CONFIGURABLE MICROPHONE RAY AND METHOD FOR CONFIGURATING A MICROPHONE RAY

Description

Field of invention

[0001] The invention relates to microphone arrays and in particular microphone arrays with automatic beam focusing. background

[0002] Microphone arrays use a variety of microphone capsules and combine their output signals to achieve a specific directionality of the array. The direction of a sound source can be detected by analyzing the delays between the sound signals arriving at the microphone capsules. Similarly, the directionality of a microphone array can be achieved by combining the delayed output signals of the microphones. While some applications only require directionality in a specific, unchanging direction, others, such as conference phones, require multiple focus areas or even variable directionality of the microphone array, where the direction of increased sensitivity can be controlled by adjusting or changing the respective delays.

[0003] Microphone arrays can also use automatic beam focusing to automatically adjust their directionality to a currently active speaker. For example, the microphone array can determine the direction of a currently active speaker by detecting the direction of a sound signal that has the maximum volume or energy among all the sound signals. In practice, the microphone array can scan sound signals from various predefined directions within a predefined search area. It can determine and compare the respective energies of sound signals from these directions, identify a direction of maximum sound energy, and adjust the delays for its microphone capsules to create a directionality in the identified direction. Such a microphone array is known, for example, from US 2017 / 164101 A1. It can, for example,It can be used for a conference phone that can focus on and track a speaker who is positioned or moving anywhere in the room.

[0004] However, problems can arise if there is a sound source within the search area whose sound should not be picked up by the microphone array. For example, interfering noise might be audible through an open window or door. For microphone arrays installed in conference rooms, for instance, interfering sound sources can be electronic devices such as projectors, speakers, fans, or air conditioners. Especially if the interfering sound source is close to the microphone array or during periods of silence in the room, such as extended pauses in speech, the microphone array can detect the interference. The interfering sound might even be the highest-energy sound signal in the search area, causing the microphone array to focus on it. During pauses in speech, the focus of the microphone array can then jump back and forth between a speaker and the interfering source, which can degrade the signal quality of the speech signal.It is particularly disruptive when the microphone array picks up and amplifies the sound from the source of the interference. Another microphone array is known from US 2016 / 012827 A1. Devices for emitting a control sound signal are known from EP 1 465 455 A2, US 7 092 763 B1, or US 2012 / 134507 A1. Summary of the invention

[0005] One of the problems addressed by the invention is therefore to prevent a microphone array with multiple focus areas or with automatic beam focusing from focusing on acoustic interference sources. In particular, the problem is to be solved flexibly for different types of acoustic interference sources, namely those whose spatial position relative to the fixed microphone array is only known after the microphone array has been installed, and which are stationary at least for a certain period of time (e.g., the typical length of a telephone call) or permanently, such as a fan. Furthermore, the problem is also to be solved for temporarily occurring interference sources whose spatial position is fixed, but which only become apparent as interference sources during a telephone call, such as an open window. In addition, the problem is also to be solved for mobile, i.e., not permanently installed, microphone arrays.

[0006] At least the aforementioned problems are solved by a microphone array according to claim 1 and by a method according to claim 12. Further embodiments are defined in the dependent claims.

[0007] The microphone array according to the invention can detect a predefined, automatically generated control sound signal played from the direction of an interfering sound source and configure itself according to the direction and the detected control sound signal. The control sound signal can, for example, contain an audio signature. In one embodiment, the invention relates to a method in which the microphone array searches a defined search area for sound sources, detects the played predefined control sound signal, determines the direction from which it originates, and then configures the microphone array according to the control sound signal and the determined direction. The configuration can, for example, include a specific amplification / attenuation or the elimination of the determined direction from the search area, or the deletion of a previously configured specific amplification / attenuation or the elimination of a direction.

[0008] A configurable microphone array according to one embodiment of the invention comprises a plurality of microphone capsules, a beamforming unit with configurable directivity, at least one direction control unit for automatically controlling and adjusting the configurable directivity, and a configuration control unit with a control signal detector. The configuration control unit is capable of recognizing an acoustic signal detected by the microphone capsules as a predefined control acoustic signal and its direction of reception, and of generating at least one control signal corresponding to the control acoustic signal. The control signal is used to configure the direction control unit according to the control acoustic signal. In some embodiments, the beamforming unit, the direction control unit, and the configuration control unit can all receive input signals directly from the microphone capsules.In some embodiments, the direction control unit can also be integrated into the beam shaping unit.

