Audio system and signal processing method for ear-worn playback devices

Abstract

An audio system (AS) for a hearable device (HP) includes a speaker (SP) and an error microphone (FB_MIC) configured to sense sound being output from the speaker (SP) and ambient sound. The audio system (AS) further includes a detection engine (DET) configured to determine a driver response between the speaker (SP) and the error microphone (FB_MIC) and estimate a leakage condition based on the determined driver response.

Description

[0001] The present disclosure relates to an audio system and a signal processing method, each for an ear-worn playback device including a speaker, such as headphones.

[0002] Today, many headphones, including earbuds, incorporate technologies that enhance the user's sound experience, such as noise cancellation. For example, this noise cancellation technology is called active noise control or ambient noise cancellation, both abbreviated as ANC. ANC typically utilizes recorded ambient noise, which is processed to generate an anti-noise signal. This anti-noise signal is then combined with the desired audio signal to be played through the headphone's speakers. ANC can also be used with other audio devices, such as cell phones or mobile phones.

[0003] Various ANC methods utilize feedback, FB, microphones, feedforward, FF, microphones, or a combination of feedback and feedforward microphones. Based on the given acoustics of the system, effective FF and FB ANC is achieved by tuning filters or by adjusting the audio signal, for example, via an equalizer.

[0004] Hybrid noise-canceling headphones are well known. For example, a microphone is placed inside a volume that is acoustically coupled directly to the eardrum, typically in front of the headphone driver. This is known as a feedback (FB) microphone. A second microphone, a feedforward (FF) microphone, can be placed outside the headphone, acoustically decoupling it from the headphone driver.

[0005] For each system to work effectively, the headset preferably forms a near-perfect seal against the user's ear / head that doesn't change while the device is worn and is consistent for any user. Any change in the seal due to a poor fit will alter the acoustics and ultimately the ANC performance. This seal is typically located between the headphone cushion and the user's head, or between the rubber tip of the earbud and the wall of the ear canal.

[0006] For most noise-canceling headphones and earbuds, efforts are made to maintain a consistent fit between wearers and from user to user to ensure that the acoustics of the headset do not change and that the noise filters are always well matched. However, "leaky" earbuds and headphones, where the earpad / earbud tip does not form a seal with the ear, can have very different acoustic effects from person to person. Additionally, due to typical everyday head movements, the acoustics can vary for a user as the headset moves in their ear. Therefore, with any leaky headset or earbud, some adjustment is required to ensure that the filters are always well matched to the acoustics.

[0007] Some headphones and earbuds already feature some form of off-ear detection, which detects whether the headphones are being worn by the user. Typically, this detection is achieved through a number of methods, including optical proximity sensors, pressure sensors, and capacitive sensors. However, off-ear detection can only distinguish between two extreme states of acoustic leakage: whether the headphones are on the ear or off. Furthermore, all listed solutions require the addition of additional sensors to the device specifically for this purpose.

[0008] The object to be achieved is to provide an improved concept for detecting acoustic leakage of an ear-worn playback device, such as a headset, earbuds or a mobile phone.

[0009] This object is achieved by the subject matter of the independent claims. Embodiments and developments of the improved concept are defined in the dependent claims.

[0010] The improved concept is based on the idea of estimating the leakage condition based on the degree of leakage. That is, during normal use of the ear-worn playback device, the degree of acoustic leakage between the ear-worn playback device and the user's ear canal is determined. This leakage condition can therefore enhance the user's sound experience, for example, by removing unwanted portions of the sound signal transmitted to the user's ear canal. For example, this enhancement can be achieved by adjusting a noise control algorithm based on the estimated leakage condition. For example, the FF and / or FB filters of a noise-canceling headphone can be tuned based on the degree of acoustic leakage.

[0011] In contrast, tuning of the aforementioned filters for conventional earbuds and headphones is currently performed only once during or at the end of ANC device production, for example by measuring the device's acoustic properties. Specifically, adjustments are performed in the ear canal of an artificial head using a measurement fixture (e.g., an artificial head) during a calibration process. The measurements (including the playback of test sounds) are coordinated by some sort of processing device, which can be a personal computer, for example. To achieve optimal ANC performance for each ANC device produced, dedicated measurements must be performed for each device under the control of the processing device, which is time-consuming, especially when calibrating a large number of ANC devices.

