Method for processing ambient sound acquired by audio device capable of being worn on or in ear, and corresponding apparatus

Abstract

In a method according to the invention for processing ambient sounds acquired by an audio device (10) capable of being worn on or in the ear, a reference signal is generated (S1) as a function of the acquired ambient sounds. A measurement signal is generated (S1) from sound acquired in an ear canal of a user of an audio device capable of being worn on or in the ear. The reference signal is processed (S2) into a first signal and the measurement signal is processed (S2) into a second signal. A factor is determined (S3) based on the first signal and the second signal such that the scale for measuring the signal strength of the second signal matches the scale for measuring the signal strength of the first signal. A third signal based on the reference signal is weighted (S4) with the determined factor. And outputting (S5) a third signal weighted with the measured factor.

Description

Technical Field

[0001] This invention relates to a method for processing ambient sound captured by an audio device that can be worn on or inside the ear. Specifically, the audio device can be a headset or hearing aid, but can also be other audio devices such as hearing protectors, audio-enabled glasses, or headsets. The invention also relates to a corresponding apparatus. Background Technology

[0002] Headphones with an ambient mode (also known as a transparency mode, open mode, or pass-through mode) compensate for the passive attenuation of ambient sound caused by headphones. The purpose of this ambient mode is, for example, to make ambient sounds sound as if no headphones were being worn when wearing headphones with an active ambient mode. This ambient mode can also be used in hearing aids to compensate for hearing loss or improve speech intelligibility in noisy environments. Headphones with ambient modes and hearing aids with ambient modes are hereinafter collectively referred to as portable audio devices with ambient modes. It is self-evident that they can be worn on or inside the ear, and therefore, for simplicity, will not be explicitly stated here. Furthermore, portable audio devices may also include, for example, hearing protectors, glasses with audio functionality, or headsets.

[0003] In a portable audio device with an ambient mode, ambient sounds are recorded by an external microphone, converted into digital signals by an analog-to-digital converter, processed on a digital signal processor, then converted back into analog signals by a digital-to-analog converter, and played through one or more speakers of the portable audio device. In addition to at least one speaker and at least one external microphone, this portable audio device also has an internal microphone, which can be used to reduce structural and ambient sounds in the ear canal via a feedback modulator.

[0004] To achieve the most natural possible hearing, the decisive factor is the type of sound reaching the eardrum. This sound depends particularly on passive sound (ambient sound passively attenuated by the portable audio device) and active sound played by the speaker. Both the passive attenuation characteristics and the transfer function from the speaker to the eardrum depend on how well the portable audio device is fitted to the individual user. For example, with classic headphones in ambient mode, the personalized fit may result in uncomfortable frequency amplification or unnatural frequency attenuation compared to open-ear headphones. To avoid such interference effects with hearing aids, fine-tuning of volume, timbre, and gain is required for different frequencies, which typically requires a hearing aid specialist to invest time and money. Summary of the Invention

[0005] The object of the present invention is to provide an improved method and a corresponding device that can achieve the most natural possible auditory effect even when the audio device is partially or completely closed off when the ear is worn on or inside the ear.

[0006] This objective is achieved by a method having the features of claim 1 and a corresponding apparatus according to claim 10. The preferred design of the invention is the subject of the dependent claims.

[0007] In the method according to the invention for processing ambient sound captured by an audio device that can be worn on or inside the ear, the following steps are performed: - Generate reference signals based on the collected ambient sounds; - Generates measurement signals based on the sound collected in the ear canal of a user who can wear an audio device on or inside the ear; - Process the reference signal into a first signal and the measured signal into a second signal; - Based on the measurement factors of the first and second signals, so that the scale used to measure the signal strength of the second signal matches the scale used to measure the signal strength of the first signal; - Weight the third signal based on the reference signal using the measured factors; and - Output a third signal weighted by the measured factors.

[0008] According to a preferred embodiment of the present invention, - Divide the first signal and the second signal into frequency bands to generate first and second frequency band signals; - A factor is determined for each frequency band based on the associated first and second frequency band signals, such that the scale used to measure the signal strength of the associated second frequency band signal matches the scale used to measure the signal strength of the associated first frequency band signal; - Apply the measured factor or a filter based on the measured factor to the third signal so that the third signal is weighted by the factor in the corresponding frequency band; - Output the weighted third signal.

