, impulse noise often appears in a cumulative manner, e.g., in a temporal clustering after an initial detection. Thus, when impulse noise has been detected at the one or more preceding times, it can be suspected that it will occur again at the subsequent time, at least with an increased probability as compared to a situation in which no impulse noise has been previously detected. This may imply that the detection of the impulse noise at the one or more preceding times is employed as an indicator of the increased likelihood of subsequent impulse noise.

[0044] Those and other implementations and advantages are further described in the description that follows.

[0045] FIG. 1 illustrates an exemplary impulse noise reduction system 100 that may be implemented according to principles described herein. As shown, system 100 may include, without limitation, a memory 102 and a processor 104 selectively and communicatively coupled to one another. Memory 102 and processor 104 may each include or be implemented by hardware and / or software components (e.g., processors, memories, communication interfaces, instructions stored in memory for execution by the processors, etc.). In some examples, memory 102 and / or processor 104 may be implemented by any suitable computing device. In other examples, memory 102 and / or processor 104 may be distributed between multiple devices and / or multiple locations as may serve a particular implementation. Illustrative implementations of system 100 are described herein.

[0046] Memory 102 may maintain (e.g., store) executable data used by processor 104 to perform any of the operations described herein. For example, memory 102 may store instructions 106 that may be executed by processor 104 to perform any of the operations described herein (e.g., instructions to detect impulse noise and / or to apply a gain reduction and / or to adjust a delay). Instructions 106 may be implemented by any suitable application, software, code, and / or other executable data instance. Memory 102 may also maintain any data received, generated, managed, used, and / or transmitted by processor 104. Memory 102 may store any other suitable data as may serve a particular implementation. For example, memory 102 may store data associated with parameters applied in an audio signal processing performed by processor 104 (e.g., information about a delay and / or a gain reduction to be applied on the audio signal and / or a delay and / or gain reduction currently applied and / or previously).

[0047] Processor 104 may be configured to perform (e.g., execute instructions 106 stored in memory 102 to perform) various processing operations associated with processing an audio signal including a reduction of impulse noise in the audio signal. For example, processor 104 may perform one or more operations on the audio signal described herein, e.g., to detect impulse noise, apply a gain reduction, and adjust a delay.

[0048] System 100 may be implemented in any suitable manner. For example, system 100 may be implemented as a hearing device, a communication device and / or a remote microphone communicatively coupled to the hearing device, e.g., via radio frequency (RF) radiation, or a combination of the hearing device and the communication device and / or the remote microphone.

[0049] FIG. 2 illustrates an exemplary implementation of system 100 as a hearing device 200 configured to be worn at an ear of a user. Hearing device 200 may be implemented by any type of hearing device configured to enable or enhance hearing or a listening experience of a user wearing hearing device 200. For example, hearing device 200 may be implemented by a hearing aid configured to provide an amplified version of audio content to a user, a sound processor included in a cochlear implant system configured to provide electrical stimulation representative of audio content to a user, a sound processor included in a bimodal hearing system configured to provide both amplification and electrical stimulation representative of audio content to a user, or any other suitable hearing prosthesis, or an earbud or an earphone or a hearable.

[0050] In certain examples, hearing device 200 may be implemented as part of a binaural hearing system. Such a binaural hearing system may include a first hearing device associated with a first ear of a user and a second hearing device associated with a second ear of a user. In such examples, the hearing devices may each be implemented by any type of hearing device configured to provide or enhance hearing to a user of a binaural hearing system. In some examples, the hearing devices in a binaural system may be of the same type. For example, the hearing devices may each be hearing aid devices. In certain alternative examples, the hearing devices may be of a different type. For example, a first hearing device may be a hearing aid and a second hearing device may be a sound processor included in a cochlear implant system.

[0051] Different types of hearing device 200 can also be distinguished by the position at which they are worn at the ear. Some hearing devices, such as behind-the-ear (BTE) hearing aids and receiver-in-the-canal (RIC) hearing aids, typically comprise an earpiece configured to be at least partially inserted into an ear canal of the ear, and an additional housing configured to be worn at a wearing position outside the ear canal, in particular behind the ear of the user. Some other hearing devices, as for instance earbuds, earphones, hearables, in-the-ear (ITE) hearing aids, invisible-in-the-canal (IIC) hearing aids, and completely-in-the-canal (CIC) hearing aids, commonly comprise such an earpiece to be worn at least partially inside the ear canal without an additional housing for wearing at the different ear position.

[0052] Hearing device 200 may include, without limitation, a memory 202 and a processor 204 selectively and communicatively coupled to one another. Memory 202 and processor 204 may each include or be implemented by hardware and / or software components (e.g., processors, memories, communication interfaces, instructions stored in memory for execution by the processors, etc.).

[0053] Memory 202 may maintain (e.g., store) executable data used by processor 204 to perform any of the operations associated with hearing device 200. For example, memory 202 may store instructions 206 that may be executed by processor 204 to perform any of the operations associated with hearing device 202 assisting a user in hearing and / or any of the operations described herein, which may include instructions 106.

[0054] In some examples, instructions 206 may include routines for audio signal processing in a time domain. In some examples, instructions 206 may include routines for audio signal processing in a frequency domain, e.g., instructions to convert an audio signal from the time domain into the frequency domain and / or vice versa. In some examples, instructions 206 may include a gain function defining an amplification of an audio signal at different frequencies and / or at different signal levels, e.g., when hearing device 200 is implemented by a hearing instrument, to compensate for a hearing loss. In some examples, instructions 206 may include audio signal processing routines providing for noise cancelling, feedback cancelling, beamforming, speech enhancement, audio signal classification, performing an audio signal processing program associated with a current acoustic scene and / or depending on the audio signal classification, own voice detection, acoustic object separation, binaural synchronization, and / or the like. Instructions 206 may be implemented by any suitable application, software, firmware, code, and / or other executable data instance.

[0055] Processor 204 may be configured to perform any suitable processing operation associated with hearing device 200. For example, when hearing device 200 is implemented by a hearing instrument, such processing operations may include monitoring ambient sound and / or presenting amplified sound to user 204 via an in-ear receiver. Processor 210 may be implemented by any suitable combination of hardware and software.

[0056] As shown in FIG. 2, hearing device 200 may further include an audio input unit 213 and an audio output unit 217 communicatively coupled to processor 204. Audio input unit 213 is configured to obtain an input audio signal. Processor 204 is configured to provide for a processing of the input audio signal to obtain an output audio signal. Audio output unit 217 is configured to output sound based on the output audio signal.

[0057] In some implementations, as illustrated, audio input unit 213 may comprise a sound detector 215 configured to detect sound in an ambient environment of the user and to provide an ambient audio signal representative of the detected sound. The input audio signal, which is received by processor 204, may then at least partially be based on the ambient audio signal. In some examples, sound detector 215 may be implemented as a microphone and / or a microphone array. In some examples, after detection of the sound in the ambient environment, audio input unit 213 may be configured to prepare the ambient audio signal for an audio signal processing by processor 204. For example, audio input unit 217 may comprise an analog-to-digital converter to convert the ambient audio signal, as detected by sound detector 215, from an analog signal into a digital signal.

[0058] In some implementations, as illustrated, audio input unit 213 may comprise a radio receiver 216 configured to receive a radio audio signal from a remote audio source via radio frequency (RF) radiation. The input audio signal, which is received by processor 204, may then at least partially be based on the radio audio signal. Radio receiver 216 may be configured for wireless data reception of the radio audio signal. For instance, the radio audio signal may be received in accordance with a Bluetooth ™< protocol and / or by any other type of RF communication. In some examples, the remote audio source may be a remote microphone, e.g., a table microphone or a clip-on microphone, configured to detect sound at a remote location and transmit the radio audio signal indicative of the detected sound to radio receiver 216. In some examples, the remote audio source may be a streaming source configured for streaming the radio audio signal to radio receiver 216. In some examples, the remote audio source may be a communication device, e.g., a portable device such as a smartphone, tablet, smartwatch and / or the like, or a computing device such as a personal computer, configured for data transmission of the radio audio signal to radio receiver 216. In some examples, after reception of the radio audio signal, audio input unit 213 may be configured to prepare the radio audio signal for an audio signal processing by processor 204. For example, when the radio audio signal received from the remote audio source comprises an encoded signal, audio input unit 213 may comprise a decoder to decode the radio audio signal.

[0059] Audio output unit 217 may be implemented by any suitable audio output device configured to output sound based on the output audio signal to the user. To this end, audio output unit 217 may include an output transducer. For example, audio output unit 119 may be implemented as a receiver of a hearing aid, a loudspeaker of an earbud, or an output electrode of a cochlear implant.

[0060] Hearing device 200 may include further components as may serve a particular implementation. E.g., hearing device 200 may further include a user interface and / or a communication port for data transmission and / or an ear-canal microphone and / or other sensors such as a motion sensor and / or a physiological sensor.

[0061] FIG. 3 illustrates an exemplary implementation of hearing device 200 as a RIC hearing aid 261. RIC hearing aid 261 comprises a BTE part 270 configured to be worn at an ear at a wearing position behind the ear, and an ITE part 280 configured to be worn at the ear at a wearing position at least partially inside an ear canal of the ear. BTE part 270 comprises a BTE housing 271 configured to be worn behind the ear. BTE housing 271 accommodates a processing unit 264, which may comprise processor 204 and memory 202, communicatively coupled to sound detector 215 and radio receiver 216. BTE part 220 further includes a battery 277 as a power source. ITE part 280 is an earpiece comprising an ITE housing 281 at least partially insertable into the ear canal. ITE housing 281 accommodates audio output unit 217 implemented as a receiver. BTE part 270 and ITE part 280 are interconnected by a cable 274. Processing unit 264 is communicatively coupled to audio output unit 217 of ITE part 280 via cable 274 and cable connectors 272, 273 provided at BTE housing 271 and ITE housing 281.

[0062] FIG. 4 is a schematic block diagram of a signal processing algorithm 301 for impulse noise reduction in an audio signal, which may be implemented by system 100 and / or hearing device 200. E.g., algorithm 301 may be executed by processor 104, 204 after receiving an audio signal from an audio input unit 302, which may include sound detector 215 and / or radio receiver 216, as an input audio signal. Processor 104, 204 may process the input audio signal to obtain an output audio signal. The output audio signal may be output by an audio output unit 303 to stimulate the user's hearing.

[0063] Algorithm 301 may comprise an audio processing module 311 and an impulse noise reduction module 321. Audio processing module 311 and impulse noise reduction module 321 may perform a processing of the audio signal in parallel to one another. E.g., as illustrated audio processing module 311 may receive the audio signal from audio input unit 302 via a first signal path A, and impulse noise reduction module 321 may receive the audio signal via a second signal path B. E.g., signal path B at which the impulse noise reduction is performed may at least partially bypass signal pass A at which the audio processing by audio processing module 311 is performed.

[0064] Audio processing module 311 may be configured to execute one or more audio processing routines as may serve a particular implementation. In some examples, the processing routines may include one or more of applying a gain function on the audio signal, which may define an amplification characteristic, e.g., to compensate for an individual hearing loss of the user; a noise cancelling algorithm; a reverberation cancelling algorithm; a wind noise cancelling algorithm; a feedback cancelling algorithm; a speech enhancement algorithm; an impulse noise cancelling algorithm; an acoustic object separation algorithm; an audio signal classification algorithm and / or one or more processing algorithms optimized for an audio signal assigned to a classification; a binaural synchronization algorithm; a beamforming algorithm, in particular adapted for static and / or adaptive beamforming; an algorithm based on a neural network (NN); a music reproduction algorithm; and / or any other algorithm for modifying and / or enhancing the audio signal. In some examples, the processing routines may also comprise a conversion of the audio signal between a time domain and a frequency domain, e.g., to provide for a processing in the frequency domain by one or more of the processing routines. In some examples, one or more of the processing routines may be executed in a time domain.

[0065] Impulse noise reduction module 321 may comprise an impulse noise detection module 323, a delay adjustment module 325, and a gain modification module 327. Impulse noise detection module 323 is configured, after receiving the audio signal, to detect impulse noise in the audio signal. Impulse noise may be defined as a transient event of sound with high intensity. Key characteristics of impulse noise may include a short duration, e.g., often in a range of milliseconds, a high amplitude, in particular an amplitude significantly higher than a background noise level, and a random occurrence, e.g., in an unpredictable manner and / or without a regular pattern. Detecting the impulse noise may comprise determining an instance of a sudden increase of a level of the audio signal and / or any other detection method known in the art, for example as disclosed in WO 2010 / 083879 A1 and / or US 2003 / 0031335.

[0066] Gain modification module 327 is configured, upon detection of impulse noise, to provide for a gain reduction which can be applied on the audio signal to reduce a signal level of the impulse noise. The gain reduction may be applied, as illustrated, in the course of the audio processing performed by audio processing module 311, e.g., during or after one or more audio processing routines executed by audio processing module 311, and / or the gain reduction may be applied on the audio signal at first signal path A which may be executed in parallel to second signal path B at which impulse noise is detected in the audio signal.

[0067] In some implementations, gain modification module 327 is configured to determine the gain reduction. Determining the gain reduction may comprise determining, e.g., calculating, a maximum value of the gain reduction; and / or a duration in which the gain reduction is maintained at the maximum value; and / or a rate and / or duration in which the gain reduction is increased over time, e.g., before the gain reduction reaches the maximum value; and / or a rate and / or duration in which the gain reduction is decreased over time, e.g., after the gain reduction reaches the maximum value. In some examples, the gain reduction may be determined depending on one or more characteristics of the impulse noise in the audio signal, e.g., as detected by impulse noise detection module 323. The characteristics may include, e.g., an amplitude and / or a duration and / or a frequency and / or a rate of the impulse noise.

[0068] In some examples, the gain reduction may be determined depending on a target gain representative of an intended signal level of the impulse noise in the audio signal. For instance, when a gain function is applied on the audio signal in the course of the audio processing performed by audio processing module 311 and / or a gain function is applied on the audio signal at first signal path A, the gain reduction may be determined depending on the gain function so as to provide for the intended target gain. In some examples, gain modification module 327 may also be configured to determine the target gain, e.g., depending on one or more characteristics of the impulse noise in the audio signal.

[0069] The audio processing performed by audio processing module 311 at signal path A, e.g., when executing one or more of the audio processing routines, may produce a latency. The processing latency generally leads to an intrinsic delay between a time at which the audio signal is received, e.g., from audio input unit 302, and a time at which the processed signal is provided as an output signal, e.g., to audio output unit 303. Detecting impulse noise in the audio signal, as performed by impulse noise detection module 323 at signal path B, and / or determining the gain reduction, which may be performed by gain modification module 327 at signal path B, also consumes time and therefore produces a latency, which may be denoted as an impulse noise reduction latency. The impulse noise reduction latency leads to an intrinsic delay between a time at which the audio signal is received and a time at which the gain reduction can be applied on the audio signal to reduce the signal level of the impulse noise. In a case in which the processing latency at signal path A is lower than the impulse noise reduction latency at signal path B, e.g., when audio processing with low latency is performed by audio processing module 311, an application of the gain reduction on the audio signal at signal path A may be too late for reducing the signal level of the impulse noise.

[0070] Delay adjustment module 325 is configured to provide for adjusting, upon detecting the impulse noise at one or more preceding times by impulse noise detection module 323, a delay of the audio signal. As illustrated, the delay of the audio signal may be adjusted in the course of the audio processing performed by audio processing module 311, e.g., during or after one or more audio processing routines executed by audio processing module 311, and / or the delay of the audio signal may be adjusted at first signal path A. By adjusting the delay of the audio signal, a timing of the applying of the gain reduction on the audio signal can be improved, at least when the impulse noise is detected at a subsequent time. To illustrate, in the case in which the processing latency at signal path A is lower than the impulse noise reduction latency at signal path B, the delay of the audio signal may be adjusted accordingly, e.g., so as to minimize a difference between the processing latency at signal path A and the impulse noise reduction latency at signal path B. In particular, a difference between a time at which the impulse noise is contained in the audio signal at signal path A and a time at which the gain function is applied on the audio signal at signal path A may thus be minimized.

[0071] In some examples, e.g., in the case in which the processing latency at signal path A is lower than a latency at signal path B for adjusting the delay at signal path A, the adjusting of the delay may be too slow for providing the improved timing for applying the gain reduction so as to reduce the impulse noise detected at one or more of the preceding times. In those cases, however, providing and / or maintaining the delay of the audio signal at signal path A can be beneficial to provide the improved timing for applying the gain reduction to reduce the impulse noise detected at a subsequent time.

[0072] In this context, the circumstance may be exploited that impulse noise often appears in a cumulative manner, e.g., in a temporal clustering. Thus, after detection of impulse noise in the audio signal at an initial time and / or after detection at two or more of the preceding times, at may be expected that impulse noise will also subsequently occur in the audio signal with an increased likelihood, at least as compared to an average occurrence of impulse noise over a longer time period such as hours or days. Detecting of impulse noise at one or more preceding times by impulse noise detection module 323 may thus be exploited as an indicator of an increased probability of impulse noise in the audio signal. In cases in which future impulse noise occurrences are expected, the benefit of impulse noise reduction in the audio signal may overweigh disadvantages associated with an adjusted delay for the audio signal processing at signal path A, e.g., an increase of the processing latency.

[0073] This may imply a trade-off in which at least one initially detected impulse noise is accepted as a loss or a sacrifice for the reducing of impulse noise which is subsequently detected in the audio signal. The benefits of such a trade-off can include, e.g., an improved performance of the audio processing, for instance with regard to a smaller delay and / or a corresponding lower latency, at least in situations in which no impulse noise is contained and / or suspected to be contained in the audio signal. In some examples, such a trade-off may be perceived as a fair compromise in view of a rare occurrence of impulse noise for the average user in daily situations and / or a more tolerable adverse effect of the impulse noise when, after prevailing at the one or more preceding times, it is reduced at one or more subsequent times. In some examples, a prevailing of the impulse noise at the initial time may also import a benefit of directing the user's attention to a potential disturbance in his environment.