[0012] The improved concept will be explained below, sometimes using headphones or earbuds as an example of a playback device. However, it should be understood that this example is not limiting, and those skilled in the art will also understand other types of playback devices that may experience different leakage scenarios during user use. Generally speaking, the term playback device should include all types of audio reproduction devices.

[0013] In an embodiment of an audio system according to the improved concept, the audio system is intended for use in an ear-worn playback device, such as a headset, earbuds, mobile phone, or cell phone. The system includes a speaker and an error microphone configured to sense the sound being output from the speaker and ambient sound. The audio system also includes a detection engine configured to determine a driver response between the speaker and the error microphone and estimate a leakage condition based on the determined driver response.

[0014] For example, the speaker of an audio system is arranged in the housing of a playback device so that a first volume is arranged on the side that is preferentially used for sound emission from the speaker. The housing may have an opening for coupling the first volume to the volume of the user's ear canal. The housing may also include a front vent covered with an acoustic resistor that couples the first volume to the surrounding environment. Due to the imperfect fit of the earplug in the user's ear, the front volume will also be coupled to the surrounding environment through acoustic leakage. This acoustic leakage varies from person to person and depends on the position of the earplug in the ear at a specific time. The error microphone is arranged within the first volume so that it detects the sound output from the speaker as well as the ambient sound. For example, it is arranged close to the opening.

[0015] Furthermore, a second volume is arranged within the housing on a side of the loudspeaker remote from the preferred side for sound emission. This second volume is acoustically coupled to the surrounding environment via a rear vent in the housing, which may also be covered with an acoustic resistor. An additional microphone may, for example, be arranged outside the rear volume, i.e., on the exterior of the housing, in order to primarily sense ambient sound.

[0016] The detection engine is configured to determine the driver response between the loudspeaker and the error microphone. The driver response corresponds to the driver (i.e., loudspeaker) and the error microphone transfer function. For example, the loudspeaker output is a desired signal, such as a music signal, and is therefore detected by the error microphone in addition to ambient noise due to acoustic leakage.

[0017] The detection engine is further configured to estimate the leakage condition based on the determined driver response. Estimating the acoustic leakage based on the determined driver response provides a solution for adjusting the noise cancellation filter of the feedforward, feedback, or hybrid ANC system for situations where the desired signal (e.g., music) is large relative to the ambient noise. In these situations, it may be impossible to determine the coherence between the ambient noise signal at the other microphone (e.g., the feedforward microphone) of the audio system and the ambient noise signal at the error microphone (i.e., the feedback microphone), which causes the adaptive noise cancellation process to deviate significantly from the optimal solution or oscillate around the optimal solution, resulting in a suboptimal noise cancellation process.

[0018] Other systems with leakage detection capabilities achieve this by adjusting and monitoring a filter that matches the driver response. In contrast, the improved concept of estimating leakage directly from the driver response eliminates the error in matching the filter to the driver response, which is particularly problematic when the filter adaptation has not yet converged.

[0019] In some embodiments, determining the driver response includes measuring a property of a first signal applied to the loudspeaker, measuring a property of a second signal detected by the error microphone, and calculating the driver response based on the first and second properties.

[0020] For example, the leakage condition is estimated based on the difference between a desired signal (e.g., a music signal) applied to the loudspeaker and a signal detected by an error microphone (i.e., an error signal). For example, this difference is proportional to the determined driver response. The properties of the first and second signals include the amplitude, energy level, or average amplitude of frequency components of the first and second signals, respectively, i.e., the average value over several bins of a Fast Fourier Transform.

[0021] In some embodiments, the property of the first signal and the second signal is the amplitude of the respective signals.

[0022] For example, the reference signal level (ie, amplitude) is taken from the signal applied to the driver, while the response signal level is taken from the signal detected by the error microphone.

[0023] In some embodiments, to calculate the driver response, the first signal and the second signal are bandpass filtered using a predetermined bandpass frequency range.

[0024] As described above, the first signal and the second signal may be referred to as a reference signal and a response signal, and may be bandpass filtered to a frequency range where the driver response differs significantly for different acoustic leaks. For example, the predetermined frequency range is low acoustic frequencies such as 80-200 Hz.

[0025] In some embodiments, the driver response is calculated as a ratio of energy levels of the first signal and the second signal.

[0026] The energy level of each signal is determined, for example, by the amplitudes of the first and second signals, and a ratio of the energy levels is calculated to determine the driver response.For example, the ratio of the energy levels results in a scalar indication of the driver response.