[0009] According to another preferred embodiment of the present invention, - Divide the third signal into frequency bands to generate a third frequency band signal; - Multiply the third frequency band signal with the relevant measurement factor; - Summate the frequency band signals multiplied with the relevant measurement factors; - Output sum signal.

[0010] According to another preferred embodiment, the reference signal is processed into a first signal and the measurement signal is processed into a second signal, such that the ratio of the signal strength of the first signal and the second signal is equivalent to the ratio of the sound pressure level at the user's eardrum when the ear canal is open and when the ear canal is closed by an audio device that can be worn on or inside the ear.

[0011] Furthermore, the first signal is processed advantageously through gain factor, dynamic processing, and / or interference noise suppression.

[0012] According to another preferred embodiment, the factor is recursively calculated as the quotient of the signal strengths of the first signal and the second signal weighted by the factor.

[0013] According to another preferred embodiment, the third signal weighted by the factor is delayed, and the factor is calculated as the quotient of the signal strength of the delayed third signal multiplied by the factor and the signal strength of the second signal.

[0014] Advantageously, the output signal is filtered by a positive filter and then played by a speaker of an audio device that can be worn on or inside the ear.

[0015] According to another embodiment of the invention, the reference signal is filtered by a positive filter before being processed into a first signal, and the output signal is played by a speaker of an audio device that can be worn on or inside the ear.

[0016] Advantageously, particularly in environmental modes where audio devices can be worn on or inside the ear, it at least partially compensates for the passive attenuation of ambient sound caused by audio devices that can be worn on or inside the ear, and this compensation is personalized according to the user of the audio device that can be worn on or inside the ear.

[0017] Accordingly, the apparatus of the present invention for processing ambient sound acquired by an audio device capable of being worn on or inside the ear includes a processor configured to: - Generate reference signals based on the collected ambient sounds; - Generates measurement signals based on the sound collected in the ear canal of a user who can wear an audio device on or inside the ear; - Process the reference signal into a first signal and the measured signal into a second signal; - Based on the measurement factors of the first and second signals, so that the scale used to measure the signal strength of the second signal matches the scale used to measure the signal strength of the first signal; - Weight the third signal based on the reference signal using the measured factors; and - Output a third signal weighted by the measured factors.

[0018] According to an embodiment of the present invention, the apparatus of the present invention includes: - A reference sensor for acquiring ambient sound and generating a reference signal based on the acquired ambient sound; - A measurement sensor for acquiring sound from the ear canal of a user who can wear an audio device on or inside the ear, and generating a measurement signal based on the acquired sound; and - Speakers, used to play audio signals based on ambient sounds processed by a processor.

[0019] The processor is advantageously configured to apply at least one filter bank to divide the first signal and the second signal into frequency bands.

[0020] Furthermore, the processor can be configured to further apply at least one filter bank to divide the third signal into frequency bands.

[0021] The device of the present invention can be integrated in particular into audio devices that can be worn on or inside the ear.

[0022] In particular, audio devices that can be worn on or inside the ear can be designed as headphones or hearing aids.

[0023] The present invention also relates to a computer program comprising instructions that cause a computer to perform the steps of the method of the present invention.

[0024] The following description and claims are consistent with the appendix. Figure 1 This reveals other features of the present invention. Attached Figure Description

[0025] Figure 1 The diagram schematically shows an in-ear headphone and its main electronic components inside the ear canal.

[0026] Figure 2 A flowchart of the method of the present invention is shown schematically.

[0027] Figure 3 A first block diagram showing the processing of ambient sound collected by a sensor according to the present invention is shown.

[0028] Figure 4 A second block diagram showing the processing of ambient sound collected by a sensor according to the present invention is shown.

[0029] Figure 5 A third block diagram showing the processing of ambient sound collected by a sensor according to the present invention is shown.

[0030] Figure 6 A block diagram showing the calculation of factors according to the present invention is shown.

[0031] Figure 7The amplitude response of each bandpass filter in the filter bank is schematically shown.