[0074] In some examples, the delay of the audio signal may be adjusted upon detecting the impulse noise at a single preceding time, e.g., at an initial time. The initial detection of impulse noise may be considered as sufficient to indicate an increased likelihood of future occurrences. In some examples, the delay of the audio signal may be adjusted upon detecting the impulse noise at two or more preceding times, e.g., at the initial time and at least one more time. A likelihood of future occurrences may then be regarded as even higher.

[0075] In some examples, a reduction of impulse noise may be more effective for impulse noise which is detected at the subsequent time as compared to the impulse noise detected at one or more of the preceding times. This may be achieved by means of the delay providing for the improved, e.g., optimized, timing for the subsequently detected impulse noise rather than the previously detected impulse noise. Conversely, in some examples, a reduction of impulse noise which is detected at one or more of the preceding times may be less effective, e.g., ineffective. To illustrate, when adjusting the delay upon detecting the impulse noise at the preceding time, the delay may be inadequate, e.g., behind schedule, for applying the gain function on the same impulse noise with an improved timing for the signal level reduction.

[0076] In some examples, when impulse noise is detected at one or more of the preceding times, e.g., when impulse noise is detected at the initial time, applying the gain function on the audio signal may be omitted, e.g., by default or in cases in which a reduced effectiveness or ineffectiveness of the delay can be anticipated.

[0077] In some examples, delay adjustment module 325 is configured to adjust the delay by applying a delay scheme on the audio signal. The delay scheme may define a duration of the delay; and / or an optimum value of the delay for reducing the signal level of the impulse noise; and / or a duration in which the delay is maintained at the optimum value; and / or a value of the delay at which the delay is kept constant; and / or a duration in which the delay is kept constant; and / or a duration and / or rate in which the delay is increased over time, e.g., before the delay reaches the optimum value and / or the delay is kept constant; and / or a duration and / or rate in which the delay is decreased over time, e.g., after the delay reaches the optimum value and / or the delay is kept constant.

[0078] In some examples, the delay scheme is updated, e.g., applied again, upon detecting the impulse noise at the subsequent time and / or the delay scheme is altered depending on a time elapsed since the impulse noise has been detected at one or more of the preceding times; and / or an intermediate time interval between impulse noise detected at two or more of the preceding times; and / or a number of the preceding times at which impulse noise has been detected; and / or a current value of the delay; and / or a latency caused by the audio signal processing; and / or one or more characteristics of impulse noise detected at one or more of the preceding times.

[0079] FIG. 5 is a schematic block diagram of a signal processing algorithm 331 for impulse noise reduction illustrating an exemplary implementation of algorithm 301 depicted in Fig. 4. Algorithm 331 may comprise an audio processing module 341 configured to process the audio signal in a time domain. In some examples, the audio signal received from audio input unit 302 at signal path A may be processed in its original form without conversion into another domain, e.g., without transformation into a frequency domain and inverse transformation into the time domain. This may result in an audio signal processing of low latency at first signal path A.

[0080] Audio processing module 341 may comprise a delay application module 343, a pre-processing module 345, a gain application module 347, and a post-processing module 349. Gain application module 347 is configured to apply a gain function on the audio signal. E.g., the gain function applied on the audio signal may include a positive or a negative gain and / or a gain model adapted to compensate for a hearing loss and / or a frequency and / or level dependent gain function and / or any other amplification of the audio signal.

[0081] Determining the gain reduction, as performed by gain modification module 327, may be performed depending on the gain function applied by gain application module 347. In some examples, when the gain reduction is determined so as to provide for an intended target gain for reducing the impulse noise, the gain reduction may be determined depending on a deviation of the gain function from the intended target gain. In some examples, the gain function, e.g., as currently applied by gain application module 347, is altered so as to reduce the gain function by the gain reduction.

[0082] Pre-processing module 345 is configured to provide for any type of audio processing prior to application of the gain function by gain application module 347. E.g., the pre-processing may comprise a processing of the raw audio signal, as received from audio input unit 302, to prepare it for the gain application and / or any other processing routines such as noise reduction, normalization, filtering, feature extraction, feedback cancelling, beamforming, and / or the like. Post-processing module 349 is configured to provide for any type of audio processing after application of the gain function by gain application module 347, e.g., processing routines for signal smoothing, artifact reduction, signal enhancement, compression, and / or the like.

[0083] Delay application module 343 is configured to perform the delaying of the audio signal according to the adjusting of the delay performed by delay adjustment module 325. The applying of the delay may be performed before the applying of the gain function by gain application module 347, e.g., before and / or after the signal pre-processing by module 345. Accordingly, by adjusting the delay, the latency at first signal path A can be adjusted so as to provide for sufficient time for the detecting of impulse noise and providing for the gain reduction at second signal path B for applying the gain function including the gain reduction at first signal path A to be effective so as to reduce the impulse noise in the audio signal, at least when the impulse noise is detected at the subsequent time.

[0084] FIG. 6 is a schematic block diagram of a signal processing algorithm 351 illustrating another exemplary implementation of algorithm 301. Algorithm 351 may comprise an audio processing module 361 configured to process the audio signal in a frequency domain. Audio processing module 361 may comprise delay application module 343, a frequency domain conversion module 365, a gain application module 367, and a time domain conversion module 369. Frequency domain conversion module 365 can transform the audio signal from the time domain into a frequency domain. Gain application module 367 is configured to apply a frequency dependent gain function on the audio signal. Time domain conversion module 369 can transform the audio signal from the frequency domain back into the time domain. Audio processing module 361 may comprise, e.g., after frequency domain conversion module 365 and before time domain conversion module 369 additional audio signal processing routines as may serve a particular purpose. Converting the audio signal between the time and frequency domain may be rather time consuming. Accordingly, a latency of frequency domain processing algorithm 361 may be increased as compared to the latency of time domain processing algorithm 341.

[0085] Delay application module 343 may be configured to perform the delaying of the audio signal before the applying of the gain function by gain application module 367 at least upon detecting the impulse noise at the subsequent time, e.g., before and / or after the frequency domain conversion by module 365. In some examples, when a frequency domain processing is performed by module 361 at first signal path A and the latency is smaller than the latency caused by impulse noise reduction module 321 at second signal path B, delay adjustment module 325 may increase the delay accordingly so as to improve the timing for the gain reduction by module 327, as described above.

[0086] In some examples, a time domain processing may be performed by module 341 at a preceding time and a frequency domain processing may be performed by module 361 at a subsequent time. E.g., a switching between time and frequency domain processing may be performed according to preferences of the user and / or a current acoustic scene and / or a current signal to noise ratio (SNR) of the audio signal. During the time domain processing, delay adjustment module 325 may increase the delay to improve the timing for applying the gain reduction. During the subsequent frequency domain processing, delay adjustment module 325 may decrease the delay to account for a latency of the frequency domain processing when optimizing the timing for applying the gain reduction. E.g., the delay may thus be optimized in each processing scheme to provide for the reduction of the impulse noise.

[0087] FIG. 7 is a schematic block diagram of a signal processing algorithm 371 illustrating another exemplary implementation of algorithm 301. Algorithm 371 may comprise audio processing module 311 and an impulse noise reduction module 381. Impulse noise reduction module 381 may comprise an impulse noise detection module 383 including a first impulse noise detection sub-module 385 and a second impulse noise detection sub-module 387. Sub-modules 385, 387 may be configured to detect impulse noise in the audio signal. For instance, each sub-module 385, 387 may be configured to detect impulse noise independently and / or sub-modules 385, 387 may be configured to detect impulse noise in a cooperative manner. E.g., one of sub-modules 385, 387 may detect a presence of impulse noise with a certain probability, wherein this information is used by the other sub-module 385, 387 to confirm the presence of impulse noise with a higher probability and / or to determine more characteristics of the impulse noise such as an amplitude and / or duration and / or frequency.

[0088] Delay adjustment module 325 may be configured to adjust the delay of the audio signal upon detection of impulse noise by first impulse noise detection sub-module 385. Sub-module 385 may thus also be denoted as an impulse noise detection module for the delay adjustment. Gain modification module 327 may be configured to provide for the gain reduction upon detection of impulse noise by second impulse noise detection sub-module 387. Sub-module 387 may thus also be denoted as an impulse noise detection module for the gain reduction.

[0089] Accordingly, detecting impulse noise for the gain reduction may be performed independently and / or complementary from detecting impulse noise for the delay adjustment. E.g., detecting impulse noise for the delay adjustment by first sub-module 385 may be performed in a more inaccurate and / or time-efficient manner as compared to detecting impulse noise for the gain reduction, which may require more information, e.g., with regard to characteristics of the impulse noise such as an amplitude and / or duration and / or frequency, and / or may require a higher certainty of the impulse noise detection. In this way, a time consumption for the delay adjustment may be minimized.

[0090] FIG. 8 is a schematic block diagram of a signal processing algorithm 401 for impulse noise reduction illustrating another exemplary implementation of algorithm 301. Algorithm 401 may comprise audio processing module 311 and an impulse noise reduction module 421 including an impulse noise management module 423. As illustrated, impulse noise management module 423 may be operationally connected with impulse noise detection module 323 and / or gain modification module 327 and / or delay adjustment module 325 and / or audio processing module 311.

[0091] In some examples, impulse noise management module 423 may be configured to receive and / or exchange information with impulse noise detection module 323 and / or gain modification module 327 and / or delay adjustment module 325 and / or audio processing module 311. To illustrate, delay management module 423 may receive information about impulse noise currently and / or previously detected by impulse noise detection module 323, e.g., an occurrence of impulse noise and / or one or more characteristics of the impulse noise such as a duration, amplitude and / or frequency. Delay management module 423 may also receive information about a gain reduction applied by gain modification module 327 and / or a delay applied by delay adjustment module 325, e.g., a currently applied delay scheme and / or a current value of the delay and / or a duration and / or rate in which the delay is currently kept constant and / or increased and / or decreased over time. Delay management module 423 may also receive information about the audio processing performed by audio processing module 311, e.g., a current latency caused by the audio processing. In some examples, delay management module 423 may log the information over time and / or store the information in a memory. E.g., a number of preceding times at which impulse noise has been detected may be determined based on the logged information.

[0092] In some examples, impulse noise management module 423 may be configured to control the detecting of impulse noise and / or the determining and / or applying of the gain reduction and / or the adjusting of the delay. To illustrate, gain modification module 327 may be controlled to modify and / or omit the applying of the gain reduction depending on information received from impulse noise detection module 323 and / or from delay adjustment module 325. In some instances, the applying of the gain reduction may be omitted in cases in which a reduced effectiveness or ineffectiveness of the delay can be anticipated. Those cases may include, e.g., instances in which impulse noise has been detected at an initial time and / or when the currently applied delay would not allow for a sufficient reduction of the impulse noise when applying the gain reduction with the current delay.

[0093] In some examples, the gain reduction may be modified depending on a current value of the delay and / or a current value of the latency of the audio processing, as performed by audio processing module 311. E.g., when the delay and / or latency is smaller, a rate and / or duration in which the gain reduction is increased over time, which may also be referred to as an attack time of the gain reduction, may be increased as compared to when the delay and / or latency is smaller.

[0094] In some examples, impulse noise detection module 323 may be controlled to adjust a a time consumption caused by the detecting of the impulse noise. The time consumption caused by the detecting of the impulse noise may also be denoted as an impulse noise detection latency. E.g., the impulse noise detection latency may be adjusted depending on one or more of a time elapsed since the impulse noise has been detected at one or more of the preceding times; an intermediate time interval between impulse noise detected at two or more of the preceding times; a number of the preceding times at which impulse noise has been detected; and / or one or more characteristics of impulse noise detected at one or more of the preceding times.

[0095] To illustrate, impulse noise detection module 323 may be configured to detect impulse noise with a larger time consumption, in particular a larger latency, which may result in an increased accuracy of the impulse noise detection, and / or with a smaller time consumption, in particular a smaller latency, which may result in a decreased accuracy of the impulse noise detection. Impulse noise detection module 323 may then be controlled to switch between detecting impulse noise with the larger time consumption and detecting impulse noise with the smaller time consumption, e.g., depending on previous information received from impulse noise detection module 323. For instance, when impulse noise has been detected at one or more of the preceding times, an increased likelihood of subsequent occurrences of impulse noise in the audio signal may be assumed. As a result, a decreased accuracy of impulse noise detection may be sufficient to determine the subsequent occurrences of impulse noise, which may be performed with the smaller time consumption. Accordingly, when impulse noise has been detected at one or more of the preceding times, detecting impulse noise with a smaller time consumption may be performed. In some examples, upon adjusting the time consumption caused by the detecting of the impulse noise, the delay of the audio signal, as performed by audio processing module 311 at first signal path A, is adjusted to account for the adjusted time consumption.

[0096] In some examples, impulse noise detection module 323 may be controlled to adjust a a time consumption caused by the determining and / or applying of the gain reduction. The time consumption caused by the determining and / or applying of the gain reduction may also be denoted as a gain reduction latency. E.g., the gain reduction latency may be adjusted depending on one or more of a time elapsed since the impulse noise has been detected at one or more of the preceding times; an intermediate time interval between impulse noise detected at two or more of the preceding times; a number of the preceding times at which impulse noise has been detected; one or more characteristics of impulse noise detected at one or more of the preceding times.

[0097] To illustrate, gain modification module 327 may be configured to determine the gain reduction with a larger time consumption, in particular a larger latency, which may result in an increased accuracy of the gain reduction and / or target gain, and / or with a smaller time consumption, in particular a smaller latency, which may result in a decreased accuracy of the gain reduction and / or target gain. For instance, determining the gain reduction with a smaller time consumption may imply to retrieve and / or reuse a gain reduction previously determined with a larger time consumption. Gain modification module 327 may then be controlled to switch between determining the gain reduction with the larger time consumption and determining the gain reduction with the smaller time consumption, e.g., depending on information received from impulse noise detection module 323. In particular, when an increased likelihood of subsequent occurrences of impulse noise in the audio signal is assumed due to a detection of impulse noise at one or more of the preceding times, it may also be assumed that the subsequently occurring impulse noise has comparable and / or similar characteristics as the impulse noise detected at one or more of the preceding times. Thus, when impulse noise has been detected at one or more of the preceding times and / or when impulse noise with comparable and / or similar characteristics has been detected at two or more of the preceding times, determining the gain reduction with a with a smaller time consumption may be performed. In some examples, upon adjusting the time consumption caused by the determining and / or applying of the gain reduction, the delay of the audio signal during the audio signal processing, as performed by audio processing module 311 at first signal path A, is adjusted to account for the adjusted time consumption.

[0098] FIGS. 9A, 9B, 9C illustrate functional plots 510, 530, 550 of exemplary audio signals 511, 531, 551 in a time domain. Time is indicated on an axis of abscissas. A signal level is indicated on an axis of ordinates. Audio signal 511 is an input audio signal, which may be obtained by audio input unit 213, 302 for audio signal processing. The audio processing may be performed, e.g., by audio processing module 311, 341, 361 at first signal path A. Audio signal 531 is a delayed audio signal, which may be obtained by adjusting a delay of input audio signal 511. Adjusting the delay of audio signal 511 may be performed, e.g., by impulse noise reduction module 321, 381, 421 at second signal path B. Audio signal 551 is an impulse noise reduced audio signal, which may be obtained by applying a gain reduction on delayed audio signal 531 at first signal path A. In some examples, when another gain function is applied on audio signal 531 during the audio signal processing at signal path A, applying the gain reduction may result in a target gain applied on delayed audio signal 531. The gain reduction and / or target gain may be determined by impulse noise reduction module 321, 381, 421 at second signal path B.

[0099] Input audio signal 511, as illustrated in FIG. 9A, comprises a number of transient events 521, 522, 523, 524 of short duration and high intensity. In particular, amplitudes A1, A2, A3, A4 of the signal level at transient events 521 - 524 may be significantly higher than amplitudes at a remaining signal portion 515 in which transient events 521 - 524 are absent. Transient events 521 - 524 are thus characteristic of impulse noise, which may be detected by impulse noise detection module 323, 383. As illustrated, a time interval 11, I2, I3 separated two subsequent events of impulse noise 521 - 524 generally varies, which may be attributed to a random occurrence of impulse noise.

[0100] As also illustrated, impulse noise 521 - 524 can appear in a cumulative manner in audio signal 510 over time. In the exemplary audio signal 511, four subsequent events 521, 522, 523, 524 of impulse noise are depicted for illustrative purposes, which may occur within a temporal distance of less than several seconds or several hundred milliseconds. Thus, upon detecting impulse noise 521 - 524 at one or more preceding times, an increased likelihood for detecting impulse noise 521 - 524 at a subsequent time, e.g., within a certain time period including a duration of time intervals 11 - 13, may be expected. To illustrate, such a time period may be selected depending on an acoustic environment and / or other characteristics of impulse noise 521 - 524, e.g., less than ten seconds, in some examples less than 1 second, or less than 0.5 seconds, or less than 0.2 seconds. As further illustrated, amplitudes A1 - A4 of subsequent impulse noise 521 - 524 may differ significantly despite their cumulative occurrence. In other examples, amplitudes A1 - A4 of subsequent impulse noise 521 - 524 may vary less, e.g., less than by 20 percent, which may be attributed to an equal sound source emitting the impulse noise.