[0027] In some embodiments, the driver response is calculated based on response values determined at predetermined frequencies or frequency ranges of the first signal and the second signal, respectively.

[0028] For example, the signal amplitudes or energy levels of the reference signal and the response signal are evaluated at a specific frequency or frequencies.

[0029] Typically, the driver response is evaluated only within a frequency band of interest, such as a frequency band that varies due to leakage variations. Therefore, the detection engine in these embodiments is configured to evaluate the first and second signals only within the frequency band of interest, while ignoring, for example, the driver response outside of that frequency band. For example, only the driver response may be evaluated and compared to a predetermined response between 100 Hz and 1 kHz.

[0030] For example, the detection engine evaluates the first signal and the second signal at multiple different frequencies, such as at least three different frequencies within the audio band. For example, the amplitude or energy level of the first signal and the second signal is monitored at each of the at least three frequencies. The results are then used to determine multiple leak conditions at the at least three frequencies. Thus, a final leak condition can be determined based on the multiple acoustic leak conditions, for example, as an average.

[0031] In one embodiment, the respective energy levels of the first and second signals are optionally calculated at a plurality of discrete frequencies through a Goertzel filter, a peak filter, or a bandpass filter. As described above, the energy levels at different frequencies can be reduced to a scalar driver response factor for use in estimating leakage conditions.

[0032] In some embodiments, the driver response is calculated by applying a process, such as a frequency transform, to the first signal and the second signal that separates the energy into at least two frequency bands.

[0033] In these embodiments, the detection engine is configured to acquire the first signal and the second signal within a specific measurement time, and then perform processing, such as applying a peak filter or a Fourier transform, on the acquired first signal and the second signal, respectively. For example, a discrete or fast Fourier transform is obtained by decomposing the sequence of values of the first signal and the second signal into components of different frequencies. For example, this processing is applied to the entire frequency range.

[0034] Thus, after applying the processing to the first and second signals, the driver response is calculated by comparing the respective acquired signals. For example, the driver response is proportional to or derived from the difference or ratio of the respective acquired signals. While this approach increases the amount of processing, it additionally provides a higher degree of accuracy in calculating the driver response.

[0035] In some embodiments, calculating the driver response further includes determining a first value by applying the process to the first signal, determining a second value by applying the process to the second signal, and comparing the first value to the second value.

[0036] For example, the detection engine is configured to determine a characteristic value, such as an average value or a maximum value, of the respective signals at a predetermined frequency or frequency range, obtained after the processing is applied to the first signal and the second signal. The driver response is then calculated based on the first value and the second value, such as a ratio of the first value and the second value.

[0037] In some embodiments, the first and second values are calculated as averages of corresponding frequency transformed data points from the first signal and the second signal.

[0038] The detection engine in these embodiments is configured to calculate, for example, a first average value of the frequency transform of a first signal over an entire or specific frequency range, and a second average value of the frequency transform of a second signal over the same frequency range. Thus, for example, the detection engine calculates the driver response based on the ratio of the first average value to the second average value.

[0039] In some embodiments, after applying the processing to the first and second signals, the first and second values are determined for a predetermined frequency or frequency range.

[0040] Similar to calculating the driver response based on the energy levels of the first and second signals, the driver response can also be calculated based on corresponding frequency-transformed values acquired at a plurality of different frequencies. The first and second values can, for example, be averages of characteristic values of the corresponding frequency transforms acquired at a number of frequencies or frequency ranges. For example, to estimate leakage conditions, the driver response at different frequency intervals (e.g., lower and upper regions of the acoustic frequency band) is of interest because varying acoustic leakage significantly affects these frequency intervals.

[0041] In some embodiments, estimating the leakage condition includes determining a leakage value based on the determined driver response.

[0042] A convenient way to describe leakage conditions is to determine an actual leakage value that quantifies the current acoustic leakage condition. For example, the leakage value is calculated as a normalized value between 0 and 1, scaling the determined driver response to a predetermined maximum and / or minimum driver response. A leakage value of 0 represents the lowest possible acoustic leakage or no leakage, while a leakage value of 1 represents the maximum acceptable acoustic leakage, i.e., if the playback device has very high leakage between the front volume and the surrounding environment.

[0043] In some embodiments, the leakage value is determined by comparing the determined driver response to a reference value in a lookup table.