[0009] One advantage of the invention is that it provides a convenient and user-friendly solution for supporting the rapid, flexible, and simple configuration of microphone arrays. In particular, regions of the search area can be marked, allowing the microphone array to process signals from different regions differently. The number of these regions is variable. The solution according to the invention is intuitive and easy to use. Another advantage is that no defined position or orientation of the microphone array in space is necessary for configuration; that is, the microphone array can be positioned arbitrarily or even contained within a mobile device. The configuration can then be adapted to the specific situation. Eliminating regions of the search area can significantly reduce the error rates of microphone arrays and improve audio quality. Brief description of the drawings

[0010] Further advantageous embodiments are presented in the following description and the accompanying drawings. These show Fig. 1 shows a conference room with a ceiling-mounted microphone array and interfering sound sources; Fig. 2 shows a block diagram of a microphone array according to a first embodiment; Fig. 3 shows a block diagram of a microphone array according to a second embodiment; Fig. 4 shows a block diagram of a microphone array according to a third embodiment; Fig. 5 shows a block diagram of the beamforming unit in the third embodiment; Fig. 6 shows a flowchart of a method according to an embodiment; and Fig. 7 shows exemplary tone sequences as control sound signals. Detailed description

[0011] Fig. 1Figure 1 shows an example of a conference room with a ceiling-mounted configurable microphone array 100 and several sources of acoustic interference, for example a projector 120, an air conditioner 130 and a pair of loudspeakers 140, 140'. In this example, the microphone array 100 is mounted in the center of the room, above a table 150. Fig. 1 a) is a schematic side view of the conference room and Fig. 1 b)This is a schematic top view. The microphone array 100 can initially search for sound sources in all directions within its search or detection range, detect one or more dominant sound sources, and record and output their sound signals. The sound sources can be speakers, e.g., participants in a telephone conference who are seated at table 150, standing in the conference room, or moving around within it. The search range can extend to the entire room. Various techniques can be used to search for sound sources, e.g., a predefined grid of test points 107. It is also possible to exclude an area 105 of the room from the search range altogether, e.g., due to the design of the microphone array.While conventional microphone arrays can be disrupted by unwanted sound from various sources 120, 130, 140, the invention offers a simple way to block one or more specific sources of interference from the microphone array's detection range: A user can place a portable electronic device 110 capable of playing back sound, such as a smartphone or an acoustic remote control, between the configurable microphone array 100 and the source of interference 130, and can play back a specific control sound signal using the portable electronic device 110. This could be, for example, a recorded sound signal or a synthetic sound signal generated by a computer program. A speech signal would be unsuitable because it would require speech recognition and training on specific speakers.

[0012] The microphone array essentially detects 100 sound signals and the spatial direction from which they are received. It can also detect the direction of the loudest sound signal. In one embodiment, only a portion of the available microphone capsules are used, thus reducing processing overhead. In another embodiment, all microphone capsules are used. The search for a dominant sound source can be repeated at regular intervals, e.g., every 30 ms, or performed continuously. In each time interval, a dominant sound source is detected, the microphone array's beamforming is focused on it in the current or next interval, and its sound signal is recorded and output. The direction can be determined, for example, as a pair of elevation and azimuth angles.If the microphone array 100 uses a grid of test points 107, it can detect the direction of incidence of the sound signal using the respective test point(s) instead of the angles. Interpolation between multiple test points is also possible to detect a direction lying between test points. In one embodiment, several or all test points of the grid can be analyzed simultaneously or at least within the aforementioned time interval. In another embodiment, the grid contains only a few test points corresponding to known positions of potential sound sources. In each case, the elevation and azimuth angle pair, or the detected or interpolated test point, represents a spatial sector whose size corresponds to the spatial resolution of the microphone array as the direction of incidence of the sound signal.

[0013] This method for detecting an acoustic signal can be used both for useful signals, such as speech signals that are output for further processing, and for the specific control acoustic signals that, according to the invention, serve to control the microphone array. These signals are specific in that they contain coded information or a signature. This could be, for example, a simple sequence of tones or a melody, an acoustic signal with a specific modulation, or a special audio watermark. The control acoustic signal can be in the audible frequency range or at least partially in the ultrasonic range. An example is given below in connection with Fig. 7The microphone array 100 can detect and decode the control sound signals to obtain the information they contain by comparing the received sound signal with one or more predefined control sound signals. If the received control sound signal corresponds to a predefined control sound signal, the microphone array 100 performs a corresponding configuration, e.g., regarding its beam steering and spatial directivity. Furthermore, since the received control sound signal is an audio signal, it can be output via the microphone array's normal audio output. In one embodiment, however, the microphone array suppresses the output of the control sound signal after recognizing that the received control sound signal corresponds to a predefined control sound signal and is therefore not a speech or useful signal.In one embodiment, the detection of the control sound signal occurs simultaneously with and independently of the search for the dominant sound signal, so that the control sound signal can be detected even if it is not the dominant sound signal. Both processes are similar and can use the same microphone signals, as described below.