[0032] Figure 8 A block diagram of the apparatus of the present invention for determining factors in a frequency band is shown.

[0033] Figure 9 A block diagram of the present invention’s apparatus for applying factors in a filter bank structure is shown.

[0034] Figure 10 A block diagram of another inventive device for applying weighting rules and processing reference signals in a filter bank structure is shown.

[0035] Figure 11 A block diagram of a hybrid portable audio device is shown, which uses the method of the present invention to process ambient sound collected by sensors. Detailed Implementation

[0036] To better understand the principles of this invention, embodiments will be described in detail below with reference to the accompanying drawings. It is to be understood that this invention is not limited to these embodiments, and the described features can be combined or modified without exceeding the scope of protection defined in the claims.

[0037] Figure 1 A portable audio device 10, in the form of an in-ear headphone, is shown as an example of a device that can implement the method of the present invention. However, the method can also be used with other types of headphones, hearing aids, hearing protectors, headsets, or audio-enabled glasses. The in-ear headphone 10 is held in the ear canal 15 of a male or female user by earplugs 17, which completely or partially insulate it from noise. Depending on the fit of the earplugs 17 to the individual user, the ear canal 15 can be opened to varying degrees by a ventilation device 18. The portable audio device 10 may also be equipped with a dedicated ventilation device in the housing or earplugs 17.

[0038] The headset 10 is equipped with at least one reference sensor in the form of an external microphone 11, which acquires ambient sound and generates a reference signal accordingly. Furthermore, at least one measurement sensor, in the form of an inward-facing microphone 12, is provided on the side of the headset facing the eardrum 16 within the ear canal. This measurement sensor acquires sound within the ear canal and generates a measurement signal accordingly. The headset 10 is equipped with at least one processor 14, which processes the ambient sound recorded by the external microphone 11 according to the present invention. Specifically, the at least one processor 14 may be one or more digital signal processors. Based on the ambient sound processed by the processor 14, an audio signal is played by a speaker 13 inside the headset. The processed ambient sound played by the speaker can, for example, allow the user to clearly hear surrounding sounds while wearing the portable audio device 10, despite passive attenuation of ambient sound caused by earplugs, or compensate for the user's hearing loss, or protect the user's hearing from transient noise.

[0039] In addition to the ventilation device 18, the acoustic impedance driven by the speaker 13 may also depend on other individual factors, such as the user's ear geometry or the impedance of their eardrum. Therefore, the transfer function from the speaker 13 to the inner microphone 12 and to the eardrum 16 may vary from person to person. Similarly, the passive attenuation characteristics of the portable audio device 10 may also vary from person to person. For example, when the ventilation device 18 is larger, more ambient sound will passively reach the eardrum 16. These individual acoustic characteristics may result in undesirable sound pressure levels at the eardrum 16. Conversely, the relative transfer function from the inner microphone 12 to the eardrum 16 and the sound pressure level at the outer microphone 11 are less affected by individual acoustic characteristics.

[0040] Based on the reference signal, a target level acting on the tympanic membrane can be determined in a manner unaffected by individual acoustic characteristics. Conversely, based on the measured signal, the actual level currently acting on the tympanic membrane, which is influenced by individual acoustic characteristics, can be estimated. One objective of the method of this invention is to adjust the actual level to the target level.

[0041] Figure 2 A flowchart of the method of the present invention is schematically shown, including basic steps for processing ambient sound collected by sensors of an audio device that can be worn on or inside the ear. This method is particularly useful for transparent modes of personalized headphones or hearing aids, but is not limited thereto.

[0042] In the first step S1 of this method, sensor signals are acquired. Specifically, ambient sound is acquired via an external microphone to generate a reference signal, and sound from the ear canal is acquired via an internal microphone to generate a measurement signal. In the second step S2, the reference signal and the measurement signal are preprocessed into a first signal and a second signal, respectively. In the third step S3, a factor is calculated based on the first and second signals. In the fourth step S4, this factor is used to weight the third signal based on the reference signal. In the fifth step S5, the third signal weighted with the measurement factor is output. Specifically, the weighting with the measurement factor can be achieved by multiplying by the factor.