[0101] Delayed audio signal 531, as illustrated in FIG. 9B, comprises events of impulse noise 541, 542, 543, 544 which are temporally separated by time intervals J1, J2, J3. Time intervals J1- J3 differ from time intervals I1 - I3 between subsequent events of impulse noise 521 - 524 in input audio signal 511. For instance, as illustrated, when the delay of audio signal 511 is adjusted to an increased value, time intervals J1 - J3 of delayed audio signal 531 may be larger than time intervals I1 - I3 of input audio signal 511. The difference between time interval J1 - J3 and time interval 11 - I3 may be a time interval D1, D2, D3 representative of a delay of audio signal 531 at a time at which impulse noise 541 - 544 occurs in audio signal 531.

[0102] In some examples, as illustrated, an initial event 521, 541 of impulse noise may occur in input audio signal 511 and in delayed audio signal 531 at an equal time. Initially occurring impulse noise 541 in audio signal 531 may thus be un-delayed relative to initially occurring impulse noise 521 in audio signal 511. This may be the case, e.g., when input audio signal 511 is only delayed upon detecting impulse noise at the initial event 521, e.g., during or after initial event 521, and / or when a latency caused by an audio processing of input audio signal 511 at first signal path A is negligible between the time of initial event 521, 541 in audio signals 511, 531. In other examples, e.g., when audio signal 511 would already be delayed before detecting impulse noise at initial event 521 and / or when a latency would be caused by the processing of audio signal 511 between the times at which initial impulse noise 521, 541 occurs in audio signals 511, 531, the initial impulse noise 521, 541 may be temporally shifted relative to one another in audio signals 511, 531. Such a temporal shift may be disregarded in the following description for illustrative purposes.

[0103] In the illustrated example, initial event 521, 541 of impulse noise precedes a first subsequent event 522, 542 of impulse noise in audio signals 511, 531. First subsequent events 522, 542 are temporally shifted by first delay D1. First subsequent event 522, 542 precedes a second subsequent event 523, 543 in audio signals 511, 531. Second subsequent events 523, 543 are temporally shifted by second delay D2. Second subsequent event 523, 543 precedes a third subsequent event 524, 544 in audio signals 511, 531. Third subsequent events 524, 544 are temporally shifted by third delay D3.

[0104] In some examples, as illustrated, one or more preceding delays D1 - D3 may vary from one or more subsequent delays D1 - D3. E.g., one or more preceding delays, as exemplified by first delay D1, may be smaller as compared to one or more subsequent delays, as exemplified by second and third delays D2, D3. The delay variation may be attributed, in some cases, to an adjustment of the delay to a desired, e.g., optimized, value in which the delay is varied, e.g., increased or decreased, until the desired value is reached, as further described below. In some examples, as also illustrated, two or more delays, as exemplified by second and third delays D2, D3, may be kept at an equal value. In some cases, this may be attributed to an adjustment of the delay in which the delay is kept constant after reaching a desired value, as further described below.

[0105] Impulse noise reduced audio signal 551, as illustrated in FIG. 9C, comprises events of impulse noise 561, 562, 563, 564, wherein, after applying delays D1 - D3 on input audio signal 511 so as to obtain delayed audio signal 531, a gain reduction for reducing a signal level of the impulse noise is applied. As a result, amplitudes A2, A3, A4 of events 522 - 524 subsequent to initial event 521 in input audio signal 511 are reduced to smaller amplitudes B2, B3, B4 of events 562 - 564 subsequent to initial event 561 in impulse noise reduced audio signal 551. In some examples, as illustrated, amplitude B1 of initial event 561 in impulse noise reduced audio signal 551 is kept equal to amplitude A1 of initial event 521 in input audio signal 511. In some cases, this may be attributed to an applying of delays D1 - D3 on audio signal 511 which is coming too late for delaying initial event 561, e.g., upon detection of initial event 561, in order to effectively apply the gain reduction for reducing amplitude A1 of initial event 521. In some cases, this may be attributed to an omission of applying the gain reduction for reducing amplitude A1 of initial event 521, e.g., in cases when a reduced effectiveness of applying the gain reduction can be anticipated. Examples of those cases are further described below.

[0106] FIG. 10B illustrates a functional plot 610 of a delay scheme 611 which may be applied on input audio signal 511 to adjust a delay of the audio signal. Time of input audio signal 511 is indicated on an axis of abscissas. A delay in units of time is indicated on an axis of ordinates. FIGS. 10A and 10C display again functional plots 510, 530 of exemplary audio signals 511, 531 for illustrative purposes. Delay scheme 611 may be applied on audio signal 511 at first signal path A, e.g., before, during, and / or after one or more other signal processing routines performed during the audio signal processing.

[0107] In some examples, when applying delay scheme 611, the delay may be adjusted from a base line value 615. In some examples, base line value 615 may correspond to a value of the delay before the adjusting of the delay, e.g., a value of the delay before applying delay scheme 611. In some examples, base line value 615 may correspond to a latency caused by an audio signal processing at first signal path A and / or a previously adjusted value of the delay and / or a time it takes to receive audio signal 511 from audio input unit 213, 302 and / or a time it takes to effectively provide for adjusting the delay. In some examples, base line value 615 may be substantially zero and / or negligible and / or disregarded, e.g., in cases in which delay scheme 611 would be effectively applied on input audio signal 511 immediately after receipt.

[0108] In some examples, delay scheme 611 comprises a first duration 621 during which the delay is increased, a second duration 623 during which the delay is kept constant, and a third duration 625 during which the delay is decreased. In other examples, a delay scheme may comprise a first duration during which the delay is decreased, a second duration during which the delay is kept constant, and a third duration during which the delay is increased. In other examples, a delay scheme may comprise a first duration during which the delay is increased, and a second duration during which the delay is decreased. In same examples a value to which the delay is increased and / or from which the delay is decreased and / or at which the delay is kept constant may correspond to an optimum value OD of the delay. Optimum value OD may be defined as a value minimizing a difference between a time at which impulse noise occurs in the audio signal, e.g., a time at which impulse noise is contained in the audio signal, and a time at which the gain reduction is applied on the audio signal.

[0109] In some examples, first duration 621 may be defined as a duration during which the delay is increased from base line value 615 to optimum value OD. First duration 621 may also be referred to as an attack time of the delay adjustment. A rate at which the delay is increased at first duration 621 may also be referred to as an attack slope of the delay adjustment. In some examples, the attack slope is constant. In other examples, the attack slope may vary over time. In some examples, third duration 625 may be defined as a duration during which the delay is decreased from optimum value OD to base line value 615. Third duration 625 may also be referred to as a release time of the delay adjustment. A rate at which the delay is decreased at third duration 625 may also be referred to as a release slope of the delay adjustment. In some examples, the release slope is constant. In other examples, the release slope may vary over time. In some examples, a constant value at which the delay is kept, e.g., at second duration 623 of delay scheme 611, may correspond to optimum value OD for applying the gain reduction on audio signal 511 to reduce impulse noise. The constant value may thus correspond to a value of the delay minimizing the difference of a time in which one or more of events 521 - 524 occur in audio signal 511 and a time at which an application of the gain reduction is performed on audio signal 511.

[0110] In some examples, delay scheme 611 comprises a delay maintenance period M1, M2. Delay maintenance period M1, M2 may define a minimum duration during which the delay adjustment is maintained. In some examples, delay maintenance period M1, M2 may define a minimum duration, e.g., a minimum time interval, at which delay scheme 611 is applied. In some examples, as illustrated, delay maintenance period M1, M2 may define a minimum duration during which the delay is kept constant, e.g., a minimum value of second duration 623. In other examples, delay maintenance period M1, M2 may define a minimum duration during which the delay is kept different from base line value 615, e.g., a minimum value of a sum of durations 621, 623, 625. In some examples, delay maintenance period M1, M2 may vary over time, e.g., when delay scheme 610 is updated upon detecting impulse noise 521 - 524 at a subsequent time and / or depending on a characteristic and / or number of preceding events 521 - 524 of impulse noise. E.g., as illustrated, a shorter delay maintenance period M1 may be extended to a longer delay maintenance period M2 when the preceding events of impulse noise exceed a threshold number, as exemplified by three preceding events 521 - 523. In other examples, delay maintenance period may be extended each time another preceding event 521 - 524 of impulse noise is detected. In this way, an increased likelihood of subsequent events of impulse noise depending on the number of preceding events may be accounted for.

[0111] In some examples, delay maintenance period M1, M2 may be selected based on a predetermined time value, e.g., an empirical value and / or a machine-learned value, in which a cumulative occurrence of impulse noise is usually observed, which may also depend, e.g., on a characteristic of the impulse noise and / or a current acoustic environment and / or individually learned acoustic patterns when a user is wearing the hearing device. In some examples, one or more properties of delay scheme 611, e.g., one or more of durations 621, 623, 625, an attack and / or release slope of the delay, a constant value of the delay and / or delay maintenance period M1, M2 may be determined by delay adjustment module 325 and / or impulse noise management module 423, e.g., depending on a current latency of the audio signal processing at first signal path A and / or a time consumption of detecting the impulse noise and / or determining the gain reduction at signal path B and / or a characteristic of the impulse noise and / or a current acoustic environment and / or individually learned acoustic patterns when a user is wearing the hearing device.

[0112] In the illustrative example shown in FIGS. 10A - 10C, delay scheme 611 is applied on audio signal 511 upon detecting one or more events of impulse noise, as exemplified by initial event 521. In practical cases, applying delay scheme 611 on audio signal 511 may lag behind a time at which event 521 occurs in audio signal 511, e.g., due to a time consumption required for detecting initial event 521 by impulse noise detection module 323, 383. For illustrative purposes, such a time lag is disregarded in the present example. Applying delay scheme 611 at first duration 621, which may define an attack time and / or attack slope of the delay adjustment, leads to an increase of the delay until reaching the constant value of the delay. Thereafter, at second duration 623 of delay scheme 611, the constant value of the delay is kept at least for the length of delay maintenance period M1.

[0113] During delay maintenance period M1, however, first subsequent event 522 of impulse noise is detected. As a result, delay scheme 611 is updated by restarting the delay maintenance period M1 for which the constant value of delay is minimally kept. This effectively leads to a prolongation of the time during which the delay is kept constant when subsequent impulse noise is detected during delay maintenance period M1. At third duration 625 of delay scheme 611, when no further subsequent event of impulse noise has been detected during delay maintenance period M1 before, the delay is decreased again. To this end, delay scheme 611 may define a release time and / or a release slope of the delay adjustment. The decrease of the delay may be continued until base line value 615 is reached again, unless a further subsequent event of impulse noise would be detected in the meantime.

[0114] In the shown example, during third duration 625 of delay scheme 611, second subsequent event 523 of impulse noise is detected. As a result, delay scheme 611 is updated again by restarting at first duration 621 of increasing the delay until reaching the constant value. Before the increase, the delay has not been minimized to base line value 615. The constant value can therefore be reached in a shorter time as compared to the initial application of delay scheme 611 upon detecting initial event 521 starting from base line value 615. Accordingly, first duration 621 is reduced as compared to the initial application. When the constant value is reached again, the delay is kept constant at second duration 623 of delay scheme 611. In this case, upon detecting a threshold number of impulse noise, as exemplified by detecting a third impulse noise at second subsequent event 523, the updated delay scheme comprises an extended delay maintenance period M2. Delay maintenance period M2, defining a minimum time at which the delay is kept constant, is larger than previously applied delay maintenance period M1.

[0115] During delay maintenance period M2, third subsequent event 524 of impulse noise is detected. As a result, delay scheme 611 is updated by a repeated application of delay maintenance period M2 so that the minimum time at which the delay is kept constant restarts upon detection of third subsequent event 524. During the repeated application of delay maintenance period M2, no further event of impulse noise is detected. Accordingly, second duration 623 of delay scheme 611 ends after the repeatedly applied delay maintenance period M2. Thereafter, at third duration 625 of delay scheme 611, the delay is decreased again.

[0116] As illustrated in FIG. 10C, applying delay scheme 611 in the above described way may lead to substantially equal delays of impulse noise, as exemplified by delays D2, D3 at subsequent events 543, 544, which may be improved, e.g., optimized, with regard to the application of the gain reduction so as to reduce the impulse noise. In some cases, the delays can vary, e.g., depending on a varying latency of the audio processing performed at first signal path A and / or depending on a varying time consumption of the impulse noise reduction at second signal path B, e.g., of the impulse noise detection and / or gain reduction determination. In some cases, the delays can also vary due to a time lag of reaching the constant delay value at second duration 623, as exemplified by delay D1 which is slightly smaller than delays D2, D3. This may be caused, e.g., by the finite attack slope during increasing the delay at first duration 621 before reaching the optimum value. Accordingly, applying the gain reduction may not be fully optimized in a case in which the delay has not yet been adjusted to the optimized value.

[0117] As illustrated in FIG. 10B, when adjusting the delay by applying delay scheme 611, the attack slope may be larger than the release slope. In particular, a rate at which the delay is increased during first duration 621 may be larger than a rate at which the delay is decreased during third duration 625. A larger attack slope can be beneficial for a quicker adjustment of the delay to the value optimized for the gain reduction. In this way, subsequent events of impulse noise may be accounted for even at a short time interval from the preceding event at which the impulse noise has been detected. A smaller release slope can be beneficial for a quicker re-adjustment of the delay to the value optimized for the gain reduction. To illustrate, the smaller release slope may provide for a longer time until the delay returns to base line value 615 during which the delay can be quicker re-adjusted to the optimized value. Thus, within this prolonged time, subsequent events of impulse noise may be effectively reduced when arriving even quicker after the preceding event.

[0118] FIG. 11B illustrates a functional plot 650 of a sequence 651 of gain reductions 661, 671, 681, 691 which may be applied on delayed audio signal 531 to reduce impulse noise. Time of delayed audio signal 531 is indicated on an axis of abscissas. A value of the gain reduction is indicated on an axis of ordinates. FIGS. 11A and 11C display again functional plots 530, 550 of exemplary audio signals 531, 551 for illustrative purposes. Gain reductions 661, 671, 681, 691 may be applied on audio signal 511 at first signal path A, e.g., before, during, and / or after one or more other signal processing routines performed during the audio signal processing.

[0119] After applying one or more gain reductions 661, 671, 681, 691 on delayed audio signal 531, impulse noise can be effectively reduced, as exemplified by events 563, 564 in impulse noise reduced audio signal 551 when compared to events 543, 544 in delayed audio signal 531. In the illustrated examples, amplitudes A3, A4 of events 543, 544 in audio signal 531 have been reduced to amplitudes B3, B4 of events 563, 564 in audio signal 551. This can be attributed to delays D2, D3 delaying impulse noise events 543, 544 to an improved, e.g., optimized, timing at which gain reductions 681, 691 are applied. One or more gain reductions 661 - 691 may then be temporally aligned with events 541 - 544 of impulse noise in delayed audio signal 531, as exemplified by gain reductions 681, 691 which are applied at an equal time at which events 543, 544 occur in delayed audio signal 531. In some examples, a maximum value of gain reductions 681, 691, e.g., a peak value or a constant maximum value, may coincide with the time of events 543, 544 in delayed audio signal 531.

[0120] In some examples, a reduction of impulse noise in delayed audio signal 531 may also occur at a less optimized timing, as exemplified by event 562 in impulse noise reduced audio signal 551 corresponding to event 542 in delayed audio signal 531. In the illustrated example, amplitude A2 of event 542 in audio signal 531 has been reduced to amplitude B2 of event 562 in audio signal 551. The reduction of impulse noise event 542 can be attributed to delay D1, which is shorter than delays D2, D3, but still leads to a reduction of the impulse noise due to an improvement of the timing of applying gain reduction 671 due to delay D1.

[0121] In some examples, after applying one or more gain reductions 661, 671, 681, 691 on delayed audio signal 531, impulse noise may remain unreduced and / or reduced less than a desired degree, as exemplified by event 561 in impulse noise reduced audio signal 551 when compared to event 541 in delayed audio signal 531. In the illustrated example, amplitude B1 of event 561 in audio signal 551 is equal to amplitude A1 of event 541 in audio signal 531 so that the impulse noise remains unreduced. This may be attributed to an applying of delay scheme 611 coming too late for an effective application of one or more gain reductions 661 - 691 so as to reduce impulse noise to a desired extent. In the illustrated example, delay scheme 611 is applied upon detection of initial event 521 in input audio signal 510 so that a delay of initial event 541 in delayed audio signal 531 remains unadjusted. As a result, applying gain reduction 661 on delayed audio signal 531 would also come too late to provide for an effective reduction of impulse noise at initial event 541.

[0122] In some examples, in cases in which applying one or more gain reductions 661, 671, 681, 691 on delayed audio signal 531 would leave the impulse noise unreduced and / or less reduced than the desired degree, applying one or more gain of reductions 661 - 691 may be omitted. In some examples, applying one or more gain reductions 661 - 691 may be omitted upon detecting one or more preceding events 521 - 524 in input audio signal 511 before and / or based on which the delay is adjusted, e.g., before and / or based on which delay scheme 611 is applied. In the illustrated example, applying gain reduction 661 on delayed audio signal 531 for reducing initial event 541 of impulse noise may be omitted.

[0123] In some examples, gain reductions 661, 671, 681, 691 may be applied on delayed audio signal 531 by adding gain reductions 661, 671, 681, 691 to a base line level 655 of gain also applied on the audio signal. As a result, base line level 655 may be reduced by gain reductions 661, 671, 681, 691. In some instances, base line level 655 may be zero gain, or positive gain, or negative gain. In some examples, a positive gain may be applied as a hearing enhancement feature, e.g., to compensate for an individual hearing loss. In some examples, a negative gain may be applied as a hearing protection feature, e.g., in a loud acoustic environment. In some examples, gain reductions 661, 671, 681, 691 may be applied on delayed audio signal 531 by adding gain reductions 661, 671, 681, 691 to a gain function, which may be applied on delayed audio signal 531 during audio processing at first signal path A, e.g., by audio processing module 311, 341, 361. As a result, the gain function may be reduced by gain reductions 661, 671, 681, 691. In some examples, the gain function may amplify the audio signal according to a gain model defining an amplification of the audio signal at different frequencies and / or at different signal levels. In some examples, the gain function may be customized to specific needs and / or an individual hearing loss of the user.