[0014] Various configuration options are possible. For example, the detected spatial direction from which a control sound signal was received can be excluded from the microphone array's search area. This makes it very easy for the user to remove an interference source from the microphone array's search area: by simply removing it. Fig. 1In the example shown, the user can hold the portable device 110, for example, between the microphone array 100 and the air conditioner 130. When the user plays a specific control sound signal, which can be generated, for example, by a smartphone using an app, the microphone array detects the control sound signal and its direction, decodes the information contained therein internally into an electrical control signal meaning "Exclude this direction," and configures itself by excluding the corresponding direction from its search area. At least any useful signals from this direction are then ignored. The direction to be excluded is a spatial direction, i.e., it can be defined by a pair of angles consisting of an elevation angle θe and an azimuth angle φa, or, for example, as a test point or a group of test points 137. It is referred to below as the exclusion sector. The microphone array 100 stores this information until, for example,is overwritten or deleted by another configuration.

[0015] In one embodiment, adjacent sectors or directions can be excluded from the search area if the microphone array receives sound from them during the described configuration. It is also possible to control the size of the exclusion sector by placing the portable electronic device 110 closer to the microphone array 100 (to enlarge the sector) or further away from the microphone array (to shrink the sector) while playing the control sound signal. The user can repeat the process for one or more additional sources of interference. The microphone array will then exclude all of these from the search area by generating and storing exclusion sectors for each, e.g., in Fig. 1a first exclusion sector 125 for a projector 120, a second exclusion sector 135 for an air conditioner 130, and a third and fourth exclusion sector 145,145' for a wall-mounted pair of loudspeakers 140,140'. While in Fig. 1 a) The elevation angles θ e of the exclusion sectors are shown, as is the case with the following: Fig. 1 b) whose azimuth angle φ a . For example, the third and fourth exclusion sectors 145,145' for the wall-mounted loudspeakers have the same elevation angle but different azimuth angles. Due to the limited spatial resolution, the terms "angle" and "direction" always refer to angular ranges. Each test point 107,137,147 in Fig. 1 a)This represents a beam with a specific elevation and azimuth angle range, or a beam through the microphone array and the test point, and the portable device 110 can be positioned anywhere along this beam. However, the detected or excluded area may be smaller if the portable device 110 is located further away from the microphone array due to the spatial resolution.

[0016] In one embodiment, the microphone array 100 can detect and distinguish several different, predefined control sound signals played back by the portable device 110, each corresponding to a different control command. The microphone array 100 can decode these signals to receive and implement the respective control command, i.e., configure itself accordingly.

[0017] For example, a previous exclusion sector configuration can be overridden. This can be achieved by defining a control sound signal containing the control information "Clear the exclusion of this sector." This is useful, for instance, if a source of interference has been moved or if an incorrect exclusion sector was accidentally defined. Such control is easier to implement in embodiments where the search for a dominant sound source and the detection of control sound signals are separate, concurrent processes, as described below. In other embodiments, the microphone array may not receive the control sound signal because it originates from a direction excluded from the search area. In this case, however, the control information can be "Clear the exclusion of adjacent sectors" and played back from an adjacent, non-excluded sector.In one embodiment, a special control sound signal is defined for deleting all exclusion sectors. In another embodiment, the microphone array can store one or more configurations for later recall, e.g., for several different conference rooms.

[0018] As another example, the microphone array can be configured via a control command to modify the processing of sound signals from a specific spatial direction; for instance, a particular gain factor can be applied. This direction is referred to below as the marker sector. This control command can also be overridden by another control command.

[0019] The control sound signal reproduced by the portable electronic device 110 is preferably not a speech signal, but in one embodiment a synthesized or electronically generated acoustic signal. This makes it easier to distinguish from useful (speech) signals that can be recorded simultaneously. In particular, speech recognition, which is usually very complex, is then not necessary for decoding.

