TW201810253A - Device and method for activating a sensor apparatus - Google Patents

Abstract

The invention relates to an activation device (1) for activating an external sensor apparatus (12), having a multiplicity of sound converter apparatuses (10-i; 10a; 10b) which have a respective resonant frequency and a respective sensitivity and which are designed to pick up a sound signal, convert the picked-up sound signal into an electrical signal and output the electrical signal, wherein the resonant frequencies and/or sensitivity levels of the sound converter apparatuses (10-i; 10a; 10b) differ at least partially from one another; and an evaluation apparatus (11) which is designed to detect a predefined sound pattern by evaluating at least one of the output electrical signals, and in this case to output an activation signal to the external sensor apparatus (12) in order to activate the external sensor apparatus (12).

Description

Apparatus and method for actuating a sensor device

The present invention relates to an actuating device for actuating an external sensor device, and a method for actuating a sensor device.

For example, an automatic speech recognition method, such as that known from DE 10 2006 039 126 A1, is highly popular because it allows for natural interaction between the user and the device. The user's hand is idle, and thus the user, for example, on the road, is not distracted by events that must perform manual operator control operations.

However, such speech recognition methods have a large power consumption level during continuous operation, which is the result of the actual microphone and is also the result of a highly computationally intensive evaluation of the received sensor data.

The present invention achieves an actuating device for actuating an external sensor device having the features of claim 1 and a method for actuating a sensor device having the features of claim 8 of the patent application.

Accordingly, the present invention provides an actuating device for actuating an external sensor device, in particular a microphone device, having a plurality of sound transducer devices having respective resonant frequencies and sensitivities Degree level. Sound converter device designed to receive The sound signal converts the received sound signal into an electrical signal and outputs the electrical signal. The resonant frequency and/or sensitivity level of the sound transducer device is at least partially different from each other. Additionally, the actuation device includes an evaluation device designed to detect a predefined sound pattern by evaluating at least one of the output electrical signals, and outputting an actuation signal to the external if the evaluation device detects the sound pattern A sensor device activates the external sensor device.

Accordingly, the present invention also provides a method for actuating a sensor device, in particular a microphone device. In this case, in a first method step, a plurality of sound signals are received by means of a sound converter device, wherein the sound converter device has a respective resonance frequency and a sensitivity level, wherein the resonance frequency and/or the sensitivity level are at least partially Different from each other. In another method step, the acoustic signals are converted into individual electrical signals and output. A predefined sound pattern is detected by evaluating at least one of the output electrical signals. If a sound pattern is detected, the sensor device is then actuated.

The preferred embodiments are the subject matter of the respective dependent claims.

Advantages of the invention

The present invention achieves an actuation device of the sensor device that consumes only a small amount of energy by itself and that is more intensive in terms of consumption, such as a microphone device having a corresponding speech recognition device (only when necessary). The actuation device outputs an actuation signal when it detects a predefined sound pattern, wherein the sound pattern is understood to be a predefined sound signal having a predefined spectrum. For example, the sound pattern may be a keyword such as the word "actuation" that the user can actuate by means of the actuation device. If the sound converter devices have different resonant frequencies, they are particularly sensitive to different frequency ranges, as a result of which a certain spectrum can be monitored by combining sound converter devices. Sound converter devices with different sensitivity levels at the same resonant frequency will The same sound pattern reacts but responds with different intensities. As a result, it is possible to determine not only the frequency mode but also the amplitude thereof. When there is noisy background noise, the detection threshold of the sound pattern can be increased by selecting a sound converter device having a relatively low intensity. Here, sensitivity is understood to mean the ratio of the output signal (electrical signal) to the input signal (sound signal). Thus, when the same sound signal is present, sound transducer devices having different levels of sensitivity output electrical signals having different intensities, and thus react at different levels of sensitivity. For example, the sensitivity level can be specified by the ratio of the electrical signal voltage (in volts) per sound signal (in Pascal). For example, the actuation device can be configured in a space and can allow for the detection of the presence of a person in the space based on the sound pattern. When a sound pattern is detected, the sensor device is actuated, for example based on its correspondingly controlling a gas sensor of the air conditioning system for measuring the air quality of the space.