[0124] In some examples, gain reductions 661, 671, 681, 691 each may comprise an attack time 663, 673, 683, 693 of the gain reduction, during which the gain is reduced until reaching a maximum level of reduced gain. The maximum level may be a peak level or may be kept constant for a certain amount of time. Further, gain reductions 661, 671, 681, 691 each may comprise a release time 665, 675, 685, 695 of the gain reduction, during which the gain is increased again until reaching an initial level of unreduced gain. Attack time 663, 673, 683, 693 may also be defined by an attack slope indicative of a slope at which the gain is reduced during the gain reduction until reaching the maximum level. Release time 665, 675, 685, 695 may also be defined by a release slope indicative of a slope at which the gain is increased during the gain increase until reaching the initial level.

[0125] In some examples, when gain reductions 661, 671, 681, 691 are applied in a sequence upon detecting a plurality of events 521 - 524 of impulse noise in input audio signal 511, a newly applied gain reduction 661, 671, 681, 691 may cancel and / or overrule a previously applied gain reduction 661, 671, 681, 691. This is exemplified, in FIG. 11B, by gain reduction 671 which is applied before gain reduction 661 ends, in particular during release time 665 of gain reduction 661.

[0126] In some examples, attack time 663, 673, 683, 693 and / or release time 665, 675, 685, 695 and / or the maximum level of reduced gain may be determined depending on one or more characteristics of events 521 - 524 of impulse noise and / or depending on the adjustment of delay D1, D2, D3 of input audio signal 531 and / or depending on a latency of the audio processing at first signal path A. To illustrate, a larger attack slope during attack time 663 - 669 may be applied for shorter delays D1, D2, D3 and / or a for a smaller latency of the audio processing.

[0127] FIG. 12 illustrates a block flow diagram for an exemplary method of processing an audio signal in a hearing device. The method may be executed, e.g., by processor 104 of system 100 and / or processor 204 of hearing device 200. At operation S11, impulse noise is detected in an input audio signal at one or more preceding times. At operation S12, upon detecting the impulse noise at the one or more preceding times, a delay of the audio signal is adjusted. At operation S13, impulse noise is detected in the audio signal at a subsequent time. At operation S14, when impulse noise is detected at the subsequent time, a gain reduction is applied on the audio signal to reduce a signal level of the impulse noise. The delay at S12 can be provided such that a timing of the applying of the gain reduction can be improved when the impulse noise is detected at the subsequent time.

[0128] While the principles of the disclosure have been described above in connection with specific devices, systems and methods, it is to be clearly understood that this description is made only by way of example and not as limitation on the scope of the invention. The above described preferred embodiments are intended to illustrate the principles of the invention, but not to limit the scope of the invention. Various other embodiments and modifications to those preferred embodiments may be made by those skilled in the art without departing from the scope of the present invention that is solely defined by the claims. In the claims, the word "comprising" does not exclude other elements or steps, and the indefinite article "a" or "an" does not exclude a plurality. A single processor or controller or other unit may fulfil the functions of several items recited in the claims. The mere fact that certain measures are recited in mutually different dependent claims does not indicate that a combination of these measures cannot be used to advantage. Any reference signs in the claims should not be construed as limiting the scope.

Examples

Embodiment Construction

[0038]Methods, systems, and devices of processing an audio signal in a hearing device for reducing impulse noise in the audio signal are described herein. As will be described in more detail below, an exemplary method may comprise detecting impulse noise in the audio signal; applying a gain reduction on the audio signal to reduce a signal level of the impulse noise; and adjusting, upon detecting the impulse noise at one or more preceding times, a delay of the audio signal to improve a timing of the applying of the gain reduction when the impulse noise is detected at a subsequent time.

[0039]By providing methods, systems, and devices such as those described herein, it may be possible to provide for a compatibility of the impulse noise reduction with various audio processing algorithms. Examples may include algorithms operating with a rather low latency, wherein the low latency may be maintained at least under normal conditions and / or at the absence of impulse noise in the audio signal...

TECHNICAL FIELD

[0001] The disclosure relates to a method of processing an audio signal, according to the preamble of claim 1. The disclosure further relates to a hearing system configured to perform the method, according to the preamble of claim 15.BACKGROUND

[0002] Hearing devices may be used to improve the hearing capability or communication capability of a user, for instance by compensating a hearing loss of a hearing-impaired user, in which case the hearing device is commonly referred to as a hearing instrument such as a hearing aid, or hearing prosthesis. A hearing device may also be used to output sound based on an audio signal which may be communicated by a wire or wirelessly to the hearing device. A hearing device may also be used to reproduce a sound in a user's ear canal detected by an input transducer such as a microphone or a microphone array. The reproduced sound may be amplified to account for a hearing loss, such as in a hearing instrument, or may be output without accounting for a hearing loss, for instance to provide for a faithful reproduction of detected ambient sound and / or to add audio features of an augmented reality in the reproduced ambient sound, such as in a hearable. A hearing device may also provide for a situational enhancement of an acoustic scene, e.g. beamforming and / or active noise cancelling (ANC), with or without amplification of the reproduced sound. A hearing device may also be implemented as a hearing protection device, such as an earplug, configured to protect the user's hearing. Different types of hearing devices configured to be be worn at an ear include earbuds, earphones, hearables, and hearing instruments such as receiver-in-the-canal (RIC) hearing aids, behind-the-ear (BTE) hearing aids, in-the-ear (ITE) hearing aids, invisible-in-the-canal (IIC) hearing aids, completely-in-the-canal (CIC) hearing aids, cochlear implant systems configured to provide electrical stimulation representative of audio content to a user, a bimodal hearing system configured to provide both amplification and electrical stimulation representative of audio content to a user, or any other suitable hearing prostheses. A hearing system comprising two hearing devices configured to be worn at different ears of the user is sometimes also referred to as a binaural hearing device. A hearing system may also comprise a hearing device, e.g., a single monaural hearing device or a binaural hearing device, and a user device, e.g., a smartphone and / or a smartwatch, communicatively coupled to the hearing device.

[0003] Hearing devices are often employed in conjunction with communication devices, such as smartphones or tablets, for instance when listening to sound data processed by the communication device and / or during a phone conversation operated by the communication device. More recently, communication devices have been integrated with hearing devices such that the hearing devices at least partially comprise the functionality of those communication devices. A hearing system may comprise, for instance, a hearing device and a communication device.

[0004] Advanced audio signal processing techniques have been developed to improve a hearing experience for users in various fields. For example, hearing instruments rely on audio signal processing to enhance auditory perception for individuals with hearing impairments.

[0005] Some important audio processing applications address impulse noise. Impulse noise refers to a type of noise which may be characterized by a transient event of a high intensity burst of sound. These bursts can occur suddenly and are usually of very brief duration, often appearing as spikes in the audio signal. Impulse noise can be caused by various factors, including electrical interference, mechanical disturbances, or environmental events such as clicks, pops, or other transient noises. Examples include noise caused by a switching of mechanical or electronic devices, gunshots or firecrackers, applause, clapping, door slams, scratching of cutlery, clinking of glasses, and / or the like. Impulse noise can be harmful for many reasons including an auditory disruption causing discomfort or even pain for the user, a signal distortion compromising a signal integrity and / or masking important audio information, a damaging of audio equipment caused by the sudden and extreme amplitude changes, a communication interference disrupting a clarity and intelligibility of speech, and a data corruption affecting a quality of audio reproduction and transmission.

[0006] Traditional audio processing methods aiming to reduce impulse noise in the audio signal typically involve a combination of multiple operations including a detection of the impulse noise, a calculation of a gain reduction, and an application of the gain reduction. Impulse noise detection involves recognizing the occurrence of an acoustic shock and assessing its intensity. This process often incorporates multiple measures to ensure accurate detection. Gain calculation aims to determine an appropriate strength and temporal slope for a change of gain to be applied to the audio signal. This calculation is crucial for ensuring that the transient event is perceived naturally by the user, without causing discomfort. In many cases, unlike complete suppression, the objective is to modulate the transient event to a level that is both noticeable and tolerable. Finally, the gain application step involves applying the calculated gain change, which may vary over time. In particular, the gain modification may be applied relative to an existing gain function within the signal path. This ensures that the transient event is handled dynamically, maintaining the integrity of the overall auditory experience.

[0007] For instance, according to WO 2010 / 083879 A1, the impulse noise detection mechanism includes measuring a peak difference of a signal upstream of a band split filter bank and comparing the peak difference against a peak difference limit. US 2003 / 0031335 A1 discloses an adaptive setting of a threshold value depending on a level of an input audio signal which can limit the transient noise contained in an output audio signal.

[0008] One of the effects of advanced audio processing techniques can be a latency leading to a delay between a time at which an audio signal is received by a processor and a time at which the processed signal is output, e.g., to an audio output unit for stimulating a user's hearing. Latency can impact the user experience because it may affect a synchronization between what users see and hear and may give rise to a comb filter effect. For instance, latency may limit an understanding of speech, a lip-synchronization of speech contained in the audio signal, an environmental sound awareness, and / or a user's listening comfort. Latency may also contribute to an increased perception of echo and feedback. Thus, to ensure optimal performance and user satisfaction, an effort is often made to minimize the latency so as to allow a faster processing, e.g., in real-time. This may involve a balancing of the need for advanced processing features, such as noise reduction and speech enhancement, with the requirement for low-latency operation.

[0009] One side effect of the processing latency is that it often provides a "look ahead" capability in a sense that a desired modification of an input audio signal can be performed rather spontaneously after the latency period while performing a required signal evaluation and / or preparation beforehand during the latency period. As an example, when performing the audio processing in a frequency domain, the inherent processing latency associated with the frequency domain transformation may significantly contribute to provide such a "look ahead" capability. In many cases, when performing a reduction of impulse noise in the audio signal, this latency also allows sufficient time for the impulse noise detection and gain calculation components to perform their respective functions effectively, ensuring timely and accurate reduction of impulse noise by the gain application component.

[0010] In modem trends, however, which aim to further reduce the processing latency as much as possible, such a timely and accurate application of a target gain for reducing impulse noise can be rather challenging, at least when a prior detection of impulse noise and / or an according determination of an appropriate gain function, which may be customized to the detected impulse noise and / or a currently applied gain function, shall be performed in a reliable manner. Nevertheless, as reduced latency and a reduction of impulse noise are both critical factors influencing the hearing experience for a user, as described above, it would be desirable to optimize the audio signal processing in a way to provide for the combined advantages of both processing schemes as much as possible whereas the inherent disadvantages of each processing scheme are kept at a minimum.SUMMARY

[0011] It is an object of the present disclosure to provide for an improvement of current audio processing methods for impulse noise reduction, in particular in a way that the desired processing outcome of reducing an impulse noise in an audio signal is less dependent on an inherent latency caused by the signal processing and / or more compatible with and / or less affected by a signal processing of a rather low and / or varying latency. It is another object to propose a method for audio processing and / or a hearing system and / or a hearing device in which transient events of impulse noise can be handled in an effective and / or adaptive way, e.g., by also optimizing an impact on other processing schemes applied on the audio signal and / or advantages thereof. It is a further object to propose a method and / or system and / or device allowing to identify an increased likelihood of forthcoming transient events of impulse noise in the audio signal and / or to provide for impulse noise reduction in the audio signal when its likelihood is increased. It is a further object to improve audio signal processing in a hearing system and / or hearing device in way that improves a hearing experience for the user.

[0012] At least one of these objects can be achieved by a method of processing an audio signal comprising the features of patent claim 1 and / or a hearing device for performing the method comprising the features of patent claim 15. Advantageous embodiments are defined by the dependent claims and the following description.

[0013] Accordingly, the present disclosure proposes a of processing an audio signal comprising detecting impulse noise in the audio signal; applying a gain reduction on the audio signal to reduce a signal level of the impulse noise; and adjusting, upon detecting the impulse noise at one or more preceding times, a delay of the audio signal to improve a timing of the applying of the gain reduction when the impulse noise is detected at a subsequent time.

[0014] Independently, the present disclosure proposes a non-transitory computer-readable medium storing instructions that, when executed by a processor, which may be included in a hearing device, cause a hearing device to perform the method.

[0015] Independently, the present disclosure proposes a hearing system comprising a hearing device configured to be worn at an ear of a user, the hearing system comprising an audio input unit for obtaining an input audio signal; a processing unit for audio signal processing of the input audio signal to obtain an output audio signal; and an audio output unit for outputting the output audio signal so as to stimulate the user's hearing, wherein the audio input is included in the hearing device and the processor is configured to perform the method based on the input audio signal. E.g., the method may be performed by providing the input audio signal obtained by the audio input unit to the processing unit for the processing of the input audio signal.

[0016] In some implementations, the hearing system may comprise the hearing device and / or a remote device communicatively coupled to the hearing device. E.g., the audio input unit and / or the processor may be included in the hearing device and / or the remote device. In some examples, the hearing system may be provided as the hearing device, wherein the audio input unit and the processor are included in the hearing device. Independently, the present disclosure thus also proposes a hearing device configured to be worn at an ear of a user, the hearing device comprising an audio input unit for obtaining an input audio signal; a processing unit for audio signal processing of the input audio signal to obtain an output audio signal; and an audio output unit for outputting the output audio signal so as to stimulate the user's hearing, wherein the processor is configured to perform the method based on the input audio signal.

[0017] Subsequently, additional features of some implementations of the method and / or the hearing device and / or the computer readable medium are described. Each of those features can be provided solely or in combination with at least another feature. The features can be correspondingly provided in some implementations of the method and / or the hearing device and / or the computer readable medium.

[0018] In some implementations, the delay is adjusted by applying a delay scheme on the audio signal, the delay scheme defining a minimum duration during which the delay adjustment is maintained. E.g., the delay scheme may define a minimum duration during which the delay is kept at an adjusted value differing from a value of the delay before the adjusting of the delay.

[0019] In some implementations, when the delay scheme is applied, the delay scheme is updated upon detecting the impulse noise at the subsequent time, wherein the updating of the delay scheme comprises extending said minimum duration. In some examples, the minimum duration is extended by restarting the delay, e.g., from time zero, for a prolongation period. The prolongation period may correspond to the minimum duration of the delay scheme applied before the updating of the delay scheme and / or may be different from this minimum duration, e.g., it may be larger.

[0020] In some implementations, the delay scheme further defines an optimum value of the delay, the optimum value minimizing a difference between a time at which the impulse noise occurs in the audio signal and a time at which the gain reduction is applied on the audio signal.

[0021] In some implementations, the delay scheme further defines a duration and / or rate in which the delay is decreased over time from the optimum value to return to a value of the delay before the adjusting of the delay; and / or a duration and / or rate in which the delay is increased over time from a value of the delay before the adjusting of the delay to the optimum value; and / or a duration during which the delay is kept constant at the optimum value. In some examples, the minimum duration during which the delay adjustment is maintained comprises the duration during which the delay is kept constant and / or the duration in which the delay is decreased and / or increased. E.g., the duration in which the delay is decreased and / or increased may be defined by the rate in which the delay is decreased and / or increased to the optimum value.

[0022] In some implementations, when updating the delay scheme, the delay scheme may be applied again with equal properties, e.g., with an equal duration during which the delay is kept constant and / or an equal duration in which the delay is decreased and / or increased, so as to extend the minimum duration by the duration during which the delay is kept constant and / or the duration in which the delay is decreased and / or increased. In some examples, when updating the delay scheme, one or more properties of the delay scheme may altered before applying the delay scheme. The minimum duration may then be extended by an altered duration during which the delay is kept constant and / or an altered duration in which the delay is decreased and / or increased.

[0023] In some implementations, the method further comprises determining the delay scheme depending on a latency caused by the audio signal processing; and / or a current value of the delay; and / or a number of the preceding times at which impulse noise has been detected; and / or an intermediate time interval between impulse noise detected at two or more of the preceding times; and / or a time elapsed since the impulse noise has been detected at one or more of the preceding times. In some examples, when a latency of the audio signal processing is larger, the optimum value of the delay may be determined to be smaller and / or the duration and / or rate in which the delay is increased over time may be determined to be smaller. In some examples, when the number of preceding times at which impulse noise has been detected is larger and / or the intermediate time interval is smaller and / or the time elapsed is smaller, the minimum duration of maintaining the delay adjustment may be determined to be larger.

[0024] In some implementations, the audio signal is processed in a time domain. In some examples, the audio signal processing may be performed exclusively in the time domain. In some implementations, the audio signal may be transformed into another domain, e.g., a frequency domain, wherein at least part of the audio signal processing is performed in the other domain.

[0025] In some implementations, when detecting the impulse noise at one or more preceding times, the applying of the gain reduction on the audio signal is omitted. In some examples, the applying of the gain reduction on the audio signal may be omitted when initially detecting the impulse noise, e.g., at the temporally first preceding time which may also be denoted as the initial time. In some examples, the applying of the gain reduction on the audio signal may be omitted when detecting the impulse noise at two or more of the preceding times, e.g., the initial time and the preceding time following the initial time.

[0026] In some implementations, the processing of the audio signal comprises applying a gain function on the audio signal, wherein the gain reduction is applied in addition to the gain function. In some examples, the gain function may provide for a frequency dependent amplification of the audio signal, e.g., to compensate for a hearing loss of the user. For instance, the gain function may be fitted to the hearing loss of an individual user. In some examples, the gain function may provide for a positive gain of the audio signal and / or a negative gain of the audio signal.

[0027] In some implementations, the method further comprises determining the gain reduction, wherein the gain reduction is determined so as to meet a target gain representative of an intended signal level of the impulse noise in the audio signal. In some examples, when the processing of the audio signal comprises applying a gain function on the audio signal, the target gain may correspond to a gain which results in applying the gain reduction in addition to the gain function, e.g., to a sum of the gain function and the gain reduction.