[0044] The detection engine may include a lookup table, for example stored in a memory of the detection engine, in which different values of the driver response are associated with corresponding leakage values. In this way, leakage values and leakage conditions can be determined without performing further calculations based on the driver response.

[0045] In some embodiments, a leak condition is determined if a ratio between the sound output from the speaker and the ambient noise ratio exceeds a set threshold.

[0046] In these embodiments, if the desired signal is large relative to, eg, greater than, the ambient noise at the error microphone location, a leakage condition may be determined solely by the driver response. In other cases, a leakage condition may be determined by another method, such as a conventional method.

[0047] In the case where the leakage condition is determined by another method, a conversion method between the leakage obtained from the driver response and the leakage determined by the other method may be necessary. Such a conversion method may be a lookup table.

[0048] In some embodiments, the audio system further includes an additional microphone, and the leakage condition is used to adjust a feedforward filter and / or a feedback filter and / or a compensation filter of the audio system. In this embodiment, when obtaining a ratio of the desired signal to the ambient noise, the ambient noise level can be obtained at the location of the additional microphone.

[0049] Particularly in situations where the desired signal (e.g., music) is large relative to the ambient noise, conventional methods of adjusting the noise cancellation process (i.e., feedforward and / or feedback filters) can result in an unstable noise cancellation process or oscillations around an optimal value, which can be perceived as disruptive by the user of the audio system. Therefore, in such situations, adjusting the noise cancellation process based on leakage conditions derived directly from the driver response provides a fast and accurate means. Furthermore, the compensation filter can be adjusted based on the leakage conditions to compensate for, for example, the desired signal, such as music, being attenuated by the feedback noise cancellation.

[0050] In some embodiments, a leakage condition is estimated when the ratio of the desired signal to the interference signal, particularly the ambient noise signal, is greater than a threshold. Otherwise, the leakage condition is estimated using a different method. In embodiments where estimating the leakage condition includes calculating leakage values, the leakage values calculated using different methods can be converted to the same scale, for example, normalized to a value between 0 and 1 using a predetermined lookup table.

[0051] The above object is also addressed by a signal processing method for an ear-worn playback device, the playback device including a speaker and an error microphone that senses sound being output from the speaker and ambient sound. The method includes generating an error signal via the error microphone and determining a driver response based on the error signal and a signal applied to the speaker. The method also includes estimating a leakage condition based on the determined driver response.

[0052] The error signal corresponds to or is derived from the signal detected by the error microphone, ie the desired signal from the loudspeaker, such as music and ambient sound.

[0053] Based on the above-described embodiments of the audio system, other embodiments of the signal processing method will become apparent to those skilled in the art.

[0054] The improved concept will be described in more detail below with the aid of the accompanying drawings. In all the accompanying drawings, elements having the same or similar functions have the same reference numerals. Therefore, their description need not be repeated in the description of the following drawings.

[0055] In the attached figure:

[0056] Figure 1 shows a schematic diagram of a headset;

[0057] Figure 2 shows a block diagram of a general adaptive ANC system;

[0058] Figure 3 An example representation of a "leaky" type earphone is shown;

[0059] Figure 4 shows an example headset worn by a user with several sound paths from ambient sound sources;

[0060] Figure 5 An example representation of an ANC-enabled mobile phone is shown;

[0061] Figure 6 A block diagram showing an exemplary embodiment of an audio system for an ear-worn playback device according to an improved concept; and

[0062] Figure 7 Signal graphs showing frequency-dependent driver responses for different acoustic leakage conditions are shown.

[0063] Figure 1A schematic diagram of an ANC-enabled playback device in the form of headphones HP is shown, which in this example are designed as ear-hook or circumaural headphones. Only a portion of the headphones HP corresponding to a single audio channel is shown. However, extension to stereo headphones will be apparent to the skilled reader. The headphones HP include a housing HS that carries a loudspeaker SP, a feedback noise microphone or error microphone FB_MIC, and an optional ambient noise microphone or feedforward microphone FF_MIC. The error microphone FB_MIC is specifically oriented or arranged so that it records both ambient noise and the sound played through the loudspeaker SP. Optionally, the error microphone FB_MIC is arranged close to the loudspeaker, for example, close to an edge of the loudspeaker SP or close to the loudspeaker's membrane. Optionally, the error microphone FB_MIC can be arranged close to the ear canal of a user of the headphones HP. The optional ambient noise / feedforward microphone FF_MIC is specifically oriented or arranged so that it primarily records ambient noise from outside the headphones HP.