[0020] Fig. 2Figure 1 shows a block diagram of a microphone array in a first embodiment. The microphone array 100 comprises an arrangement 200 of a plurality of microphone capsules 210, which has a configurable directivity. The arrangement 200 also includes control electronics (not shown), e.g., an analog-to-digital converter (ADC) and adjustable delay elements. The microphone array 100 further comprises a direction control unit 250 and a configuration control unit 240. The direction control unit 250 is configured to control the configurable directivity of the microphone arrangement 200, e.g., to adjust the adjustable delay elements. The configuration control unit 240 is configured to determine whether an audio signal detected by the microphone arrangement 200 is a control audio signal and corresponds to a predefined control audio signal.The configuration control unit 240 is further adapted to send one or more control signals CTR to the direction control unit 250 in order to configure it according to the detected sound signal or the predefined control sound signal. The configuration control unit 240 and / or the direction control unit 250 can be implemented by one or more processors, e.g., signal processor units. In the illustrated embodiment, the microphone arrangement 200 contains delay elements for microphone signals picked up by the microphone capsules 210 and therefore has an inherent directionality. To change the directionality or beamforming, the delay elements are configurable, and the direction control unit 250 generates control signals 255 for their configuration. However, the picked-up sound signals, and thus the output signal 230 of the microphone arrangement 200, contain not only control signals but also speech or other audio signals.Useful signals are received from one or more directions determined by beam focusing. These are output at output 220 for speech output and further processing, such as filtering (not shown). In one embodiment, the configuration control unit 240 can detect a dominant sound source and adjust the directionality of the microphone array accordingly by beam shaping using the direction control unit 250, as described above.

[0021] In the illustrated exemplary embodiment, the configuration control unit 240 includes a comparator 241, a control signal generator 242, and a direction detector 243. The direction detector 243 is configured to detect a first direction from which the microphone arrangement 200 has received a sound signal. The comparator 241 is configured to compare the received sound signal with at least one predefined control sound signal and to detect that the received sound signal corresponds to the control sound signal. The control signal generator 242 is configured to generate at least one electrical control signal CTR corresponding to the type and direction of the received control sound signal. If the comparator 241 has not detected a predefined control sound signal, the received sound signal can be considered a useful signal and output at output 220. However, if the comparator 241, for example, detects a predefined control sound signal, the received sound signal can be considered a useful signal and output at output 220.Once the first predefined control sound signal has been detected, it is decoded and processed, for example, by the control signal generator 242 outputting the information corresponding to the control sound signal via the electrical control signal CTR. Alternatively, the control signal generator 242 can output the information corresponding to the control sound signal as a separate control signal. The electrical control signal CTR, and optionally also the separate control signal, is output to the direction control unit 250 to configure the microphone array 200 with respect to the first direction according to the received (useful) sound signal and, if applicable, according to the received control sound signal.

[0022] In one embodiment, the microphone array 100 includes, or is connected to, an electronic memory 500 in which data is stored defining one or more predefined control sound signals and their meaning or configuration commands. The configuration control unit 240 can compare the recorded sound signal, or certain parameters derived from it, with the one or more predefined control sound signals or their parameters stored in the memory 500 and detect that the recorded sound signal matches one of the predefined control sound signals, e.g., the first control sound signal. This can be done in the comparator 241. Subsequently, the configuration control unit 240 can generate the electrical control signal CTR according to the detected control sound signal. The direction control unit 250 can be configured using the electrical control signal CTR.

[0023] Searching for sound signals in the direction detector 243 and receiving sound signals from previously located or precisely defined sound sources as useful signals, which are then output, may require focusing in different directions for beamforming. This can be done alternately in time-division multiplexing or simultaneously. In the first case, short pauses in speech can be used for the search. In the second case, for example, some of the microphone capsules 210 can be used for the search and others for receiving the useful signals. Alternatively, in the second case, at least some of the microphone capsules 210 can have two or more different, individually configurable delay elements, so that these microphone capsules can be used for both processes simultaneously. Preferably, fewer microphone capsules are used for the search than for receiving the useful signals.

[0024] Several configurations of the directional control unit 250 are possible via the configuration control unit 240. In one embodiment, the directional control unit 250 can be configured to generate a directional effect for the microphone arrangement 200, suppressing the first direction. This means that sound signals from the direction from which the control sound signal was received are ignored. In embodiments where the search for a dominant sound source and the detection of control sound signals are separate, simultaneous processes, at least the search for a dominant sound source in the first direction is omitted or suppressed. In one variant, this can also mean that no sound is received from the first direction at all, while in another variant, the detection of control sound signals remains possible even from the first direction.In another embodiment, the direction control unit 250 can be configured to disable the search for a dominant sound source. A current focus can be maintained, or a default configuration can be set, such as omnidirectional sensitivity or a specific directionality. In a further embodiment, a previously made configuration of the direction control unit 250 can be reversed, such as the suppression of a first direction or the suppression of a second direction that differs from the first.