[0028] In some implementations, the determining the gain reduction comprises determining a maximum value of the gain reduction; and / or a duration in which the gain reduction is maintained at the maximum value; and / or a rate and / or duration in which the gain reduction is increased over time, e.g., before the gain reduction reaches the maximum value; and / or a rate and / or duration in which the gain reduction is decreased over time, e.g., after the gain reduction reaches the maximum value.

[0029] In some implementations, the gain reduction is determined depending on one or more characteristics of the impulse noise; and / or a latency caused by the audio signal processing; and / or a current value of the delay. In some examples, one or more characteristics of impulse noise may comprise a signal level and / or a duration of the impulse noise and / or a frequency of sound contained in the impulse noise and / or a frequency in which the impulse noise occurs over time in the audio signal. In some examples, when the one or more characteristics of the impulse noise comprise a signal level of the impulse noise and the signal level is larger, the maximum value of the gain reduction is determined to be larger. In some implementations, when the latency caused by the audio signal processing and / or the current value of the delay is smaller, the rate in which the gain reduction is increased over time is determined to be larger.

[0030] In some implementations, a time consumption caused by the detecting of the impulse noise is adjusted depending on a time elapsed since the impulse noise has been detected at one or more of the preceding times; and / or an intermediate time interval between impulse noise detected at two or more of the preceding times; and / or a number of the preceding times at which impulse noise has been detected; and / or one or more characteristics of impulse noise detected at one or more of the preceding times. In some examples, the time consumption caused by the detecting of the impulse noise may be reduced upon detecting the impulse noise at one or more preceding times, e.g., at the initial time. To illustrate, after detecting the impulse noise at the one or more preceding times, an increased likelihood of future occurrences of impulse noise may be employed to reduce the time consumption caused by the detecting of the impulse noise.

[0031] In some implementations, a time consumption caused by the determining of the gain reduction is adjusted depending on a time elapsed since the impulse noise has been detected at one or more of the preceding times; and / or an intermediate time interval between impulse noise detected at two or more of the preceding times; and / or a number of the preceding times at which impulse noise has been detected; and / or one or more characteristics of impulse noise detected at one or more of the preceding times. In some examples, the time consumption caused by the determining of the gain reduction may be reduced upon detecting the impulse noise at one or more preceding times, e.g., at the initial time. To illustrate, after detecting the impulse noise at the one or more preceding times, an increased likelihood of future occurrences of impulse noise with similar characteristics, which may require corresponding properties of the gain reduction, may be employed to reduce the time consumption caused by the determining of the gain reduction.

[0032] In some implementations, when adjusting the time consumption caused by the detecting of the impulse noise and / or by the determining of the gain reduction, the delay of the audio signal is adjusted to account for the adjusted time consumption.

[0033] In some implementations, the processing of the audio signal is performed at a first signal path at which the delay of the audio signal is adjusted and at which the gain reduction is applied on the audio signal, and the impulse noise in the audio signal is detected at a second signal path bypassing the first signal path. In some examples, the determining the delay scheme and / or the determining the gain reduction is also performed at the second signal path.

[0034] In some implementations, the delay may be adjusted upon detecting the impulse noise at the initial time. In some implementations, the delay may be adjusted upon detecting the impulse noise at two or more preceding times, e.g., at the initial time and the following preceding time.

[0035] In some implementations, the hearing device is configured to be worn at an ear of the user. In some examples, the hearing device may be configured to be at least partially inserted into an ear canal of the user.

[0036] In some implementations, by the adjusting of the delay, a latency of the audio processing prior to the adjusted delay may be arbitrary, e.g., rather low and / or vary over time. In this way, advantages of low latency and / or varying latency audio processing may be preserved and / or exploited. Moreover, transient events of impulse noise, which may occur in the audio signal in a random and / or unpredictable manner, can be handled effectively and / or in an adaptive manner. In this way, an impact on other processing schemes applied on the audio signal and / or advantages thereof, e.g., with regard to a desired latency, may be reduced. In particular, an impact may be restricted to the events at which impulse noise is detected at the one or more preceding times and / or at which impulse noise occurs with the increased likelihood. This may contribute to an improved listening experience for the user.BRIEF DESCRIPTION OF THE DRAWINGS

[0037] Reference will now be made in detail to embodiments, examples of which are illustrated in the accompanying drawings. The drawings illustrate various embodiments and are a part of the specification. The illustrated embodiments are merely examples and do not limit the scope of the disclosure. Throughout the drawings, identical or similar reference numbers designate identical or similar elements. In the drawings: Fig. 1schematically illustrates an exemplary impulse noise reduction system; Fig. 2schematically illustrates an exemplary hearing device implementing the impulse noise reduction system illustrated in Fig. 1; Fig. 3schematically illustrates an embodiment of the hearing device illustrated in Fig. 2 as a RIC hearing aid; Figs. 4-8schematically illustrate block diagrams of exemplary signal processing algorithms for impulse noise reduction; Figs. 9A - Cillustrate exemplary audio signals which are processed according to principles described herein; Figs. 10A - Cillustrate, in Figs. 10A, C, exemplary audio signals, and, in Fig. 10B, a delay scheme for adjusting a delay of the audio signal illustrated in Fig. 10A to obtain the audio signal illustrated in Fig. 10C; Figs. 11A - Cillustrate, in Figs. 11A, C, exemplary audio signals, and, in Fig. 10B, a sequence of gain reductions applied on the audio signal illustrated in Fig. 11A to obtain the audio signal illustrated in Fig. 11C; and Fig. 12schematically illustrates an exemplary method of processing an audio signal according to principles described herein. DETAILED DESCRIPTION OF THE DRAWINGS

[0038] Methods, systems, and devices of processing an audio signal in a hearing device for reducing impulse noise in the audio signal are described herein. As will be described in more detail below, an exemplary method may comprise detecting impulse noise in the audio signal; applying a gain reduction on the audio signal to reduce a signal level of the impulse noise; and adjusting, upon detecting the impulse noise at one or more preceding times, a delay of the audio signal to improve a timing of the applying of the gain reduction when the impulse noise is detected at a subsequent time.

[0039] By providing methods, systems, and devices such as those described herein, it may be possible to provide for a compatibility of the impulse noise reduction with various audio processing algorithms. Examples may include algorithms operating with a rather low latency, wherein the low latency may be maintained at least under normal conditions and / or at the absence of impulse noise in the audio signal. Examples may also include algorithms causing variable processing delays, for instance when activating and deactivating optional processing features. Examples may also include algorithms for audio processing in a time domain, algorithms for audio processing in a frequency domain, algorithms providing for hearing loss compensation, beamforming, noise cancellation, speech enhancement, audio reproduction and / or augmentation, and / or the like.

[0040] In some implementations, methods, systems, and devices such as those described herein may be employed in audio processing with arbitrary low latency, wherein the latency lowness may only be limited in situations in which a presence of impulse noise in the audio signal is more likely than in other situations and / or wherein the latency can be selected not to exceed a value which is minimally required for an efficient reduction of the impulse noise.

[0041] To illustrate, the latency may be reduced by turning off time-consuming processing operations depending on specific circumstances, e.g., a current acoustic environment and / or listening intention of the user, in which their presumed benefit is rather low. In some examples, latency may be reduced by performing the processing without converting the audio signal to another domain, such as a frequency domain, but processing the audio signal in its original temporal form, which may also be referred to as a processing in a time domain.

[0042] In some implementations, methods, systems, and devices such as those described herein may be employed to determine situations in which impulse noise is more likely to be contained in the audio signal than in other situations and / or to modify impulse noise when its likelihood is increased. This may result in an improved hearing experience for a user, e.g., by optimizing an impact of impulse noise reduction in a processing scheme optimized for the user in various other aspects.

[0043] To illustrate, by adjusting the delay of the audio signal upon detecting the impulse noise at the one or more preceding times, an impact of the delay adjustment on the audio signal processing can be automatically adapted, e.g., reserved, to situations of an increased likelihood in which impulse noise may occur. An assumption of an increased likelihood of impulse noise after a preceding detection may be based on the observation, that, in real life situations, impulse noise often appears in a cumulative manner, e.g., in a temporal clustering after an initial detection. Thus, when impulse noise has been detected at the one or more preceding times, it can be suspected that it will occur again at the subsequent time, at least with an increased probability as compared to a situation in which no impulse noise has been previously detected. This may imply that the detection of the impulse noise at the one or more preceding times is employed as an indicator of the increased likelihood of subsequent impulse noise.

[0044] Those and other implementations and advantages are further described in the description that follows.

[0045] FIG. 1 illustrates an exemplary impulse noise reduction system 100 that may be implemented according to principles described herein. As shown, system 100 may include, without limitation, a memory 102 and a processor 104 selectively and communicatively coupled to one another. Memory 102 and processor 104 may each include or be implemented by hardware and / or software components (e.g., processors, memories, communication interfaces, instructions stored in memory for execution by the processors, etc.). In some examples, memory 102 and / or processor 104 may be implemented by any suitable computing device. In other examples, memory 102 and / or processor 104 may be distributed between multiple devices and / or multiple locations as may serve a particular implementation. Illustrative implementations of system 100 are described herein.

[0046] Memory 102 may maintain (e.g., store) executable data used by processor 104 to perform any of the operations described herein. For example, memory 102 may store instructions 106 that may be executed by processor 104 to perform any of the operations described herein (e.g., instructions to detect impulse noise and / or to apply a gain reduction and / or to adjust a delay). Instructions 106 may be implemented by any suitable application, software, code, and / or other executable data instance. Memory 102 may also maintain any data received, generated, managed, used, and / or transmitted by processor 104. Memory 102 may store any other suitable data as may serve a particular implementation. For example, memory 102 may store data associated with parameters applied in an audio signal processing performed by processor 104 (e.g., information about a delay and / or a gain reduction to be applied on the audio signal and / or a delay and / or gain reduction currently applied and / or previously).

[0047] Processor 104 may be configured to perform (e.g., execute instructions 106 stored in memory 102 to perform) various processing operations associated with processing an audio signal including a reduction of impulse noise in the audio signal. For example, processor 104 may perform one or more operations on the audio signal described herein, e.g., to detect impulse noise, apply a gain reduction, and adjust a delay.

[0048] System 100 may be implemented in any suitable manner. For example, system 100 may be implemented as a hearing device, a communication device and / or a remote microphone communicatively coupled to the hearing device, e.g., via radio frequency (RF) radiation, or a combination of the hearing device and the communication device and / or the remote microphone.

[0049] FIG. 2 illustrates an exemplary implementation of system 100 as a hearing device 200 configured to be worn at an ear of a user. Hearing device 200 may be implemented by any type of hearing device configured to enable or enhance hearing or a listening experience of a user wearing hearing device 200. For example, hearing device 200 may be implemented by a hearing aid configured to provide an amplified version of audio content to a user, a sound processor included in a cochlear implant system configured to provide electrical stimulation representative of audio content to a user, a sound processor included in a bimodal hearing system configured to provide both amplification and electrical stimulation representative of audio content to a user, or any other suitable hearing prosthesis, or an earbud or an earphone or a hearable.

[0050] In certain examples, hearing device 200 may be implemented as part of a binaural hearing system. Such a binaural hearing system may include a first hearing device associated with a first ear of a user and a second hearing device associated with a second ear of a user. In such examples, the hearing devices may each be implemented by any type of hearing device configured to provide or enhance hearing to a user of a binaural hearing system. In some examples, the hearing devices in a binaural system may be of the same type. For example, the hearing devices may each be hearing aid devices. In certain alternative examples, the hearing devices may be of a different type. For example, a first hearing device may be a hearing aid and a second hearing device may be a sound processor included in a cochlear implant system.

[0051] Different types of hearing device 200 can also be distinguished by the position at which they are worn at the ear. Some hearing devices, such as behind-the-ear (BTE) hearing aids and receiver-in-the-canal (RIC) hearing aids, typically comprise an earpiece configured to be at least partially inserted into an ear canal of the ear, and an additional housing configured to be worn at a wearing position outside the ear canal, in particular behind the ear of the user. Some other hearing devices, as for instance earbuds, earphones, hearables, in-the-ear (ITE) hearing aids, invisible-in-the-canal (IIC) hearing aids, and completely-in-the-canal (CIC) hearing aids, commonly comprise such an earpiece to be worn at least partially inside the ear canal without an additional housing for wearing at the different ear position.

[0052] Hearing device 200 may include, without limitation, a memory 202 and a processor 204 selectively and communicatively coupled to one another. Memory 202 and processor 204 may each include or be implemented by hardware and / or software components (e.g., processors, memories, communication interfaces, instructions stored in memory for execution by the processors, etc.).

[0053] Memory 202 may maintain (e.g., store) executable data used by processor 204 to perform any of the operations associated with hearing device 200. For example, memory 202 may store instructions 206 that may be executed by processor 204 to perform any of the operations associated with hearing device 202 assisting a user in hearing and / or any of the operations described herein, which may include instructions 106.

[0054] In some examples, instructions 206 may include routines for audio signal processing in a time domain. In some examples, instructions 206 may include routines for audio signal processing in a frequency domain, e.g., instructions to convert an audio signal from the time domain into the frequency domain and / or vice versa. In some examples, instructions 206 may include a gain function defining an amplification of an audio signal at different frequencies and / or at different signal levels, e.g., when hearing device 200 is implemented by a hearing instrument, to compensate for a hearing loss. In some examples, instructions 206 may include audio signal processing routines providing for noise cancelling, feedback cancelling, beamforming, speech enhancement, audio signal classification, performing an audio signal processing program associated with a current acoustic scene and / or depending on the audio signal classification, own voice detection, acoustic object separation, binaural synchronization, and / or the like. Instructions 206 may be implemented by any suitable application, software, firmware, code, and / or other executable data instance.

[0055] Processor 204 may be configured to perform any suitable processing operation associated with hearing device 200. For example, when hearing device 200 is implemented by a hearing instrument, such processing operations may include monitoring ambient sound and / or presenting amplified sound to user 204 via an in-ear receiver. Processor 210 may be implemented by any suitable combination of hardware and software.

[0056] As shown in FIG. 2, hearing device 200 may further include an audio input unit 213 and an audio output unit 217 communicatively coupled to processor 204. Audio input unit 213 is configured to obtain an input audio signal. Processor 204 is configured to provide for a processing of the input audio signal to obtain an output audio signal. Audio output unit 217 is configured to output sound based on the output audio signal.

[0057] In some implementations, as illustrated, audio input unit 213 may comprise a sound detector 215 configured to detect sound in an ambient environment of the user and to provide an ambient audio signal representative of the detected sound. The input audio signal, which is received by processor 204, may then at least partially be based on the ambient audio signal. In some examples, sound detector 215 may be implemented as a microphone and / or a microphone array. In some examples, after detection of the sound in the ambient environment, audio input unit 213 may be configured to prepare the ambient audio signal for an audio signal processing by processor 204. For example, audio input unit 217 may comprise an analog-to-digital converter to convert the ambient audio signal, as detected by sound detector 215, from an analog signal into a digital signal.

[0058] In some implementations, as illustrated, audio input unit 213 may comprise a radio receiver 216 configured to receive a radio audio signal from a remote audio source via radio frequency (RF) radiation. The input audio signal, which is received by processor 204, may then at least partially be based on the radio audio signal. Radio receiver 216 may be configured for wireless data reception of the radio audio signal. For instance, the radio audio signal may be received in accordance with a Bluetooth ™< protocol and / or by any other type of RF communication. In some examples, the remote audio source may be a remote microphone, e.g., a table microphone or a clip-on microphone, configured to detect sound at a remote location and transmit the radio audio signal indicative of the detected sound to radio receiver 216. In some examples, the remote audio source may be a streaming source configured for streaming the radio audio signal to radio receiver 216. In some examples, the remote audio source may be a communication device, e.g., a portable device such as a smartphone, tablet, smartwatch and / or the like, or a computing device such as a personal computer, configured for data transmission of the radio audio signal to radio receiver 216. In some examples, after reception of the radio audio signal, audio input unit 213 may be configured to prepare the radio audio signal for an audio signal processing by processor 204. For example, when the radio audio signal received from the remote audio source comprises an encoded signal, audio input unit 213 may comprise a decoder to decode the radio audio signal.

[0059] Audio output unit 217 may be implemented by any suitable audio output device configured to output sound based on the output audio signal to the user. To this end, audio output unit 217 may include an output transducer. For example, audio output unit 119 may be implemented as a receiver of a hearing aid, a loudspeaker of an earbud, or an output electrode of a cochlear implant.

[0060] Hearing device 200 may include further components as may serve a particular implementation. E.g., hearing device 200 may further include a user interface and / or a communication port for data transmission and / or an ear-canal microphone and / or other sensors such as a motion sensor and / or a physiological sensor.

[0061] FIG. 3 illustrates an exemplary implementation of hearing device 200 as a RIC hearing aid 261. RIC hearing aid 261 comprises a BTE part 270 configured to be worn at an ear at a wearing position behind the ear, and an ITE part 280 configured to be worn at the ear at a wearing position at least partially inside an ear canal of the ear. BTE part 270 comprises a BTE housing 271 configured to be worn behind the ear. BTE housing 271 accommodates a processing unit 264, which may comprise processor 204 and memory 202, communicatively coupled to sound detector 215 and radio receiver 216. BTE part 220 further includes a battery 277 as a power source. ITE part 280 is an earpiece comprising an ITE housing 281 at least partially insertable into the ear canal. ITE housing 281 accommodates audio output unit 217 implemented as a receiver. BTE part 270 and ITE part 280 are interconnected by a cable 274. Processing unit 264 is communicatively coupled to audio output unit 217 of ITE part 280 via cable 274 and cable connectors 272, 273 provided at BTE housing 271 and ITE housing 281.