[0064] Depending on the type of ANC to be performed, the ambient noise microphone FF_MIC can be omitted when only feedback ANC is performed. According to an improved concept, the error microphone FB_MIC can be used to provide an error signal, which is used as the basis for determining the wearing condition (correspondingly, the leakage condition) of the headset HP when the user wears the headset HP.

[0065] exist Figure 1 In an embodiment, the detection engine DET is located inside the headset HP for performing various signal processing operations, examples of which will be described in the following disclosure. The detection engine DET may also be located outside the headset HP, for example in an external device located in a mobile handset or phone, or in the cable of the headset HP.

[0066] Figure 2 A block diagram of a general adaptive ANC system is shown. The system includes an error microphone FB_MIC and a feedforward microphone FF_MIC, both of which provide their output signals to an adaptation engine ADP. The noise signal recorded by the feedforward microphone FF_MIC is also provided to a feedforward filter F, which generates an anti-noise signal output via a loudspeaker SP. The sound output from the loudspeaker SP is combined with the ambient noise at the error microphone FB_MIC and recorded as an error signal that includes the remaining portion of the ambient noise after ANC. This error signal is used by the sound adaptation engine ADP to adjust the filter response of the feedforward filter.

[0067] Figure 3An example representation of a "leaky" type earplug is shown, i.e. an earphone with some leakage between the surroundings and the ear canal EC. In particular, there is an acoustic path between the surroundings and the ear canal EC, which is indicated as "acoustic leakage" in the figure.

[0068] Figure 4 An example configuration of a headset HP worn by a user with several sound paths is shown. Figure 4 The headset HP shown in FIG2 can be used as an example of any ear-worn playback device of an audio system capable of noise cancellation, and can include, for example, in-ear headphones or earbuds, on-ear headphones or earhook headphones. In addition to the headset, the ear-worn playback device can also be a mobile phone or similar device.

[0069] The headset HP in this example has a loudspeaker SP, a feedback noise microphone FB_MIC and an optional ambient noise microphone FF_MIC which is designed as a feedforward noise cancellation microphone, for example. For a better overview, the internal processing details of the headset HP are not shown here.

[0070] exist Figure 4 In the configuration shown, there are several sound paths, each of which can be represented by its own acoustic response function or acoustic transfer function. For example, the first acoustic transfer function DFBM represents the sound path between the speaker SP and the feedback noise microphone FB_MIC, and can be referred to as the driver-to-feedback response function. The first acoustic transfer function DFBM may include the response of the speaker SP itself. The second acoustic transfer function DE represents the acoustic sound path between the speaker SP of the headphone (potentially including the response of the speaker SP itself) and the user's eardrum ED exposed to the speaker SP, and can be referred to as the driver-ear response function. The third acoustic transfer function AE represents the acoustic sound path between the ambient sound source and the eardrum ED through the user's ear canal EC, and can be referred to as the environment-ear response function. The fourth acoustic transfer function AFBM represents the acoustic sound path between the ambient sound source and the feedback noise microphone FB_MIC, and can be referred to as the environment-to-feedback response function. The driver response subject to the present disclosure is generated by the first acoustic transfer function DFBM, that is, the ratio of the total sound signal detected by the error microphone FB_MIC to the total signal driving the speaker SP.

[0071] If the ambient noise microphone FF_MIC is present, the fifth acoustic transfer function AFFM represents the acoustic sound path between the ambient sound source and the ambient noise microphone FF_MIC, and may be referred to as an ambient-to-feedforward microphone response function.

[0072] The response function or transfer function of the headset HP, in particular the response function or transfer function between the microphones FB_MIC and FF_MIC and the loudspeaker SP, can be used together with the feedback filter function B and the feedforward filter function F, which can be parameterized as noise cancellation filters during operation.

[0073] A headset HP, as an example of an ear-worn playback device, can be implemented with both microphones FB_MIC and FF_MIC activated or enabled to perform hybrid ANC, or as an FB ANC device in which only the feedback noise microphone FB_MIC is activated while the ambient noise microphone FF_MIC is not present or at least not activated. Therefore, in the following, if a signal or acoustic transfer function involving the ambient noise microphone FF_MIC is used, it is assumed that the microphone is present; otherwise, it is assumed to be optional.