[0025] As described above, the first and second directions represent spatial areas and can be considered marker sectors. Another possible application for marker sectors is speaker detection or identification of speakers or participants in a telephone conference. For example, brands or labels can be inserted as metadata into the audio signal.

[0026] Fig. 3Figure 1 shows a block diagram of a microphone array according to a second embodiment. The output signals 315 of an arrangement of microphone capsules 310 are output to a sound signal direction detector 350, a control signal detector 30, and a beamforming unit 320. Digitization using analog-to-digital converters (ADCs) can be performed directly at the microphone capsules 310 or in the other blocks 320, 30, and 350. The beamforming unit 320 contains, for example, adjustable delay elements to obtain a microphone signal with a specific directionality. The delay elements within the beamforming unit 320 are controlled according to a direction signal 355, which indicates the current direction of a dominant speech or sound signal that is to be detected and output by the beamforming unit 320.The directional signal 355 is generated by the sound signal direction detector 350, which scans various directions simultaneously or sequentially for a dominant sound source. For this purpose, the output signals 315 from different microphone capsules are correlated with each other, e.g., in pairs, and the results are analyzed (GCC, Generalized Cross Correlation). For this, the so-called SRP-PHAT (Steered Response Power - PHASE Transform) algorithm, as described in US2017 / 164101 A1, can be used. To reduce computational effort, it may be sufficient to use the output signals from only some of the microphone capsules. In one embodiment, the output signals from less than half of the available microphone capsules are used for directional detection. By specifying fixed grid points, the corresponding signal delays are also fixed, and thus the signals for different directions can be queried, accumulated, and compared simultaneously.

[0027] According to this embodiment of the invention, each grid point can be individually omitted during this search by the acoustic signal direction detector 350 deactivating it or ignoring the corresponding measured value. In this case, each deactivated grid point corresponds to an exclusion sector. Alternatively, a marker can be set for each grid point individually, thereby defining a marker sector. The marker can cause special processing of acoustic signals from the corresponding marker sector, e.g., increased or reduced amplification, activation of speech recognition, and / or automatic translation. A control signal corresponding to the marker can be output to the beamforming unit or to a subsequent unit (not shown) together with the direction control signal 355. For an exclusion sector, it is not necessary for the acoustic signal direction detector 350 to output a specific control signal.To generate or implement the exclusion sector or marking sector, the acoustic signal direction detector 350 receives a direction signal XS, which indicates the receiving direction of the control acoustic signal and thus the affected direction. If several different configuration commands are possible, e.g., several different markings in the case of a marking sector, the acoustic signal direction detector 350 can receive a control signal in addition to the direction signal XS, which indicates the type of marking, e.g., "gain". Alternatively, a control signal CS can be sent directly to the beam shaping unit 320 or to other external units.

[0028] Such a control signal CS, like the direction signal XS, is provided by the control signal detector 30. This detector also receives the (optionally digitized) output signals 315 from the microphone capsules to detect whether and from which direction control sound signals are received. Here, too, it is possible to use the output signals of only some of the available microphone capsules 310 and / or to perform the detection of control signals only at specific grid points. However, improved detection of the control sound signal and / or the direction of reception can be achieved if the output signals of all available microphone capsules 310 are used. In one embodiment, the control signal detector 30 can, for example, contain several individual detectors 330, which detect the reception of a control sound signal by means of cross-correlation of the received signal of each microphone capsule with a predefined control sound signal.The detection signals 335 of the individual detectors 330 are correlated with each other in a correlation unit 340 to determine the reception direction XS of the control sound signal. One or more control sound signals or their parameters can be stored in a memory 500 to provide them to the individual detectors. The control signal detector 30 can also check the input sound signal 315 sequentially for several different control sound signals using time-division multiplexing, and / or there can be several control signal detectors 30 that output separate direction signals XS and control signals CS.

[0029] In the Fig. 3In the illustrated embodiment, the detection of the control sound signals in the control signal detector 30 occurs simultaneously with the audio output at output 325 of the beamforming unit 320, and independently of its orientation. Therefore, the control sound signal, provided it has a certain minimum volume, can be played from any direction within the search area to be detected, and it does not need to be the sound signal with the highest energy in the search area.