[0062] FIG. 4 is a schematic block diagram of a signal processing algorithm 301 for impulse noise reduction in an audio signal, which may be implemented by system 100 and / or hearing device 200. E.g., algorithm 301 may be executed by processor 104, 204 after receiving an audio signal from an audio input unit 302, which may include sound detector 215 and / or radio receiver 216, as an input audio signal. Processor 104, 204 may process the input audio signal to obtain an output audio signal. The output audio signal may be output by an audio output unit 303 to stimulate the user's hearing.

[0063] Algorithm 301 may comprise an audio processing module 311 and an impulse noise reduction module 321. Audio processing module 311 and impulse noise reduction module 321 may perform a processing of the audio signal in parallel to one another. E.g., as illustrated audio processing module 311 may receive the audio signal from audio input unit 302 via a first signal path A, and impulse noise reduction module 321 may receive the audio signal via a second signal path B. E.g., signal path B at which the impulse noise reduction is performed may at least partially bypass signal pass A at which the audio processing by audio processing module 311 is performed.

[0064] Audio processing module 311 may be configured to execute one or more audio processing routines as may serve a particular implementation. In some examples, the processing routines may include one or more of applying a gain function on the audio signal, which may define an amplification characteristic, e.g., to compensate for an individual hearing loss of the user; a noise cancelling algorithm; a reverberation cancelling algorithm; a wind noise cancelling algorithm; a feedback cancelling algorithm; a speech enhancement algorithm; an impulse noise cancelling algorithm; an acoustic object separation algorithm; an audio signal classification algorithm and / or one or more processing algorithms optimized for an audio signal assigned to a classification; a binaural synchronization algorithm; a beamforming algorithm, in particular adapted for static and / or adaptive beamforming; an algorithm based on a neural network (NN); a music reproduction algorithm; and / or any other algorithm for modifying and / or enhancing the audio signal. In some examples, the processing routines may also comprise a conversion of the audio signal between a time domain and a frequency domain, e.g., to provide for a processing in the frequency domain by one or more of the processing routines. In some examples, one or more of the processing routines may be executed in a time domain.

[0065] Impulse noise reduction module 321 may comprise an impulse noise detection module 323, a delay adjustment module 325, and a gain modification module 327. Impulse noise detection module 323 is configured, after receiving the audio signal, to detect impulse noise in the audio signal. Impulse noise may be defined as a transient event of sound with high intensity. Key characteristics of impulse noise may include a short duration, e.g., often in a range of milliseconds, a high amplitude, in particular an amplitude significantly higher than a background noise level, and a random occurrence, e.g., in an unpredictable manner and / or without a regular pattern. Detecting the impulse noise may comprise determining an instance of a sudden increase of a level of the audio signal and / or any other detection method known in the art, for example as disclosed in WO 2010 / 083879 A1 and / or US 2003 / 0031335.

[0066] Gain modification module 327 is configured, upon detection of impulse noise, to provide for a gain reduction which can be applied on the audio signal to reduce a signal level of the impulse noise. The gain reduction may be applied, as illustrated, in the course of the audio processing performed by audio processing module 311, e.g., during or after one or more audio processing routines executed by audio processing module 311, and / or the gain reduction may be applied on the audio signal at first signal path A which may be executed in parallel to second signal path B at which impulse noise is detected in the audio signal.

[0067] In some implementations, gain modification module 327 is configured to determine the gain reduction. Determining the gain reduction may comprise determining, e.g., calculating, a maximum value of the gain reduction; and / or a duration in which the gain reduction is maintained at the maximum value; and / or a rate and / or duration in which the gain reduction is increased over time, e.g., before the gain reduction reaches the maximum value; and / or a rate and / or duration in which the gain reduction is decreased over time, e.g., after the gain reduction reaches the maximum value. In some examples, the gain reduction may be determined depending on one or more characteristics of the impulse noise in the audio signal, e.g., as detected by impulse noise detection module 323. The characteristics may include, e.g., an amplitude and / or a duration and / or a frequency and / or a rate of the impulse noise.

[0068] In some examples, the gain reduction may be determined depending on a target gain representative of an intended signal level of the impulse noise in the audio signal. For instance, when a gain function is applied on the audio signal in the course of the audio processing performed by audio processing module 311 and / or a gain function is applied on the audio signal at first signal path A, the gain reduction may be determined depending on the gain function so as to provide for the intended target gain. In some examples, gain modification module 327 may also be configured to determine the target gain, e.g., depending on one or more characteristics of the impulse noise in the audio signal.

[0069] The audio processing performed by audio processing module 311 at signal path A, e.g., when executing one or more of the audio processing routines, may produce a latency. The processing latency generally leads to an intrinsic delay between a time at which the audio signal is received, e.g., from audio input unit 302, and a time at which the processed signal is provided as an output signal, e.g., to audio output unit 303. Detecting impulse noise in the audio signal, as performed by impulse noise detection module 323 at signal path B, and / or determining the gain reduction, which may be performed by gain modification module 327 at signal path B, also consumes time and therefore produces a latency, which may be denoted as an impulse noise reduction latency. The impulse noise reduction latency leads to an intrinsic delay between a time at which the audio signal is received and a time at which the gain reduction can be applied on the audio signal to reduce the signal level of the impulse noise. In a case in which the processing latency at signal path A is lower than the impulse noise reduction latency at signal path B, e.g., when audio processing with low latency is performed by audio processing module 311, an application of the gain reduction on the audio signal at signal path A may be too late for reducing the signal level of the impulse noise.

[0070] Delay adjustment module 325 is configured to provide for adjusting, upon detecting the impulse noise at one or more preceding times by impulse noise detection module 323, a delay of the audio signal. As illustrated, the delay of the audio signal may be adjusted in the course of the audio processing performed by audio processing module 311, e.g., during or after one or more audio processing routines executed by audio processing module 311, and / or the delay of the audio signal may be adjusted at first signal path A. By adjusting the delay of the audio signal, a timing of the applying of the gain reduction on the audio signal can be improved, at least when the impulse noise is detected at a subsequent time. To illustrate, in the case in which the processing latency at signal path A is lower than the impulse noise reduction latency at signal path B, the delay of the audio signal may be adjusted accordingly, e.g., so as to minimize a difference between the processing latency at signal path A and the impulse noise reduction latency at signal path B. In particular, a difference between a time at which the impulse noise is contained in the audio signal at signal path A and a time at which the gain function is applied on the audio signal at signal path A may thus be minimized.

[0071] In some examples, e.g., in the case in which the processing latency at signal path A is lower than a latency at signal path B for adjusting the delay at signal path A, the adjusting of the delay may be too slow for providing the improved timing for applying the gain reduction so as to reduce the impulse noise detected at one or more of the preceding times. In those cases, however, providing and / or maintaining the delay of the audio signal at signal path A can be beneficial to provide the improved timing for applying the gain reduction to reduce the impulse noise detected at a subsequent time.

[0072] In this context, the circumstance may be exploited that impulse noise often appears in a cumulative manner, e.g., in a temporal clustering. Thus, after detection of impulse noise in the audio signal at an initial time and / or after detection at two or more of the preceding times, at may be expected that impulse noise will also subsequently occur in the audio signal with an increased likelihood, at least as compared to an average occurrence of impulse noise over a longer time period such as hours or days. Detecting of impulse noise at one or more preceding times by impulse noise detection module 323 may thus be exploited as an indicator of an increased probability of impulse noise in the audio signal. In cases in which future impulse noise occurrences are expected, the benefit of impulse noise reduction in the audio signal may overweigh disadvantages associated with an adjusted delay for the audio signal processing at signal path A, e.g., an increase of the processing latency.

[0073] This may imply a trade-off in which at least one initially detected impulse noise is accepted as a loss or a sacrifice for the reducing of impulse noise which is subsequently detected in the audio signal. The benefits of such a trade-off can include, e.g., an improved performance of the audio processing, for instance with regard to a smaller delay and / or a corresponding lower latency, at least in situations in which no impulse noise is contained and / or suspected to be contained in the audio signal. In some examples, such a trade-off may be perceived as a fair compromise in view of a rare occurrence of impulse noise for the average user in daily situations and / or a more tolerable adverse effect of the impulse noise when, after prevailing at the one or more preceding times, it is reduced at one or more subsequent times. In some examples, a prevailing of the impulse noise at the initial time may also import a benefit of directing the user's attention to a potential disturbance in his environment.

[0074] In some examples, the delay of the audio signal may be adjusted upon detecting the impulse noise at a single preceding time, e.g., at an initial time. The initial detection of impulse noise may be considered as sufficient to indicate an increased likelihood of future occurrences. In some examples, the delay of the audio signal may be adjusted upon detecting the impulse noise at two or more preceding times, e.g., at the initial time and at least one more time. A likelihood of future occurrences may then be regarded as even higher.

[0075] In some examples, a reduction of impulse noise may be more effective for impulse noise which is detected at the subsequent time as compared to the impulse noise detected at one or more of the preceding times. This may be achieved by means of the delay providing for the improved, e.g., optimized, timing for the subsequently detected impulse noise rather than the previously detected impulse noise. Conversely, in some examples, a reduction of impulse noise which is detected at one or more of the preceding times may be less effective, e.g., ineffective. To illustrate, when adjusting the delay upon detecting the impulse noise at the preceding time, the delay may be inadequate, e.g., behind schedule, for applying the gain function on the same impulse noise with an improved timing for the signal level reduction.

[0076] In some examples, when impulse noise is detected at one or more of the preceding times, e.g., when impulse noise is detected at the initial time, applying the gain function on the audio signal may be omitted, e.g., by default or in cases in which a reduced effectiveness or ineffectiveness of the delay can be anticipated.

[0077] In some examples, delay adjustment module 325 is configured to adjust the delay by applying a delay scheme on the audio signal. The delay scheme may define a duration of the delay; and / or an optimum value of the delay for reducing the signal level of the impulse noise; and / or a duration in which the delay is maintained at the optimum value; and / or a value of the delay at which the delay is kept constant; and / or a duration in which the delay is kept constant; and / or a duration and / or rate in which the delay is increased over time, e.g., before the delay reaches the optimum value and / or the delay is kept constant; and / or a duration and / or rate in which the delay is decreased over time, e.g., after the delay reaches the optimum value and / or the delay is kept constant.

[0078] In some examples, the delay scheme is updated, e.g., applied again, upon detecting the impulse noise at the subsequent time and / or the delay scheme is altered depending on a time elapsed since the impulse noise has been detected at one or more of the preceding times; and / or an intermediate time interval between impulse noise detected at two or more of the preceding times; and / or a number of the preceding times at which impulse noise has been detected; and / or a current value of the delay; and / or a latency caused by the audio signal processing; and / or one or more characteristics of impulse noise detected at one or more of the preceding times.

[0079] FIG. 5 is a schematic block diagram of a signal processing algorithm 331 for impulse noise reduction illustrating an exemplary implementation of algorithm 301 depicted in Fig. 4. Algorithm 331 may comprise an audio processing module 341 configured to process the audio signal in a time domain. In some examples, the audio signal received from audio input unit 302 at signal path A may be processed in its original form without conversion into another domain, e.g., without transformation into a frequency domain and inverse transformation into the time domain. This may result in an audio signal processing of low latency at first signal path A.

[0080] Audio processing module 341 may comprise a delay application module 343, a pre-processing module 345, a gain application module 347, and a post-processing module 349. Gain application module 347 is configured to apply a gain function on the audio signal. E.g., the gain function applied on the audio signal may include a positive or a negative gain and / or a gain model adapted to compensate for a hearing loss and / or a frequency and / or level dependent gain function and / or any other amplification of the audio signal.

[0081] Determining the gain reduction, as performed by gain modification module 327, may be performed depending on the gain function applied by gain application module 347. In some examples, when the gain reduction is determined so as to provide for an intended target gain for reducing the impulse noise, the gain reduction may be determined depending on a deviation of the gain function from the intended target gain. In some examples, the gain function, e.g., as currently applied by gain application module 347, is altered so as to reduce the gain function by the gain reduction.

[0082] Pre-processing module 345 is configured to provide for any type of audio processing prior to application of the gain function by gain application module 347. E.g., the pre-processing may comprise a processing of the raw audio signal, as received from audio input unit 302, to prepare it for the gain application and / or any other processing routines such as noise reduction, normalization, filtering, feature extraction, feedback cancelling, beamforming, and / or the like. Post-processing module 349 is configured to provide for any type of audio processing after application of the gain function by gain application module 347, e.g., processing routines for signal smoothing, artifact reduction, signal enhancement, compression, and / or the like.

[0083] Delay application module 343 is configured to perform the delaying of the audio signal according to the adjusting of the delay performed by delay adjustment module 325. The applying of the delay may be performed before the applying of the gain function by gain application module 347, e.g., before and / or after the signal pre-processing by module 345. Accordingly, by adjusting the delay, the latency at first signal path A can be adjusted so as to provide for sufficient time for the detecting of impulse noise and providing for the gain reduction at second signal path B for applying the gain function including the gain reduction at first signal path A to be effective so as to reduce the impulse noise in the audio signal, at least when the impulse noise is detected at the subsequent time.

[0084] FIG. 6 is a schematic block diagram of a signal processing algorithm 351 illustrating another exemplary implementation of algorithm 301. Algorithm 351 may comprise an audio processing module 361 configured to process the audio signal in a frequency domain. Audio processing module 361 may comprise delay application module 343, a frequency domain conversion module 365, a gain application module 367, and a time domain conversion module 369. Frequency domain conversion module 365 can transform the audio signal from the time domain into a frequency domain. Gain application module 367 is configured to apply a frequency dependent gain function on the audio signal. Time domain conversion module 369 can transform the audio signal from the frequency domain back into the time domain. Audio processing module 361 may comprise, e.g., after frequency domain conversion module 365 and before time domain conversion module 369 additional audio signal processing routines as may serve a particular purpose. Converting the audio signal between the time and frequency domain may be rather time consuming. Accordingly, a latency of frequency domain processing algorithm 361 may be increased as compared to the latency of time domain processing algorithm 341.

[0085] Delay application module 343 may be configured to perform the delaying of the audio signal before the applying of the gain function by gain application module 367 at least upon detecting the impulse noise at the subsequent time, e.g., before and / or after the frequency domain conversion by module 365. In some examples, when a frequency domain processing is performed by module 361 at first signal path A and the latency is smaller than the latency caused by impulse noise reduction module 321 at second signal path B, delay adjustment module 325 may increase the delay accordingly so as to improve the timing for the gain reduction by module 327, as described above.

[0086] In some examples, a time domain processing may be performed by module 341 at a preceding time and a frequency domain processing may be performed by module 361 at a subsequent time. E.g., a switching between time and frequency domain processing may be performed according to preferences of the user and / or a current acoustic scene and / or a current signal to noise ratio (SNR) of the audio signal. During the time domain processing, delay adjustment module 325 may increase the delay to improve the timing for applying the gain reduction. During the subsequent frequency domain processing, delay adjustment module 325 may decrease the delay to account for a latency of the frequency domain processing when optimizing the timing for applying the gain reduction. E.g., the delay may thus be optimized in each processing scheme to provide for the reduction of the impulse noise.

[0087] FIG. 7 is a schematic block diagram of a signal processing algorithm 371 illustrating another exemplary implementation of algorithm 301. Algorithm 371 may comprise audio processing module 311 and an impulse noise reduction module 381. Impulse noise reduction module 381 may comprise an impulse noise detection module 383 including a first impulse noise detection sub-module 385 and a second impulse noise detection sub-module 387. Sub-modules 385, 387 may be configured to detect impulse noise in the audio signal. For instance, each sub-module 385, 387 may be configured to detect impulse noise independently and / or sub-modules 385, 387 may be configured to detect impulse noise in a cooperative manner. E.g., one of sub-modules 385, 387 may detect a presence of impulse noise with a certain probability, wherein this information is used by the other sub-module 385, 387 to confirm the presence of impulse noise with a higher probability and / or to determine more characteristics of the impulse noise such as an amplitude and / or duration and / or frequency.

[0088] Delay adjustment module 325 may be configured to adjust the delay of the audio signal upon detection of impulse noise by first impulse noise detection sub-module 385. Sub-module 385 may thus also be denoted as an impulse noise detection module for the delay adjustment. Gain modification module 327 may be configured to provide for the gain reduction upon detection of impulse noise by second impulse noise detection sub-module 387. Sub-module 387 may thus also be denoted as an impulse noise detection module for the gain reduction.

[0089] Accordingly, detecting impulse noise for the gain reduction may be performed independently and / or complementary from detecting impulse noise for the delay adjustment. E.g., detecting impulse noise for the delay adjustment by first sub-module 385 may be performed in a more inaccurate and / or time-efficient manner as compared to detecting impulse noise for the gain reduction, which may require more information, e.g., with regard to characteristics of the impulse noise such as an amplitude and / or duration and / or frequency, and / or may require a higher certainty of the impulse noise detection. In this way, a time consumption for the delay adjustment may be minimized.

[0090] FIG. 8 is a schematic block diagram of a signal processing algorithm 401 for impulse noise reduction illustrating another exemplary implementation of algorithm 301. Algorithm 401 may comprise audio processing module 311 and an impulse noise reduction module 421 including an impulse noise management module 423. As illustrated, impulse noise management module 423 may be operationally connected with impulse noise detection module 323 and / or gain modification module 327 and / or delay adjustment module 325 and / or audio processing module 311.