According to a preferred development of the actuating device, the evaluation device is designed to evaluate the electrical signals of the acoustic transducer devices whose resonant frequencies substantially correspond to the frequency peaks of the spectrum of the sound pattern. The keyword has a unique spectrum with a characteristic frequency peak. If the sound transducers whose resonant frequencies substantially correspond to the peaks of the frequencies detect an electrical signal that exceeds a predefined threshold, a sound pattern or keyword is detected and the sensor device is actuated.

According to a preferred development of the actuating device, the sound transducer device comprises a piezoelectric element which is arranged on a deflection element which is deflectable by a sound signal and which is designed to output an electrical signal which varies with the deflection of the deflection element. Such sound transducer devices, preferably microelectromechanical sound transducer devices, have only very low power consumption and can therefore remain actuated in a continuous mode. Since the piezoelectric element is used, an external voltage is not required to operate the sound transducer device, and as a result, only the power consumption of the device is evaluated. Various sound converters with the same resonant frequency but different sensitivity levels Devices (eg, different deflections of a piezoelectric element for a given sound level at the same resonant frequency), in one embodiment, the magnitude of the sound level at this frequency can also be compared without having to compare the resources by a comparator In the case of a reference signal, it is roughly determined. For example, different sensitivity levels having the same resonant frequency can be implemented by sound converter devices of different widths.

According to a preferred development of the actuation device, the evaluation device is designed to detect the sound pattern when the signal strength of the at least one evaluation electrical signal exceeds a respective predefined threshold. For different sound transducer devices (in other words, different resonant frequencies), the thresholds can be different and the threshold can be determined as the sound pattern changes.

According to a preferred development of the actuating device, the evaluation device is designed to detect the sound pattern when the time profile of the signal strength of the at least one evaluation electrical signal substantially corresponds to a predefined time feature. In this way, a specific sound pattern or verb can be reliably detected.

The present invention is also directed to a speech recognition device having a sensor device and an actuation device that receives sound, the actuation device being designed to actuate the sensor device when the actuation device detects a predefined sound pattern. According to a preferred development of the speech recognition device, the sound pattern can be defined by user input. Thus, the user can determine the sound pattern or the verb itself by means of text input, selection or voice input.

According to a preferred development of the method, the electrical signals of the sound transducer devices whose resonant frequencies substantially correspond to the frequency peaks of the spectrum of the sound pattern are evaluated.

According to a preferred development of the method, the sound pattern is detected when the signal strength of at least one of the evaluated electrical signals exceeds a respective predefined threshold.

1‧‧‧Acoustic device

2‧‧‧Voice recognition device

10a‧‧‧Sound converter equipment

10b‧‧‧Sound converter equipment

10-1‧‧‧Sound Converter Equipment

10-2‧‧‧Sound Converter Equipment

10-3‧‧‧Sound Converter Equipment

10-n‧‧‧Sound Converter Equipment

11‧‧‧Evaluation equipment

12‧‧‧External sensor equipment

20‧‧‧Substrate

21‧‧‧ deflection element

22‧‧‧Piezoelectric components

22a‧‧‧Piezoelectric components

22b‧‧‧Piezoelectric components

23‧‧‧Signal generating equipment

24‧‧‧ cavity

41‧‧‧Substrate

42‧‧‧Piezoelectric layer

43‧‧‧Contact electrode

51‧‧‧ lower electrode

52‧‧‧Intermediate electrode

53‧‧‧Upper electrode

54‧‧‧Second piezoelectric layer

55‧‧‧Upper

71‧‧‧ deflection element

91‧‧‧ comparator

92‧‧‧ pattern recognition equipment

93‧‧‧Microphone

94‧‧‧Signal Processing Unit

95‧‧‧ Analog/Digital Converter

96‧‧‧Digital Signal Processing Unit

97‧‧‧Voice recognition module

A-A‧‧‧Axis

B-B‧‧‧Axis

y‧‧‧Vertical direction

z‧‧‧Vertical direction

S1‧‧‧ first method steps

S2‧‧‧ second method step

S3‧‧‧ third method steps

S4‧‧‧Fourth method steps

In the drawings: FIG. 1 shows a schematic circuit block diagram of an actuator according to a specific example of the present invention; and FIG. 2 shows a schematic plan view and a cross-sectional view of a sound converter device according to an embodiment; 5 shows a schematic plan view and a cross-sectional view of a sound converter device according to another specific example; FIGS. 6 and 7 show schematic cross-sectional views of a piezoelectric element according to a specific example of the present invention; FIG. 8 shows an exemplary sound conversion FIG. 9 shows a circuit block diagram of a speech recognition apparatus according to an embodiment of the present invention; FIG. 10 shows a flowchart explaining a method for actuating a sensor apparatus according to an embodiment; and FIG. A schematic flow diagram illustrating a method for actuating a sensor device in accordance with another specific example is shown.