[0030] Fig. 4Figure 1 shows a block diagram of a microphone array according to a third embodiment. In this embodiment, the search area contains several detection sectors S1, S2, S3, which are not movable but permanently fixed or at least fixed for a certain period of time, e.g., for the typical duration of a telephone call. For example, each detection sector S1, S2, S3 can point to a participant in a telephone conference. The detection sectors can also be pre-programmed and remain fixed until they are reprogrammed or reconfigured. This feature is useful for detecting only sound sources whose directions are temporarily fixed but not yet known when the microphone array is installed. There are also one or more control signal detectors 40, which in one embodiment are the same control signal detectors 30 as described above. In this case, control sound signals can be received from all directions.In another embodiment, however, each control signal detector 40 is directed only at one of the detection sectors S1, S2, S3, so that its output signal S1x, S2x, S3x relates only to the respective detection sector. Each control signal detector 40 contains several individual detectors 430, each of which detects a control sound signal, as well as a correlation unit 440, which calculates the direction of reception of the control sound signal from the output signals 435 of the individual detectors and / or assigns a detected control sound signal to one of the detection sectors S1, S2, S3. In one embodiment, the beamforming unit 420 contains one submodule 421, 422, 423 per detection sector S1, S2, S3, which focuses on the respective detection sector and receives sound signals from there.The output signals of submodules 421, 422, and 423 are combined, superimposed, or added together in a combiner 424 to obtain an audio output signal 425 from the microphone array. Each of the submodules 421, 422, and 423 can be configured, modified, and / or switched on or off for the respective detection sector by the corresponding output signal S1x, S2x, and S3x. If the position of the detection sectors is programmable, e.g., by control sound signals as described above, then both the correlation units 440 and the submodules 421, 422, and 423 of the beamforming unit are programmed accordingly. In principle, however, control by control sound signals is independent of whether the position of the detection sectors S1, S2, and S3 is fixed or variable.

[0031] Fig. 5Figure 1 shows a block diagram of the beamforming unit 420 for a microphone array of the third embodiment. The beamforming unit 420 can be considered a multi-focus beamforming unit and contains one or more submodules 421, 422, 423, each of which receives several or all of the microphone signals 315 as input signals. In this example, each submodule 421, 422, 423 is connected via a configuration bus (CFB) to a controller 4201, e.g., a processor, through which the respective configuration of each submodule can be read or changed. Each submodule is essentially identical in structure and contains a single-focus beamforming unit 421b and a configuration control unit 421d. The single-focus beamforming unit 421b forms a directional characteristic from the microphone signals 315, thereby covering the first detection sector S1 and enabling sound signals from there to be received and output at the output of the submodule.The position of each detection sector S1, S2, S3, i.e., the directional characteristics of the single-focus beamforming units and also of the correlation unit 440, and other parameters can be programmed by means of the controller 4201. This can be arranged inside or outside the multi-focus beamforming unit 420.

[0032] The configuration control unit 421d receives the output signals S1x, S2x, and S3x from the respective control signal detector 40. This control signal contains information based on a control sound signal received in the respective detection sector. This could be, for example, a gain or attenuation factor to be applied to (useful) sound signals received in this sector, or a cutoff signal that turns the respective detection sector into an exclusion sector. The configuration control unit 421d stores this configuration and generates a control signal DC1 for the single-focus beam shaping unit 421b, enabling it to shape the beam accordingly. The configuration control unit 421d can be configured or read from the controller 4201 via the CFB bus.However, this is normally only a security or backup access, since the inventive method of configuring the microphone array during operation is easier and faster than configuration via the 4201 controller.

[0033] Fig. 6Figure 600 shows a flowchart of a method for configuring a microphone array in one embodiment of the invention. The method 600 is executed automatically by the microphone array 600 and comprises the (simultaneous) scanning 610 of a plurality of directions for sound signals by means of an arrangement of several microphone capsules, detecting 620 a sound signal from a first direction, determining 625 that the detected sound signal corresponds to a first predefined control sound signal, generating 640 a first electrical control signal according to the first predefined control sound signal, and configuring 650 the microphone array according to the first predefined control sound signal. The scanning 610 of the search area for sound signals or sound sources can be performed in several directions simultaneously or sequentially.In one embodiment, determining 630 that the detected sound signal corresponds to a first predefined control sound signal can also include the detection of a signature. In this case, too, the first electrical control signal is generated by the portable electronic device 110 according to the first predefined control sound signal.

[0034] In one embodiment, the configuration process 650 includes excluding the first direction from the search area. In another embodiment, the configuration process 650 includes limiting the search area to the first direction, so that directions other than the first are not searched for sound signals. In one embodiment, the configuration process 650 includes deleting or reversing a current configuration or setting a default configuration. In a further embodiment, a computer with a user interface (e.g., a screen and input keys) and control programs can additionally be used to delete or revise a previously made configuration.