[0091] In some examples, impulse noise management module 423 may be configured to receive and / or exchange information with impulse noise detection module 323 and / or gain modification module 327 and / or delay adjustment module 325 and / or audio processing module 311. To illustrate, delay management module 423 may receive information about impulse noise currently and / or previously detected by impulse noise detection module 323, e.g., an occurrence of impulse noise and / or one or more characteristics of the impulse noise such as a duration, amplitude and / or frequency. Delay management module 423 may also receive information about a gain reduction applied by gain modification module 327 and / or a delay applied by delay adjustment module 325, e.g., a currently applied delay scheme and / or a current value of the delay and / or a duration and / or rate in which the delay is currently kept constant and / or increased and / or decreased over time. Delay management module 423 may also receive information about the audio processing performed by audio processing module 311, e.g., a current latency caused by the audio processing. In some examples, delay management module 423 may log the information over time and / or store the information in a memory. E.g., a number of preceding times at which impulse noise has been detected may be determined based on the logged information.

[0092] In some examples, impulse noise management module 423 may be configured to control the detecting of impulse noise and / or the determining and / or applying of the gain reduction and / or the adjusting of the delay. To illustrate, gain modification module 327 may be controlled to modify and / or omit the applying of the gain reduction depending on information received from impulse noise detection module 323 and / or from delay adjustment module 325. In some instances, the applying of the gain reduction may be omitted in cases in which a reduced effectiveness or ineffectiveness of the delay can be anticipated. Those cases may include, e.g., instances in which impulse noise has been detected at an initial time and / or when the currently applied delay would not allow for a sufficient reduction of the impulse noise when applying the gain reduction with the current delay.

[0093] In some examples, the gain reduction may be modified depending on a current value of the delay and / or a current value of the latency of the audio processing, as performed by audio processing module 311. E.g., when the delay and / or latency is smaller, a rate and / or duration in which the gain reduction is increased over time, which may also be referred to as an attack time of the gain reduction, may be increased as compared to when the delay and / or latency is smaller.

[0094] In some examples, impulse noise detection module 323 may be controlled to adjust a a time consumption caused by the detecting of the impulse noise. The time consumption caused by the detecting of the impulse noise may also be denoted as an impulse noise detection latency. E.g., the impulse noise detection latency may be adjusted depending on one or more of a time elapsed since the impulse noise has been detected at one or more of the preceding times; an intermediate time interval between impulse noise detected at two or more of the preceding times; a number of the preceding times at which impulse noise has been detected; and / or one or more characteristics of impulse noise detected at one or more of the preceding times.

[0095] To illustrate, impulse noise detection module 323 may be configured to detect impulse noise with a larger time consumption, in particular a larger latency, which may result in an increased accuracy of the impulse noise detection, and / or with a smaller time consumption, in particular a smaller latency, which may result in a decreased accuracy of the impulse noise detection. Impulse noise detection module 323 may then be controlled to switch between detecting impulse noise with the larger time consumption and detecting impulse noise with the smaller time consumption, e.g., depending on previous information received from impulse noise detection module 323. For instance, when impulse noise has been detected at one or more of the preceding times, an increased likelihood of subsequent occurrences of impulse noise in the audio signal may be assumed. As a result, a decreased accuracy of impulse noise detection may be sufficient to determine the subsequent occurrences of impulse noise, which may be performed with the smaller time consumption. Accordingly, when impulse noise has been detected at one or more of the preceding times, detecting impulse noise with a smaller time consumption may be performed. In some examples, upon adjusting the time consumption caused by the detecting of the impulse noise, the delay of the audio signal, as performed by audio processing module 311 at first signal path A, is adjusted to account for the adjusted time consumption.

[0096] In some examples, impulse noise detection module 323 may be controlled to adjust a a time consumption caused by the determining and / or applying of the gain reduction. The time consumption caused by the determining and / or applying of the gain reduction may also be denoted as a gain reduction latency. E.g., the gain reduction latency may be adjusted depending on one or more of a time elapsed since the impulse noise has been detected at one or more of the preceding times; an intermediate time interval between impulse noise detected at two or more of the preceding times; a number of the preceding times at which impulse noise has been detected; one or more characteristics of impulse noise detected at one or more of the preceding times.

[0097] To illustrate, gain modification module 327 may be configured to determine the gain reduction with a larger time consumption, in particular a larger latency, which may result in an increased accuracy of the gain reduction and / or target gain, and / or with a smaller time consumption, in particular a smaller latency, which may result in a decreased accuracy of the gain reduction and / or target gain. For instance, determining the gain reduction with a smaller time consumption may imply to retrieve and / or reuse a gain reduction previously determined with a larger time consumption. Gain modification module 327 may then be controlled to switch between determining the gain reduction with the larger time consumption and determining the gain reduction with the smaller time consumption, e.g., depending on information received from impulse noise detection module 323. In particular, when an increased likelihood of subsequent occurrences of impulse noise in the audio signal is assumed due to a detection of impulse noise at one or more of the preceding times, it may also be assumed that the subsequently occurring impulse noise has comparable and / or similar characteristics as the impulse noise detected at one or more of the preceding times. Thus, when impulse noise has been detected at one or more of the preceding times and / or when impulse noise with comparable and / or similar characteristics has been detected at two or more of the preceding times, determining the gain reduction with a with a smaller time consumption may be performed. In some examples, upon adjusting the time consumption caused by the determining and / or applying of the gain reduction, the delay of the audio signal during the audio signal processing, as performed by audio processing module 311 at first signal path A, is adjusted to account for the adjusted time consumption.

[0098] FIGS. 9A, 9B, 9C illustrate functional plots 510, 530, 550 of exemplary audio signals 511, 531, 551 in a time domain. Time is indicated on an axis of abscissas. A signal level is indicated on an axis of ordinates. Audio signal 511 is an input audio signal, which may be obtained by audio input unit 213, 302 for audio signal processing. The audio processing may be performed, e.g., by audio processing module 311, 341, 361 at first signal path A. Audio signal 531 is a delayed audio signal, which may be obtained by adjusting a delay of input audio signal 511. Adjusting the delay of audio signal 511 may be performed, e.g., by impulse noise reduction module 321, 381, 421 at second signal path B. Audio signal 551 is an impulse noise reduced audio signal, which may be obtained by applying a gain reduction on delayed audio signal 531 at first signal path A. In some examples, when another gain function is applied on audio signal 531 during the audio signal processing at signal path A, applying the gain reduction may result in a target gain applied on delayed audio signal 531. The gain reduction and / or target gain may be determined by impulse noise reduction module 321, 381, 421 at second signal path B.

[0099] Input audio signal 511, as illustrated in FIG. 9A, comprises a number of transient events 521, 522, 523, 524 of short duration and high intensity. In particular, amplitudes A1, A2, A3, A4 of the signal level at transient events 521 - 524 may be significantly higher than amplitudes at a remaining signal portion 515 in which transient events 521 - 524 are absent. Transient events 521 - 524 are thus characteristic of impulse noise, which may be detected by impulse noise detection module 323, 383. As illustrated, a time interval 11, I2, I3 separated two subsequent events of impulse noise 521 - 524 generally varies, which may be attributed to a random occurrence of impulse noise.

[0100] As also illustrated, impulse noise 521 - 524 can appear in a cumulative manner in audio signal 510 over time. In the exemplary audio signal 511, four subsequent events 521, 522, 523, 524 of impulse noise are depicted for illustrative purposes, which may occur within a temporal distance of less than several seconds or several hundred milliseconds. Thus, upon detecting impulse noise 521 - 524 at one or more preceding times, an increased likelihood for detecting impulse noise 521 - 524 at a subsequent time, e.g., within a certain time period including a duration of time intervals 11 - 13, may be expected. To illustrate, such a time period may be selected depending on an acoustic environment and / or other characteristics of impulse noise 521 - 524, e.g., less than ten seconds, in some examples less than 1 second, or less than 0.5 seconds, or less than 0.2 seconds. As further illustrated, amplitudes A1 - A4 of subsequent impulse noise 521 - 524 may differ significantly despite their cumulative occurrence. In other examples, amplitudes A1 - A4 of subsequent impulse noise 521 - 524 may vary less, e.g., less than by 20 percent, which may be attributed to an equal sound source emitting the impulse noise.

[0101] Delayed audio signal 531, as illustrated in FIG. 9B, comprises events of impulse noise 541, 542, 543, 544 which are temporally separated by time intervals J1, J2, J3. Time intervals J1- J3 differ from time intervals I1 - I3 between subsequent events of impulse noise 521 - 524 in input audio signal 511. For instance, as illustrated, when the delay of audio signal 511 is adjusted to an increased value, time intervals J1 - J3 of delayed audio signal 531 may be larger than time intervals I1 - I3 of input audio signal 511. The difference between time interval J1 - J3 and time interval 11 - I3 may be a time interval D1, D2, D3 representative of a delay of audio signal 531 at a time at which impulse noise 541 - 544 occurs in audio signal 531.

[0102] In some examples, as illustrated, an initial event 521, 541 of impulse noise may occur in input audio signal 511 and in delayed audio signal 531 at an equal time. Initially occurring impulse noise 541 in audio signal 531 may thus be un-delayed relative to initially occurring impulse noise 521 in audio signal 511. This may be the case, e.g., when input audio signal 511 is only delayed upon detecting impulse noise at the initial event 521, e.g., during or after initial event 521, and / or when a latency caused by an audio processing of input audio signal 511 at first signal path A is negligible between the time of initial event 521, 541 in audio signals 511, 531. In other examples, e.g., when audio signal 511 would already be delayed before detecting impulse noise at initial event 521 and / or when a latency would be caused by the processing of audio signal 511 between the times at which initial impulse noise 521, 541 occurs in audio signals 511, 531, the initial impulse noise 521, 541 may be temporally shifted relative to one another in audio signals 511, 531. Such a temporal shift may be disregarded in the following description for illustrative purposes.

[0103] In the illustrated example, initial event 521, 541 of impulse noise precedes a first subsequent event 522, 542 of impulse noise in audio signals 511, 531. First subsequent events 522, 542 are temporally shifted by first delay D1. First subsequent event 522, 542 precedes a second subsequent event 523, 543 in audio signals 511, 531. Second subsequent events 523, 543 are temporally shifted by second delay D2. Second subsequent event 523, 543 precedes a third subsequent event 524, 544 in audio signals 511, 531. Third subsequent events 524, 544 are temporally shifted by third delay D3.

[0104] In some examples, as illustrated, one or more preceding delays D1 - D3 may vary from one or more subsequent delays D1 - D3. E.g., one or more preceding delays, as exemplified by first delay D1, may be smaller as compared to one or more subsequent delays, as exemplified by second and third delays D2, D3. The delay variation may be attributed, in some cases, to an adjustment of the delay to a desired, e.g., optimized, value in which the delay is varied, e.g., increased or decreased, until the desired value is reached, as further described below. In some examples, as also illustrated, two or more delays, as exemplified by second and third delays D2, D3, may be kept at an equal value. In some cases, this may be attributed to an adjustment of the delay in which the delay is kept constant after reaching a desired value, as further described below.

[0105] Impulse noise reduced audio signal 551, as illustrated in FIG. 9C, comprises events of impulse noise 561, 562, 563, 564, wherein, after applying delays D1 - D3 on input audio signal 511 so as to obtain delayed audio signal 531, a gain reduction for reducing a signal level of the impulse noise is applied. As a result, amplitudes A2, A3, A4 of events 522 - 524 subsequent to initial event 521 in input audio signal 511 are reduced to smaller amplitudes B2, B3, B4 of events 562 - 564 subsequent to initial event 561 in impulse noise reduced audio signal 551. In some examples, as illustrated, amplitude B1 of initial event 561 in impulse noise reduced audio signal 551 is kept equal to amplitude A1 of initial event 521 in input audio signal 511. In some cases, this may be attributed to an applying of delays D1 - D3 on audio signal 511 which is coming too late for delaying initial event 561, e.g., upon detection of initial event 561, in order to effectively apply the gain reduction for reducing amplitude A1 of initial event 521. In some cases, this may be attributed to an omission of applying the gain reduction for reducing amplitude A1 of initial event 521, e.g., in cases when a reduced effectiveness of applying the gain reduction can be anticipated. Examples of those cases are further described below.

[0106] FIG. 10B illustrates a functional plot 610 of a delay scheme 611 which may be applied on input audio signal 511 to adjust a delay of the audio signal. Time of input audio signal 511 is indicated on an axis of abscissas. A delay in units of time is indicated on an axis of ordinates. FIGS. 10A and 10C display again functional plots 510, 530 of exemplary audio signals 511, 531 for illustrative purposes. Delay scheme 611 may be applied on audio signal 511 at first signal path A, e.g., before, during, and / or after one or more other signal processing routines performed during the audio signal processing.

[0107] In some examples, when applying delay scheme 611, the delay may be adjusted from a base line value 615. In some examples, base line value 615 may correspond to a value of the delay before the adjusting of the delay, e.g., a value of the delay before applying delay scheme 611. In some examples, base line value 615 may correspond to a latency caused by an audio signal processing at first signal path A and / or a previously adjusted value of the delay and / or a time it takes to receive audio signal 511 from audio input unit 213, 302 and / or a time it takes to effectively provide for adjusting the delay. In some examples, base line value 615 may be substantially zero and / or negligible and / or disregarded, e.g., in cases in which delay scheme 611 would be effectively applied on input audio signal 511 immediately after receipt.

[0108] In some examples, delay scheme 611 comprises a first duration 621 during which the delay is increased, a second duration 623 during which the delay is kept constant, and a third duration 625 during which the delay is decreased. In other examples, a delay scheme may comprise a first duration during which the delay is decreased, a second duration during which the delay is kept constant, and a third duration during which the delay is increased. In other examples, a delay scheme may comprise a first duration during which the delay is increased, and a second duration during which the delay is decreased. In same examples a value to which the delay is increased and / or from which the delay is decreased and / or at which the delay is kept constant may correspond to an optimum value OD of the delay. Optimum value OD may be defined as a value minimizing a difference between a time at which impulse noise occurs in the audio signal, e.g., a time at which impulse noise is contained in the audio signal, and a time at which the gain reduction is applied on the audio signal.

[0109] In some examples, first duration 621 may be defined as a duration during which the delay is increased from base line value 615 to optimum value OD. First duration 621 may also be referred to as an attack time of the delay adjustment. A rate at which the delay is increased at first duration 621 may also be referred to as an attack slope of the delay adjustment. In some examples, the attack slope is constant. In other examples, the attack slope may vary over time. In some examples, third duration 625 may be defined as a duration during which the delay is decreased from optimum value OD to base line value 615. Third duration 625 may also be referred to as a release time of the delay adjustment. A rate at which the delay is decreased at third duration 625 may also be referred to as a release slope of the delay adjustment. In some examples, the release slope is constant. In other examples, the release slope may vary over time. In some examples, a constant value at which the delay is kept, e.g., at second duration 623 of delay scheme 611, may correspond to optimum value OD for applying the gain reduction on audio signal 511 to reduce impulse noise. The constant value may thus correspond to a value of the delay minimizing the difference of a time in which one or more of events 521 - 524 occur in audio signal 511 and a time at which an application of the gain reduction is performed on audio signal 511.

[0110] In some examples, delay scheme 611 comprises a delay maintenance period M1, M2. Delay maintenance period M1, M2 may define a minimum duration during which the delay adjustment is maintained. In some examples, delay maintenance period M1, M2 may define a minimum duration, e.g., a minimum time interval, at which delay scheme 611 is applied. In some examples, as illustrated, delay maintenance period M1, M2 may define a minimum duration during which the delay is kept constant, e.g., a minimum value of second duration 623. In other examples, delay maintenance period M1, M2 may define a minimum duration during which the delay is kept different from base line value 615, e.g., a minimum value of a sum of durations 621, 623, 625. In some examples, delay maintenance period M1, M2 may vary over time, e.g., when delay scheme 610 is updated upon detecting impulse noise 521 - 524 at a subsequent time and / or depending on a characteristic and / or number of preceding events 521 - 524 of impulse noise. E.g., as illustrated, a shorter delay maintenance period M1 may be extended to a longer delay maintenance period M2 when the preceding events of impulse noise exceed a threshold number, as exemplified by three preceding events 521 - 523. In other examples, delay maintenance period may be extended each time another preceding event 521 - 524 of impulse noise is detected. In this way, an increased likelihood of subsequent events of impulse noise depending on the number of preceding events may be accounted for.

[0111] In some examples, delay maintenance period M1, M2 may be selected based on a predetermined time value, e.g., an empirical value and / or a machine-learned value, in which a cumulative occurrence of impulse noise is usually observed, which may also depend, e.g., on a characteristic of the impulse noise and / or a current acoustic environment and / or individually learned acoustic patterns when a user is wearing the hearing device. In some examples, one or more properties of delay scheme 611, e.g., one or more of durations 621, 623, 625, an attack and / or release slope of the delay, a constant value of the delay and / or delay maintenance period M1, M2 may be determined by delay adjustment module 325 and / or impulse noise management module 423, e.g., depending on a current latency of the audio signal processing at first signal path A and / or a time consumption of detecting the impulse noise and / or determining the gain reduction at signal path B and / or a characteristic of the impulse noise and / or a current acoustic environment and / or individually learned acoustic patterns when a user is wearing the hearing device.

[0112] In the illustrative example shown in FIGS. 10A - 10C, delay scheme 611 is applied on audio signal 511 upon detecting one or more events of impulse noise, as exemplified by initial event 521. In practical cases, applying delay scheme 611 on audio signal 511 may lag behind a time at which event 521 occurs in audio signal 511, e.g., due to a time consumption required for detecting initial event 521 by impulse noise detection module 323, 383. For illustrative purposes, such a time lag is disregarded in the present example. Applying delay scheme 611 at first duration 621, which may define an attack time and / or attack slope of the delay adjustment, leads to an increase of the delay until reaching the constant value of the delay. Thereafter, at second duration 623 of delay scheme 611, the constant value of the delay is kept at least for the length of delay maintenance period M1.