[0043] After steps S1 to S5, other processing steps not shown in the flowchart may be performed, such as digital-to-analog conversion, and subsequent playback through one or more speakers.

[0044] The invention will be further explained below with several block diagrams that illustrate different implementations of the method. These block diagrams illustrate the signal flow and the elements for processing the signals, represented by the illustrated blocks. The functions of the various elements shown in the figures can be provided by corresponding software modules or dedicated hardware units of the processor. For clarity, if a block has the same function in different block diagrams, it will be described only once.

[0045] Figure 3 A first block diagram is shown illustrating the processing of ambient sound acquired by a sensor according to the present invention. The external microphone's reference signal is also shown. The signal is filtered by equalizer 25 to generate an open tympanic membrane signal. The open tympanic membrane signal is equivalent to an estimate of the signal reaching the tympanic membrane when no portable audio device is worn. Measurement signal. The signal is filtered by equalizer 24 to generate a closed tympanic membrane signal. The closed-eardrum signal is equivalent to an estimate of the signal currently reaching the eardrum when wearing a portable audio device.

[0046] For example, the equalizer can be determined through measurements using a dummy head. To do this, the signal at the open-drum microphone of the dummy head can first be measured to obtain a clear external excitation signal, without using a portable audio device. Then, the device can be used to record the signals from the external and internal microphones, as well as the signal from the closed-drum microphone, under the same excitation. The equalizer 25 can then be determined as a relative transfer function from the external microphone signal to the open-drum signal, and the equalizer 24 can be determined as a relative transfer function from the internal microphone signal to the closed-drum signal.

[0047] The signal strength estimators 22 and 22' can be used to estimate the open and closed tympanic membrane signals. and Calculate signal strength and These ratios of signal intensities correspond to the ratios of sound pressure levels at the open and closed tympanic membranes. Estimators 22, 22' can, for example, be implemented as exponential smoothers of the signal absolute values, with a smoothing factor of... : Alternatively, the squared signal can be smoothed to calculate the short-term power, and then the square root can be obtained.

[0048] Now, the reference signal With factors Multiply, with Generate the modified signal. It can then undergo further processing and be played back by the speaker 13 of the portable audio device 10. Here, it is assumed that the signal... The sound pressure level generated at the eardrum is higher than the sound pressure level of ambient sound reaching the eardrum through other pathways. When the portable audio device 10 passively attenuates and suppresses high-frequency components, while, for example, a feedback modulator 52 (such as...) Figure 11 This assumption is particularly applicable when suppressing low-frequency components of ambient sound (as shown). In this case, the measured signal... Signal strength and modified signal The signal strength is proportional to the signal strength of the reference signal after factor scaling. Since the modified signal calculated using the factor is played by speaker 13 and correspondingly acquired by internal microphone 12, the signal strength of the measured signal and the signal strength of the occult membrane signal have been affected by the factor according to the aforementioned equation. For example, to compensate for this effect, either the occult membrane signal can be calculated... Before the signal strength is determined, the factor is divided by the factor delayed by the delay element 21 by the estimator 22', or as follows: Figure 3 As shown, in calculating the open tympanic membrane signal Before calculating the signal strength, the factor is multiplied by the estimator 22'. .if If it is established, then and The ratio corresponds to the ratio of the sound pressure level at the open and closed tympanic membranes. Then, the factor can be calculated in block 23. For example, as a quotient of these signal strengths: By factor With reference signal Multiply by 26 to produce And played through the speaker 13 of the portable audio device 10 Then, using the aforementioned equalizers 24 and 25, the sound pressure level (actual level) at the eardrum when wearing a portable audio device is adjusted to the sound pressure level (target level) at the eardrum when not wearing a portable audio device.

[0049] The equalizers 24 and 25 may also be designed differently from those described above to achieve purposes other than acoustic transparency. For example, block 25 may also include dynamic processing or interference noise suppression to determine the improved target level, thereby protecting the user's hearing from noise interference or achieving hearing improvement. Figure 10 Relevant examples are shown.