[0035] In one embodiment, the method 600 further comprises determining 660 that the detected sound signal corresponds to a second predefined control sound signal that differs from the first predefined control sound signal, generating 670 a second electrical control signal according to the second control sound signal, and configuring 680 the microphone array according to the second electrical control signal. Configuring 680 the microphone array according to the second electrical control signal may include storing a current configuration or modifying processing for sound signals received from the first direction, e.g., modifying a gain value for sound signals from the first direction. In one embodiment, determining 660 that the detected sound signal corresponds to a second predefined control sound signal includes detecting a signature in a sound signal.In one embodiment, determining 630,660 that the received sound signal corresponds to a predefined control sound signal includes comparing parameters of the received sound signal with stored parameters that correspond to the predefined control sound signal.

[0036] In one embodiment, the portable electronic device is a portable computer or a smartphone. The control sound signal or signals can be automatically generated according to an algorithm or program, for example, in an app. In one embodiment, a user can use a sound-generating device, such as a smartphone or tablet computer, to emit structured audible signals. These can be received and translated by a microphone array to mark areas within its search or detection range. For example, one structured audible signal might contain a signature meaning "Mark this area." Another structured audible signal might contain a signature meaning "Clear the mark on this area" or "Save the current configuration." In one embodiment, the control sound signal might also contain variable, unpredefined information, such as...a name under which a configuration should be saved.

[0037] Fig. 7This shows frequency responses of two exemplary tone sequences, which can be contained as structured audible signals in a control sound signal or signature. In this example, each tone sequence begins with an identifier that makes the sound signal more easily detectable as a control sound signal. In the examples shown, the identifier consists of a tone of a fixed frequency f4 for a duration of t1 - t0, which serves to synchronize the following tone sequence. This is followed by a sequence of four tones with different frequencies, each with a predetermined (in this case, approximately the same) length, which contain the actual information. In one case, the tone sequence has the frequencies f3 - f5 - f4 - f3, and in the second case, the frequencies f2 - f3 - f1 - f4. The sequences can be recognized by the microphone array and assigned to control commands or control signals. The tone sequence can be very short, e.g.,The sequence can be 30 ms or 100 ms long, and only needs to be long enough for the microphone array to identify and locate it. The frequencies lie within a range detectable by the microphone array, typically between 16 Hz and 20 kHz or slightly higher. Higher frequencies above 12-15 kHz have the advantage of being easier to locate and are barely or inaudible to many people, and therefore less disruptive. The sequence can also be hidden as an audio watermark within another audio signal that serves as the carrier. It should be noted that the audio sequences shown may be greatly simplified to illustrate the principle. The control audio signals can also contain longer sequences with more information, which may, for example, be encrypted.

[0038] The invention simplifies and improves the operation and configuration of microphone arrays compared to conventional solutions, such as graphical input using a site plan. According to the invention, no site plan is required. Furthermore, unlike GPS-based solutions, the solution according to the invention can also be used indoors, and the probability of errors is reduced. Another advantage of the invention is that a feedback channel from the microphone array to the portable electronic device 110 used for configuration is not strictly necessary.

[0039] In one embodiment, the invention relates to a non-transient, computer-readable storage medium with instructions stored thereon, in the execution of which a computer performs a method for configuring a microphone array as described above.

[0040] The invention is advantageously applicable to configurable microphone arrays, in particular those that use automatic beam focusing.

[0041] It is understood and expressly intended that the various described and illustrated embodiments can be combined with one another. Even if such a combination is not explicitly mentioned, it is considered an embodiment of the invention. Components such as the direction control unit or the beam shaping unit can be implemented as hardware modules, software modules, or mixed modules. The subject matter of the invention is defined by the following claims.

Claims

1. A configurable microphone array (100) comprising: a microphone array (200) with a plurality of microphone capsules (210, 310) and a configurable directional pattern; at least one sound signal direction detector (250, 350, 421d-423d) configured to detect a current direction of a dominant desired signal and generate a corresponding first direction signal (255, 355, DC1) which serves as a preferred direction for detecting the desired sound; at least one beamforming unit (200, 320, 421b, 422b, 423b) configured to receive output signals (230, 315) from the microphone array (310) and the first direction signal (255, 355, DC1) for directional control of the configurable directional pattern of the microphone array (200) in order to adapt the configurable directional pattern to the preferred direction for the desired sound based on the first directional signal (255, 355, DC1), a control signal detector (240, 30, 40) which receives output signals (230, 315) from the microphone array (310) and is configured to detect at least one predefined control sound signal and its reception direction in the output signals (230, 315) from the microphone array (200), and is configured to output at least one second direction signal (CTR, XS, S1x, S2x, S3x) corresponding to the direction of reception of the control sound signal, wherein the control sound signal is a synthetically generated tone sequence and not a speech signal and is generated by an electronic device placed between a source of interference (130) and the microphone array (200), wherein the second direction signal (CTR, XS, S1x, S2x, S3x) represents a direction of a disturbance source (130); and wherein the sound signal direction detector (250, 350, 421d-423d) is configured to exclude the second direction according to the second direction signal as a preferred direction for detecting the desired sound and to filter out the second direction according to the second direction signal from the reception range of the microphone array (200).