[0113] During delay maintenance period M1, however, first subsequent event 522 of impulse noise is detected. As a result, delay scheme 611 is updated by restarting the delay maintenance period M1 for which the constant value of delay is minimally kept. This effectively leads to a prolongation of the time during which the delay is kept constant when subsequent impulse noise is detected during delay maintenance period M1. At third duration 625 of delay scheme 611, when no further subsequent event of impulse noise has been detected during delay maintenance period M1 before, the delay is decreased again. To this end, delay scheme 611 may define a release time and / or a release slope of the delay adjustment. The decrease of the delay may be continued until base line value 615 is reached again, unless a further subsequent event of impulse noise would be detected in the meantime.

[0114] In the shown example, during third duration 625 of delay scheme 611, second subsequent event 523 of impulse noise is detected. As a result, delay scheme 611 is updated again by restarting at first duration 621 of increasing the delay until reaching the constant value. Before the increase, the delay has not been minimized to base line value 615. The constant value can therefore be reached in a shorter time as compared to the initial application of delay scheme 611 upon detecting initial event 521 starting from base line value 615. Accordingly, first duration 621 is reduced as compared to the initial application. When the constant value is reached again, the delay is kept constant at second duration 623 of delay scheme 611. In this case, upon detecting a threshold number of impulse noise, as exemplified by detecting a third impulse noise at second subsequent event 523, the updated delay scheme comprises an extended delay maintenance period M2. Delay maintenance period M2, defining a minimum time at which the delay is kept constant, is larger than previously applied delay maintenance period M1.

[0115] During delay maintenance period M2, third subsequent event 524 of impulse noise is detected. As a result, delay scheme 611 is updated by a repeated application of delay maintenance period M2 so that the minimum time at which the delay is kept constant restarts upon detection of third subsequent event 524. During the repeated application of delay maintenance period M2, no further event of impulse noise is detected. Accordingly, second duration 623 of delay scheme 611 ends after the repeatedly applied delay maintenance period M2. Thereafter, at third duration 625 of delay scheme 611, the delay is decreased again.

[0116] As illustrated in FIG. 10C, applying delay scheme 611 in the above described way may lead to substantially equal delays of impulse noise, as exemplified by delays D2, D3 at subsequent events 543, 544, which may be improved, e.g., optimized, with regard to the application of the gain reduction so as to reduce the impulse noise. In some cases, the delays can vary, e.g., depending on a varying latency of the audio processing performed at first signal path A and / or depending on a varying time consumption of the impulse noise reduction at second signal path B, e.g., of the impulse noise detection and / or gain reduction determination. In some cases, the delays can also vary due to a time lag of reaching the constant delay value at second duration 623, as exemplified by delay D1 which is slightly smaller than delays D2, D3. This may be caused, e.g., by the finite attack slope during increasing the delay at first duration 621 before reaching the optimum value. Accordingly, applying the gain reduction may not be fully optimized in a case in which the delay has not yet been adjusted to the optimized value.

[0117] As illustrated in FIG. 10B, when adjusting the delay by applying delay scheme 611, the attack slope may be larger than the release slope. In particular, a rate at which the delay is increased during first duration 621 may be larger than a rate at which the delay is decreased during third duration 625. A larger attack slope can be beneficial for a quicker adjustment of the delay to the value optimized for the gain reduction. In this way, subsequent events of impulse noise may be accounted for even at a short time interval from the preceding event at which the impulse noise has been detected. A smaller release slope can be beneficial for a quicker re-adjustment of the delay to the value optimized for the gain reduction. To illustrate, the smaller release slope may provide for a longer time until the delay returns to base line value 615 during which the delay can be quicker re-adjusted to the optimized value. Thus, within this prolonged time, subsequent events of impulse noise may be effectively reduced when arriving even quicker after the preceding event.

[0118] FIG. 11B illustrates a functional plot 650 of a sequence 651 of gain reductions 661, 671, 681, 691 which may be applied on delayed audio signal 531 to reduce impulse noise. Time of delayed audio signal 531 is indicated on an axis of abscissas. A value of the gain reduction is indicated on an axis of ordinates. FIGS. 11A and 11C display again functional plots 530, 550 of exemplary audio signals 531, 551 for illustrative purposes. Gain reductions 661, 671, 681, 691 may be applied on audio signal 511 at first signal path A, e.g., before, during, and / or after one or more other signal processing routines performed during the audio signal processing.

[0119] After applying one or more gain reductions 661, 671, 681, 691 on delayed audio signal 531, impulse noise can be effectively reduced, as exemplified by events 563, 564 in impulse noise reduced audio signal 551 when compared to events 543, 544 in delayed audio signal 531. In the illustrated examples, amplitudes A3, A4 of events 543, 544 in audio signal 531 have been reduced to amplitudes B3, B4 of events 563, 564 in audio signal 551. This can be attributed to delays D2, D3 delaying impulse noise events 543, 544 to an improved, e.g., optimized, timing at which gain reductions 681, 691 are applied. One or more gain reductions 661 - 691 may then be temporally aligned with events 541 - 544 of impulse noise in delayed audio signal 531, as exemplified by gain reductions 681, 691 which are applied at an equal time at which events 543, 544 occur in delayed audio signal 531. In some examples, a maximum value of gain reductions 681, 691, e.g., a peak value or a constant maximum value, may coincide with the time of events 543, 544 in delayed audio signal 531.

[0120] In some examples, a reduction of impulse noise in delayed audio signal 531 may also occur at a less optimized timing, as exemplified by event 562 in impulse noise reduced audio signal 551 corresponding to event 542 in delayed audio signal 531. In the illustrated example, amplitude A2 of event 542 in audio signal 531 has been reduced to amplitude B2 of event 562 in audio signal 551. The reduction of impulse noise event 542 can be attributed to delay D1, which is shorter than delays D2, D3, but still leads to a reduction of the impulse noise due to an improvement of the timing of applying gain reduction 671 due to delay D1.

[0121] In some examples, after applying one or more gain reductions 661, 671, 681, 691 on delayed audio signal 531, impulse noise may remain unreduced and / or reduced less than a desired degree, as exemplified by event 561 in impulse noise reduced audio signal 551 when compared to event 541 in delayed audio signal 531. In the illustrated example, amplitude B1 of event 561 in audio signal 551 is equal to amplitude A1 of event 541 in audio signal 531 so that the impulse noise remains unreduced. This may be attributed to an applying of delay scheme 611 coming too late for an effective application of one or more gain reductions 661 - 691 so as to reduce impulse noise to a desired extent. In the illustrated example, delay scheme 611 is applied upon detection of initial event 521 in input audio signal 510 so that a delay of initial event 541 in delayed audio signal 531 remains unadjusted. As a result, applying gain reduction 661 on delayed audio signal 531 would also come too late to provide for an effective reduction of impulse noise at initial event 541.

[0122] In some examples, in cases in which applying one or more gain reductions 661, 671, 681, 691 on delayed audio signal 531 would leave the impulse noise unreduced and / or less reduced than the desired degree, applying one or more gain of reductions 661 - 691 may be omitted. In some examples, applying one or more gain reductions 661 - 691 may be omitted upon detecting one or more preceding events 521 - 524 in input audio signal 511 before and / or based on which the delay is adjusted, e.g., before and / or based on which delay scheme 611 is applied. In the illustrated example, applying gain reduction 661 on delayed audio signal 531 for reducing initial event 541 of impulse noise may be omitted.

[0123] In some examples, gain reductions 661, 671, 681, 691 may be applied on delayed audio signal 531 by adding gain reductions 661, 671, 681, 691 to a base line level 655 of gain also applied on the audio signal. As a result, base line level 655 may be reduced by gain reductions 661, 671, 681, 691. In some instances, base line level 655 may be zero gain, or positive gain, or negative gain. In some examples, a positive gain may be applied as a hearing enhancement feature, e.g., to compensate for an individual hearing loss. In some examples, a negative gain may be applied as a hearing protection feature, e.g., in a loud acoustic environment. In some examples, gain reductions 661, 671, 681, 691 may be applied on delayed audio signal 531 by adding gain reductions 661, 671, 681, 691 to a gain function, which may be applied on delayed audio signal 531 during audio processing at first signal path A, e.g., by audio processing module 311, 341, 361. As a result, the gain function may be reduced by gain reductions 661, 671, 681, 691. In some examples, the gain function may amplify the audio signal according to a gain model defining an amplification of the audio signal at different frequencies and / or at different signal levels. In some examples, the gain function may be customized to specific needs and / or an individual hearing loss of the user.

[0124] In some examples, gain reductions 661, 671, 681, 691 each may comprise an attack time 663, 673, 683, 693 of the gain reduction, during which the gain is reduced until reaching a maximum level of reduced gain. The maximum level may be a peak level or may be kept constant for a certain amount of time. Further, gain reductions 661, 671, 681, 691 each may comprise a release time 665, 675, 685, 695 of the gain reduction, during which the gain is increased again until reaching an initial level of unreduced gain. Attack time 663, 673, 683, 693 may also be defined by an attack slope indicative of a slope at which the gain is reduced during the gain reduction until reaching the maximum level. Release time 665, 675, 685, 695 may also be defined by a release slope indicative of a slope at which the gain is increased during the gain increase until reaching the initial level.

[0125] In some examples, when gain reductions 661, 671, 681, 691 are applied in a sequence upon detecting a plurality of events 521 - 524 of impulse noise in input audio signal 511, a newly applied gain reduction 661, 671, 681, 691 may cancel and / or overrule a previously applied gain reduction 661, 671, 681, 691. This is exemplified, in FIG. 11B, by gain reduction 671 which is applied before gain reduction 661 ends, in particular during release time 665 of gain reduction 661.

[0126] In some examples, attack time 663, 673, 683, 693 and / or release time 665, 675, 685, 695 and / or the maximum level of reduced gain may be determined depending on one or more characteristics of events 521 - 524 of impulse noise and / or depending on the adjustment of delay D1, D2, D3 of input audio signal 531 and / or depending on a latency of the audio processing at first signal path A. To illustrate, a larger attack slope during attack time 663 - 669 may be applied for shorter delays D1, D2, D3 and / or a for a smaller latency of the audio processing.

[0127] FIG. 12 illustrates a block flow diagram for an exemplary method of processing an audio signal in a hearing device. The method may be executed, e.g., by processor 104 of system 100 and / or processor 204 of hearing device 200. At operation S11, impulse noise is detected in an input audio signal at one or more preceding times. At operation S12, upon detecting the impulse noise at the one or more preceding times, a delay of the audio signal is adjusted. At operation S13, impulse noise is detected in the audio signal at a subsequent time. At operation S14, when impulse noise is detected at the subsequent time, a gain reduction is applied on the audio signal to reduce a signal level of the impulse noise. The delay at S12 can be provided such that a timing of the applying of the gain reduction can be improved when the impulse noise is detected at the subsequent time.

[0128] While the principles of the disclosure have been described above in connection with specific devices, systems and methods, it is to be clearly understood that this description is made only by way of example and not as limitation on the scope of the invention. The above described preferred embodiments are intended to illustrate the principles of the invention, but not to limit the scope of the invention. Various other embodiments and modifications to those preferred embodiments may be made by those skilled in the art without departing from the scope of the present invention that is solely defined by the claims. In the claims, the word "comprising" does not exclude other elements or steps, and the indefinite article "a" or "an" does not exclude a plurality. A single processor or controller or other unit may fulfil the functions of several items recited in the claims. The mere fact that certain measures are recited in mutually different dependent claims does not indicate that a combination of these measures cannot be used to advantage. Any reference signs in the claims should not be construed as limiting the scope.

Examples

Embodiment Construction

[0038]Methods, systems, and devices of processing an audio signal in a hearing device for reducing impulse noise in the audio signal are described herein. As will be described in more detail below, an exemplary method may comprise detecting impulse noise in the audio signal; applying a gain reduction on the audio signal to reduce a signal level of the impulse noise; and adjusting, upon detecting the impulse noise at one or more preceding times, a delay of the audio signal to improve a timing of the applying of the gain reduction when the impulse noise is detected at a subsequent time.

[0039]By providing methods, systems, and devices such as those described herein, it may be possible to provide for a compatibility of the impulse noise reduction with various audio processing algorithms. Examples may include algorithms operating with a rather low latency, wherein the low latency may be maintained at least under normal conditions and / or at the absence of impulse noise in the audio signal...

Claims

1. A method of processing an audio signal in a hearing device, the method comprising - detecting impulse noise (521 - 524, 541 - 544, 561 - 564) in the audio signal (510, 530, 550); and, - applying a gain reduction (661, 671, 681, 691) on the audio signal (510, 530, 550) to reduce a signal level of the impulse noise (521 - 524, 541 - 544, 561 - 564), characterized by - adjusting, upon detecting the impulse noise (521 - 524, 541 - 544, 561 - 564) at one or more preceding times, a delay of the audio signal (510, 530, 550) to improve a timing of the applying of the gain reduction (661, 671, 681, 691) when the impulse noise (521 - 524, 541 - 544, 561 - 564) is detected at a subsequent time.

2. The method of claim 1, wherein the delay is adjusted by applying a delay scheme (610) on the audio signal (510, 530, 550), the delay scheme (610) defining a minimum duration during which the delay adjustment is maintained.

3. The method of claim 2, wherein, when the delay scheme is applied, the delay scheme (610) is updated upon detecting the impulse noise (521 - 524, 541 - 544, 561 - 564) at the subsequent time, wherein the updating of the delay scheme (610) comprises extending said minimum duration.

4. The method of claim 2 or 3, wherein the delay scheme (610) defines an optimum value of the delay, the optimum value minimizing a difference between a time at which the impulse noise (521 - 524, 541 - 544, 561 - 564) occurs in the audio signal and a time at which the gain reduction (661, 671, 681, 691) is applied on the audio signal.

5. The method of claim 4, wherein the delay scheme (610) defines - a duration and / or rate in which the delay is decreased over time from the optimum value to return to a value of the delay before the adjusting of the delay; and / or - a duration and / or rate in which the delay is increased over time from a value of the delay before the adjusting of the delay to the optimum value; and / or - a duration during which the delay is kept constant at the optimum value.

6. The method of any of claims 2 to 5, further comprising determining the delay scheme (610) depending on - a latency caused by the audio signal processing; and / or - a current value of the delay; and / or - a number of the preceding times at which impulse noise (521 - 524, 541 - 544, 561 - 564) has been detected; and / or - an intermediate time interval between impulse noise (521 - 524, 541 - 544, 561 - 564) detected at two or more of the preceding times; and / or - a time elapsed since the impulse noise (521 - 524, 541 - 544, 561 - 564) has been detected at one or more of the preceding times.

7. The method of any of the preceding claims, wherein the audio signal (510, 530, 550) is processed in a time domain.

8. The method of any of the preceding claims, wherein, when detecting the impulse noise (521 - 524, 541 - 544, 561 - 564) at one or more preceding times, the applying of the gain reduction (661, 671, 681, 691) on the audio signal (510, 530, 550) is omitted.

9. The method of any of the preceding claims, wherein the processing of the audio signal (510, 530, 550) comprises applying a gain function on the audio signal (510, 530, 550), wherein the gain reduction (661, 671, 681, 691) is applied in addition to the gain function.

10. The method of any of the preceding claims, further comprising determining the gain reduction (661, 671, 681, 691), wherein the gain reduction (661, 671, 681, 691) is determined so as to meet a target gain representative of an intended signal level of the impulse noise (521 - 524, 541 - 544, 561 - 564) in the audio signal (510, 530, 550).

11. The method of claim 10, wherein the determining the gain reduction (661, 671, 681, 691) comprises determining - a maximum value of the gain reduction (661, 671, 681, 691); and / or - a duration in which the gain reduction (661, 671, 681, 691) is maintained at the maximum value; and / or - a rate and / or duration in which the gain reduction (661, 671, 681, 691) is increased over time; and / or - a rate and / or duration in which the gain reduction (661, 671, 681, 691) is decreased over time.

12. The method of claim 11, wherein the gain reduction (661, 671, 681, 691) is determined depending on - one or more characteristics of the impulse noise (521 - 524, 541 - 544, 561 - 564); and / or - a latency caused by the audio signal processing; and / or - a current value of the delay.

13. The method of any of the preceding claims, wherein a time consumption caused by the detecting of the impulse noise (521 - 524, 541 - 544, 561 - 564) is adjusted depending on - a time elapsed since the impulse noise (521 - 524, 541 - 544, 561 - 564) has been detected at one or more of the preceding times; and / or - an intermediate time interval between impulse noise (521 - 524, 541 - 544, 561 - 564) detected at two or more of the preceding times; and / or - a number of the preceding times at which impulse noise (521 - 524, 541 - 544, 561 - 564) has been detected; and / or - one or more characteristics of impulse noise (521 - 524, 541 - 544, 561 - 564) detected at one or more of the preceding times.

14. The method of any of the preceding claims, wherein the processing of the audio signal (510, 530, 550) is performed at a first signal path (A) at which the delay of the audio signal (510, 530, 550) is adjusted and the gain reduction (661, 671, 681, 691) is applied on the audio signal (510, 530, 550), and the impulse noise (521 - 524, 541 - 544, 561 - 564) in the audio signal (510, 530, 550) is detected at a second signal path (B) bypassing the first signal path (A).

15. A hearing system comprising a hearing device configured to be worn at an ear of a user, the hearing system comprising - an audio input unit (213, 302) for obtaining an input audio signal (510, 530, 550); - a processor (104, 204) for audio signal processing of the input audio signal (510, 530, 550) to obtain an output audio signal (510, 530, 550); and - an audio output unit (217, 303) for outputting the output audio signal (510, 530, 550) so as to stimulate the user's hearing, wherein the audio output unit (217, 303) is included in the hearing device, characterized in that the processor (104, 204) is configured to perform the method of any of claims 1 to 14 based on the input audio signal (510, 530, 550).

Images