[0050] Figure 4 A second block diagram showing the processing of ambient sound collected by a sensor according to the present invention is shown. Here, the equalizer 24 can implement a transfer function, which is equivalent to... Figure 2 The ratio of the transfer functions of equalizers 25 and 24. Therefore, the ratio of the signal strength of the equalized measured signal to the reference signal remains equal to the ratio of the sound pressure levels at the open and closed tympanic membranes, thus eliminating the need for equalizer 25. This saving is particularly advantageous when the method is used in a filter bank structure, as it eliminates either the need for an additional filter bank or a copy of equalizer 25.

[0051] Figure 5 A third block diagram illustrating the processing of ambient sound acquired by a sensor according to the present invention is shown. Here, the factor is not calculated based on the signal strength of a reference signal multiplied by a delay factor, but rather on a modified signal delayed by delay element 21. The signal strength is calculated from this. Therefore, the multiplication operation can be omitted.

[0052] Figure 6 The calculation factor according to the present invention is shown. The block diagram of block 20 shows that this calculation can be used, for example, in the following filter bank implementation. This structure is characterized by receiving a measurement signal that has been modified, for example, by equalizer 24. and reference signal And only output factors .

[0053] Figure 7 An example is shown of the amplitude response 30 of a filter bank that divides the input signal into K frequency bands. Each filter in the filter bank has an impulse response. Ideally, a filter should be designed to block frequencies below the lower limit. and frequencies above the upper limit The frequency components of the input signal, and allow and The frequency components between these frequencies pass through without distortion. From this, the following objective function for the filter amplitude response is derived, which is related to the frequency... Related: In addition, the total number of bandpass filters is 30. Ideally, the following conditions should be met: This ensures that, especially when bandpass filters are connected in parallel, no undesirable cancellation or enhancement phenomena occur in the transition region.

[0054] Figure 8 A block diagram of an apparatus according to the present invention for applying block 20 in a filter bank structure 40 is shown. Figure 3 Similarly, the reference signal and the measurement signal are filtered by the relevant equalizers 24 and 25, and converted into modified reference signals and measurement signals. The modified reference signal and the modified measurement signal are then separated into frequency band signals by the filter group analysis 41. and Then, these frequency band signals are input into algorithm 20 (e.g., as shown in the image). Figure 6 As shown in the figure, the factor for each frequency band is calculated. The thick lines in the block diagram indicate that these are multi-channel signals and factors. These factors are then input into unit 45, which... Perform filtering so that factors can be used in the corresponding frequency band. for Weighting. This can be based on factors. The coefficients of a filter can be calculated, selected, or otherwise determined. Similarly, filtering in the frequency domain can be achieved by weighting the frequency coefficients accordingly. Figure 9 Another form of implementing unit 45 is shown by weighting the frequency band signal.

[0055] Implementing this method within a filter bank structure has advantages because different factors can be applied across different frequency bands. For example, in a single-channel implementation of this method, it may be more common for low-frequency measurements to be below the reference level while high-frequency measurements are above the reference level, but these deviations cannot be compensated for simultaneously.

[0056] Figure 9 Another block diagram of an apparatus according to the present invention for applying block 20 in a filter bank structure 40 is shown. Figure 4Similarly, the measurement signal is filtered by equalizer 24. The reference signal and the modified measurement signal are respectively filtered and analyzed by filter group 41 to be divided into frequency band signals. and And it is input into Algorithm 20. In this case, Figure 8 The design of unit 45, used for filtering the reference signal, involves combining the factor of each frequency band with the reference frequency band signal. Multiply, then sum in step 42 when the filters are combined, and then... Output in the form of .

[0057] Figure 10 Another block diagram of an apparatus according to the present invention for applying a weighting algorithm 20 in a filter bank structure 44 is shown. Here, the reference frequency band signal input into the algorithm 20 is first processed by block 43. The signal strength of the reference frequency band signal input into the algorithm is modified through processing. This block may correspond to, for example, dynamic compression, interference noise suppression, or simple scaling. This can protect the user's hearing from noise interference, and / or improve hearing by eliminating interfering environmental signals, and / or adjust sound perception according to user settings.