[0074] For a better overview, Figure 4 Any processing of the microphone signals or any signal transmission is omitted. However, the processing of the microphone signals for the purpose of performing ANC may be implemented in a processor located within the headset or other ear-worn playback device, or in a dedicated processing unit external to the headset. The processor or processing unit may be referred to as an adaptation engine. If the processing unit is integrated into the playback device, the playback device itself may form an audio system capable of noise cancellation. If the processing is performed externally, the external device or processor together with the playback device may form an audio system with noise cancellation enabled. For example, the processing may be performed in a mobile device like a mobile phone or mobile audio player, to which the headset is connected either wired or wirelessly.

[0075] In various embodiments, the FB or error microphone FB_MIC may be located in a dedicated cavity, such as described in detail in ams application EP17208972.4.

[0076] Now refer to Figure 5 , another example of a noise cancellation enabled audio system is given. In this example embodiment, the system is formed by a mobile device, such as a mobile phone MP, which includes a playback device with a loudspeaker SP, a feedback or error microphone FB_MIC, an ambient noise or feedforward microphone FF_MIC, and an adaptation engine ADP for performing ANC and / or other signal processing during operation.

[0077] In another embodiment not shown, the headset HP (e.g. similar to Figure 1 or Figure 4A headset (as shown in FIG) can be connected to a mobile phone MP, wherein signals from the microphones FB_MIC, FF_MIC are transmitted from the headset to the mobile phone MP, in particular to the processor PROC of the mobile phone, for generating an audio signal to be played through the loudspeaker of the headset. For example, depending on whether the headset is connected to a mobile phone, ANC is performed using the internal components of the mobile phone (i.e., the loudspeaker and microphone) or using the loudspeaker and microphone of the headset, thereby using different sets of filter parameters in each case.

[0078] Hereinafter, several embodiments of the improved concept will be described in conjunction with specific use cases. However, it is obvious to those skilled in the art that the details described in this embodiment can still be applied to other embodiments.

[0079] Figure 6 A block diagram of a hybrid ANC audio system according to an improved concept is shown. The system includes an error microphone FB_MIC and a feedforward microphone FF_MIC. The noise signal recorded by the feedforward microphone FF_MIC is supplied to a first feedforward noise filter F for generating an anti-noise signal, which is output through a speaker SP along with a desired signal (e.g., music). At the error microphone FB_MIC, the sound output from the speaker SP is combined with the ambient noise and recorded as an error signal that includes the remaining portion of the ambient noise after ANC. This error signal is supplied to a second feedback noise filter B for generating another anti-noise signal, which is added to the anti-noise signal and the desired signal and also output through the speaker SP.

[0080] The total signal applied to the loudspeaker SP and the error signal from the error microphone FB_MIC are further provided to the detection engine DET for determining a driver response and a subsequent estimate of the leakage condition. For example, the driver response is calculated based on these two signals and subsequently evaluated and compared with known driver responses for different leakage conditions, for example, stored in a lookup table, to determine a leakage value that quantifies the actual leakage condition of the earphone. The adaptation engine ADP then uses the leakage value to adjust the filter response of the feedforward filter F and / or the feedback filter b.

[0081] As detailed in ams patent US 9,779,718 B2, the hybrid system in this embodiment also includes an optional music compensation filter C. In this case, the desired signal (e.g., music) is provided to the music compensation filter C in order to compensate for the desired signal attenuated by, for example, feedback noise cancellation.

[0082] Figure 7A signal graph showing the amplitude of the frequency-dependent driver response for different acoustic leakage conditions is shown. For example, the labeled low leakage driver response corresponds to a no-leakage, i.e., close-to-the-ear condition with no or insignificant acoustic leakage between the ear canal and the surroundings, while the labeled high leakage driver response corresponds to a maximum, i.e., a condition with significant acoustic leakage between the ear canal and the surroundings. Intermediate leakage conditions then result in driver response amplitudes between the aforementioned high and low leakage conditions, as shown in FIG. Figure 7 . For example, a typical range of possible amplitudes of the driver response between a minimum and a maximum value is about 30 dB, which may also be highly frequency dependent. For example, the driver response exhibits a significant (i.e., largest) leakage dependence at low frequencies. Therefore, the detection engine can be configured to evaluate only the signal applied to the loudspeaker SP and the error signal from the error microphone FB_MIC within this frequency range, e.g., between 10 Hz and 200 Hz. For example, this can be achieved by bandpass filtering or a fast Fourier transform of the signals.