2. Configurable microphone array (100) according to claim 1, wherein a plurality of control sound signals are predefined, and wherein the control signal detector (240, 30, 40) is configured to compare the output signals (230, 315) of the microphone array (310) with the multiple control sound signals and to detect a respective control sound signal, and wherein the at least one second direction signal (CTR, XS, S1x, S2x, S3x) corresponds to the respective detected control sound signal.

3. Configurable microphone array (100) according to one of claims 1-2, wherein, prior to detecting the control sound signal, the sound signal direction detector (250, 350, 421d-423d) is configured such that a direction or an adjacent direction is excluded as a preferred direction of the beamforming unit according to the second direction signal, and wherein the first direction signal (CTR, XS, S1x, S2x, S3x) causes the exclusion to be lifted.

4. Configurable microphone array (100) according to claim 1, wherein the sound signal direction detector (350) uses the output signals (230, 315) of only a portion of the microphone capsules in the microphone array (310).

5. Configurable microphone array (100) according to claim 1 or 4, wherein the sound signal direction detector (350 ) uses a grid of test points (107) with fixed positions to detect a direction of a dominant sound source.

6. Configurable microphone array (100) according to claim 5, wherein individual test points (107) of the grid of test points can be defined as not being used for the search.

7. Configurable microphone array (100) according to one of claims 1-3, wherein the sound signal direction detector (421d, 422d, 423d) generates a directional characteristic with a fixed direction independent of the output signals (230, 315) of the microphone array (310).

8. Configurable microphone array (100) according to claim 7, wherein the beamforming unit (421b-423b) is a single-focus beamforming unit, and wherein the microphone array contains several single-focus beamforming units with different directional pattern, whose output signals are superimposed or added in a combiner (424) to generate the output signal (425) to be output.

9. Configurable microphone array (100) according to one of claims 1-8, wherein the control signal detector (240, 30, 40) is configured to detect (330, 430) at least one control sound signal in each of the output signals (230, 315) of the microphone array (310) supplied to it, to generate (355) a second directional signal (335) for each of the control sound signals, respectively and to correlate (340, 440) the second directional signals (335) with each other in order to determine the direction of reception, generating a second direction signal (355) in each case, and correlating (340, 440) the second direction signals (335) with each other in order to detect the direction of reception of the control sound signal.

10. Configurable microphone array (100) according to any one of claims 1-9, wherein the control signal detector (240, 30, 40) is configured to detect (330, 430) the at least one control sound signal in at least one of the output signals (230, 315) of the microphone array (310) supplied to it, and is further configured to correlate the output signals (230, 315) supplied to it with each other in order to detect the direction of reception of the sound signal.

11. Configurable microphone array (100) according to one of claims 1-10, wherein the control signal detector outputs at least one control signal (CS) to the beamforming unit (320, 420) in order to suppress output of the control sound signal as a desired signal.

12. A method for configuring a microphone array (100) for speech signals, wherein the microphone array includes an array (200) of a plurality of microphone capsules (210, 310), and wherein the method is performed automatically and includes the following steps: detecting a current direction of a dominant desired signal and generating a corresponding first direction signal (255, 355, DC1) which serves as a preferred direction for detecting the desired sound, combining output signals from the microphone capsules with each other by means of beamforming in dependence on the first direction signal, wherein a directional pattern is created and wherein sound signals detected in accordance with the directional characteristic are output as desired signals at an output (325, 425) of the microphone array; automatically searching (610) within a spatial search area with a control signal detector for at least one predefined control sound signal and its direction of reception, and outputting at least one second direction signal (CTR, XS, S1x, S2x, S3x) that corresponds to the direction of reception of the control sound signal, wherein the control sound signal is a synthetically generated tone sequence and not a speech signal, and is generated by an electronic device placed between a disturbance source (130) and the microphone array (200), wherein the second directional signal (CTR, XS, S1x, S2x, S3x) represents a direction of a disturbance source (130); and automatically configuring (650) the microphone array by excluding the second direction according to the second direction signal as a preferred direction and by suppressing the second direction according to the second direction signal from the reception range of the microphone array (200).