[0058] The parameters of block 43 can also be configured by the user. For example, the user can select a factor to scale the reference band signal in block 43. Since the algorithm factor in block 20 is calculated based on the scaled reference band signal, and the reference band signal is scaled again using the algorithm factor before synthesis 42, the measurement level and the corresponding sound pressure level at the eardrum when the user wears the portable audio device are scaled according to the user-modified reference level. Accordingly, the reference band signal can be cleaned up by interference noise suppression or compressed using a dynamic compressor so that algorithm 20 can adjust the measurement level accordingly based on this interference noise suppression or dynamic compression.

[0059] Such a process 43 can also be integrated into Figure 3 and Figure 4 In the single-channel method shown, for example in Figure 3 In the middle, after equalizer 25; or in Figure 4 In the middle, after the reference signal branch and before multiplying by the delay factor of 26.

[0060] Figure 11 A block diagram of a hybrid portable audio device 10 employing method 44 of the present invention is shown, hereinafter referred to as Figure 10 Taking the filter bank structure in the example, the portable audio device is equipped with an external microphone 11, an internal microphone 12, and a speaker 13.

[0061] Among them, audio signals (For example, music or a phone call) can be filtered by equalizer 50 and then played by speaker 13. Furthermore, the signal from internal microphone 12 is calculated using the output signal of quadratic path estimation 53. The signal is filtered by feedback regulator 52 and then played back by speaker 13. Since the signal played from speaker 13 into the ear canal is recorded by internal microphone 12, regulator 52 forms a closed regulation loop. Regulator 52 can be designed to attenuate sounds in the ear canal (e.g., passively attenuated ambient or structural sounds) within a specific frequency range. For example, this can compensate for the blockage effect caused by ear canal closure.

[0062] The quadratic path estimation 53 is equivalent to estimating the transfer function from the speaker 13 to the internal microphone 12 (i.e., the so-called quadratic path). The quadratic path estimation 53 can be used to reduce the influence of the regulator 52 on the audio signal. The effect of the output signal of the forward filter 51. Switch 54 can be set as follows: in switch position 1, only the equalized audio signal is transmitted to the secondary path estimation; in position 2, no signal is transmitted to the secondary path estimation; in position 3, the sum of the equalized audio signal and the output signal of the forward filter is transmitted to the secondary path estimation.

[0063] Signals from the external microphone 11 and the internal microphone 12 are transmitted to the device 44. The processed ambient sound signal is output from the device 44 and filtered by the positive filter 51. It is then played through the speaker along with signals from the regulator 52 and the equalizer 50. The positive filter can advantageously be designed such that the sound pressure level at the eardrum when the portable audio device 10 is worn is comparable to the sound pressure level at the eardrum when the portable audio device is not worn. However, the positive filter can also be of other types. For example, it can only convert the sound pressure level according to the open ear in a specific frequency domain, or it can be set for targeted enhancement / attenuation.

[0064] and Figure 11 Conversely, as shown in the block diagram, the signal calculated from the output signal of the inner microphone 12 using the secondary path estimation 53 can be used in advance by performing an addition operation 55, replacing the signal of the inner microphone 12 before being sent for processing 44. For example, this can reduce the impact of the audio signal on the processing of ambient sound.

[0065] Similarly, a positive filter can be applied to the signal from the external microphone first, and then provided to method 44. After this method, other positive filters can be omitted. Ideally, the effect of the positive filter should be considered when measuring equalizers 24 and 25.

[0066] Alternatively, the sum of the audio signal optionally equalized by equalizer 50 and the signal from external microphone 11 optionally filtered by forward filter 51 can be provided to the method instead of the signal from external microphone 11. For example, through the processing 44 according to the invention, not only the level of ambient sound can be adjusted, but also the level of the audio signal. Furthermore, only the equalized audio signal can be input to processing unit 44. In this case, the method of the invention will only adjust the level of the audio signal.

[0067] Reference number list

Claims

1. A method for processing ambient sound acquired by an audio device (10) that can be worn on or inside the ear, wherein - Generate a reference signal (S1) based on the collected ambient sound; - A measurement signal (S1) is generated based on the sound collected in the ear canal of the user of the audio device that can be worn on or inside the ear; - Process the reference signal (S2) into a first signal, and process the measurement signal (S2) into a second signal; - Based on the first signal and the second signal, a (S3) factor is determined so that the scale used to measure the signal strength of the second signal matches the scale used to measure the signal strength of the first signal; - Weight the third signal based on the reference signal using the measured factors (S4); and - Output (S5) the third signal weighted by the measured factors.