[0083] The detection engine DET may be configured to evaluate the determined driver response and compare it to predetermined minimum and maximum driver responses over a frequency range or a number of different frequencies. Thus, a leakage value quantifying the leakage condition may be determined, for example as a normalized value between 0 and 1, where 0 represents a minimum leakage condition and 1 corresponds to a maximum leakage condition.

[0084] Reference numerals

[0085] HP Headphones

[0086] HS housing

[0087] SP driver or speaker

[0088] FB_MIC Error or feedback microphone

[0089] FF_MIC Ambient or feed-forward microphone

[0090] F Feedforward filter

[0091] B Feedback filter

[0092] C Compensation filter

[0093] ADP Adaptation Engine

[0094] DET Detection Engine

[0095] EC ear canal

[0096] ED eardrum

[0097] DFBM drive to feedback response function

[0098] DE Driver to Ear Response Function

[0099] AE Environment to Ear Response Function

[0100] AFBM environment to feedback response function

[0101] AFFM Environment to Feedforward Response Function

[0102] MP Mobile Phone

Claims

1. An audio system (AS) for a wearable device (HP), comprising - loudspeaker (SP); - an error microphone (FB_MIC) configured to sense the sound being output from the speaker (SP) and ambient sound; and - a detection engine (DET) configured to - measure a property of a first signal applied to said loudspeaker (SP); - measuring a property of a second signal detected by said error microphone (FB_MIC); - determining a driver response between the loudspeaker (SP) and the error microphone (FB_MIC) by calculating the driver response based on properties of the first signal and properties of the second signal; and - Estimating the leakage condition by comparing the determined actuator response with known actuator responses under different leakage conditions.

2. The audio system (AS) according to claim 1, wherein The property of the first signal and the second signal is the amplitude of the respective signal.

3. The audio system (AS) according to claim 1, wherein To calculate the driver response, the first and second signals are bandpass filtered at a predetermined bandpass frequency range.

4. The audio system (AS) according to claim 1, wherein The driver response is calculated as a ratio of energy levels of the first signal and the second signal.

5. The audio system (AS) according to claim 1, wherein The driver response is calculated based on response values determined at predetermined frequencies or frequency ranges of the first and second signals, respectively.

6. The audio system (AS) according to claim 1, wherein The driver response is calculated by applying a process to the first and second signals that separates energy into at least two frequency bands.

7. The audio system (AS) according to claim 6, wherein The processing is frequency conversion.

8. The audio system (AS) according to claim 6, wherein Calculating the driver response also includes - determining a first value by applying said processing to said first signal; - determining a second value by applying said processing to said second signal; and - comparing said first value with said second value.

9. The audio system (AS) according to claim 8, wherein After applying the processing to the first and second signals, a first value and a second value are determined for a predetermined frequency or frequency range.

10. The audio system (AS) according to claim 1, wherein Estimating the leakage condition includes determining a leakage value based on the determined driver response.

11. The audio system (AS) according to claim 10, wherein The leakage value is determined by comparing the determined driver response with reference values in a lookup table.

12. The audio system (AS) according to claim 1, wherein: The audio system further comprises a further microphone (FF_MIC), and the leakage condition is used to adjust a feedforward filter (F) and / or a feedback filter (B) and / or a compensation filter (C) of the audio system (AS).

13. The audio system (AS) according to claim 1, wherein: The leakage condition is estimated when a ratio of the desired signal to the interference signal is greater than a threshold.

14. The audio system (AS) according to claim 1, wherein: The driver response is determined without adjusting and monitoring the filter to match the driver response.

15. An ear-worn playback device (HP) comprising an audio system (AS) according to any one of claims 1 to 14.

16. A signal processing method for an ear-worn playback device (HP), the ear-worn playback device comprising a speaker (SP) and an error microphone (FB_MIC), the error microphone sensing a sound being output from the speaker (SP) and an ambient sound, the method comprising: - generating an error signal via the error microphone (FB_MIC); - measuring a property of said error signal; - measuring a property of a signal applied to said loudspeaker (SP); - determining a driver response between a loudspeaker (SP) and an error microphone (FB_MIC) by calculating the driver response based on properties of the error signal and properties of a signal applied to the loudspeaker (SP); as well as - Estimating the leakage condition by comparing the determined actuator response with known actuator responses under different leakage conditions.

Images