2. The method according to claim 1, wherein - Divide the first signal and the second signal into frequency bands to generate a first frequency band signal and a second frequency band signal; - For each frequency band, a factor is determined based on the relevant first frequency band signal and the second frequency band signal, so that the scale used to measure the signal strength of the relevant second frequency band signal matches the scale used to measure the signal strength of the relevant first frequency band signal; - Apply the measured factor or a filter based on the measured factor to the third signal so as to weight the third signal with the factor in the corresponding frequency band; - Output the weighted third signal.

3. The method according to claim 2, wherein - Divide the third signal into frequency bands to generate a third frequency band signal; - Multiply the third frequency band signal by the relevant measurement factor; - Summate the frequency band signals multiplied with the relevant measurement factors; - Output sum signal.

4. The method according to any one of the preceding claims, wherein the reference signal is processed (S2) into a first signal and the measurement signal is processed (S2) into a second signal such that the ratio of the signal strength of the first signal and the second signal is equivalent to the ratio of the sound pressure level at the user's tympanic membrane when the ear canal is open and when the ear canal is closed by the audio device that can be worn on or inside the ear.

5. The method of claim 4, wherein the first signal is further processed by gain factor, dynamic processing and / or interference noise suppression.

6. The method according to any one of the preceding claims, wherein the factor is recursively calculated (S3) as the quotient of the signal strengths of the first signal weighted by the factor and the second signal.

7. The method of claim 5, wherein the third signal weighted by the factor is delayed, and the factor is calculated (S3) as the quotient of the signal strength of the delayed third signal multiplied by the factor and the signal strength of the second signal.

8. The method according to any one of the preceding claims, wherein the output signal is filtered by a positive filter and then played by a speaker of the audio device that can be worn on or inside the ear.

9. The method according to any one of claims 1 to 7, wherein the reference signal is filtered by a positive filter before being processed into the first signal, and the output signal is played by a speaker of the audio device that can be worn on or inside the ear.

10. The method according to any one of the preceding claims, wherein, in an ambient mode, passive attenuation of ambient sound caused by the audio device that can be worn on or in the ear is at least partially compensated, the compensation being personalized for the user of the audio device that can be worn on or in the ear.

11. An apparatus for processing ambient sound acquired by an audio device (10) that can be worn on or inside the ear, comprising a processor (14) configured to: - Generate a reference signal based on the collected ambient sound; - A measurement signal is generated based on the sound collected in the ear canal of the user of the audio device that can be worn on or inside the ear; - Process the reference signal into a first signal and the measurement signal into a second signal; - Based on the measurement factors of the first signal and the second signal, such that the scale used to measure the signal strength of the second signal matches the scale used to measure the signal strength of the first signal; - Weight the third signal based on the reference signal using the measured factors; and - Output a third signal weighted by the measured factors.

12. The apparatus of claim 11, wherein the apparatus comprises: - Reference sensor (11), the reference sensor is used to acquire the ambient sound and generate a reference signal based on the acquired ambient sound; - Measurement sensor (12), the measurement sensor is used to collect sound in the ear canal of the user of the audio device that can be worn on or inside the ear, and generate a measurement signal based on the collected sound; as well as - A speaker (13) for playing audio signals based on ambient sounds processed by the processor (14).

13. The apparatus of claim 11 or 12, wherein the processor (14) is further configured to apply at least one filter bank (41) to divide the first signal and the second signal into frequency bands.

14. The apparatus of claim 13, wherein the processor (14) is further configured to apply the at least one filter bank (41) to divide the third signal into frequency bands.

15. The device according to any one of claims 11 to 14, wherein the device is integrated into the audio device that can be worn on or inside the ear.

16. The device of claim 15, wherein the audio device capable of being worn on or inside the ear is designed as a headset (10) or a hearing aid.

17. A computer program comprising instructions that cause a computer to perform the steps of the method according to any one of claims 1 to 10.

Images