In the following, the same or structurally identical elements have the same reference symbols. Various specific examples may be combined with each other as needed.

1 shows a block diagram of an actuation device 1 for actuating an external sensor device 12 in accordance with an embodiment of the present invention. The sensor device 12 preferably includes a microphone device, but may also include, for example, a sensor device for sensing gas, temperature, humidity, or brightness. Actuating device 1 comprises a plurality of sound transducer devices 10-1 to 10-n having respective resonant frequencies and respective sensitivity levels. The actuating device 1 comprises at least two and preferably at least ten sound transducer devices 10-i. The sound converter device 10-i is designed to receive a sound signal and convert the received sound signal into The electrical signal is output and the electrical signal is output.

2 is a plan view, and an exemplary acoustic transducer device 10a is schematically illustrated in FIG. 3 in a cross-sectional view along the A-A axis.

The deflection element 21 is connected to the substrate 20 on one side or (specifically) formed therefrom by etching, and the other three sides and the bottom surface of the element are not connected to the substrate, with the result that a cavity 24 partially covered by the deflection element 21 is produced. The deflection element 21 is in the shape of a cantilever bracket and can be deflected in the vertical direction y. The resonant frequency and sensitivity level of the deflection element 21 can be set by suitably selecting the thickness, length and material of the deflection element 21.

The deflection element 21 has a piezoelectric element 22 integrated into or disposed on the deflection element 21. The piezoelectric element 22 is preferably disposed adjacent to the connection region of the deflection element 21 that is coupled to the substrate 20 because the deflection of the deflection element 21 is largely due to deflection in the connection region. The deflection element 21 is correspondingly deflected as a function of the spectral change of the sound signal or sound impinging on the actuating device 1. The maximum deflection occurs at the resonant frequency of the deflection element 21. The piezoelectric element 22 is connected to a signal generating device 23 which outputs an electrical signal that varies with the deflection of the deflection element 21.

4 illustrates a plan view of the sound converter device 10b, and FIG. 5 illustrates a schematic cross-sectional view along the B-B axis according to another specific example. Here, the opposite sides of the deflection element 21 are connected to the substrate 20 and the remaining two opposite sides are not attached, resulting in a bridge element that can be deflected by the sound in the vertical direction z. The corresponding piezoelectric elements 22 connected to the signal generating device 23 are each disposed in the vicinity of the connection region of the deflection element 71 connected to the substrate 20.

Figure 6 illustrates an exemplary layered structure of piezoelectric element 22a having upper contact electrode 43 on piezoelectric layer 42, wherein piezoelectric layer 42 is preferably comprised of a material such as aluminum nitride AlN or PZT ceramic. The piezoelectric layer 42 is disposed on the substrate 41 preferably made of tantalum, and can be deflected by itself Piece 21 is formed and acts as a lower contact electrode. In order to achieve a resonance frequency in the frequency range of the speech signal, the thickness of the piezoelectric layer 42 is preferably selected to be less than 1 μm, and the thickness of the contact electrode 43 is preferably selected to be less than 100 nm.

According to another embodiment, the two contact electrodes can be disposed on the piezoelectric layer 42, and the electrodes can be disposed over a passivation layer (eg, made of SiO ₂ ) on the deflection element 21 .

Fig. 7 illustrates a piezoelectric element 22b according to another specific example, in which an upper layer 55 made of a piezoelectric material is disposed between the upper electrode 53 and the intermediate electrode 52, and a second piezoelectric layer 54 is disposed between the lower electrode 51 and the middle Between the electrodes 52. This particular example is characterized by an improved output signal.

The sound converter device 10-i is coupled to an evaluation device 11 that is designed to evaluate the electrical signals output by the signal generating device 23 and to detect predefined sound patterns.

Figure 8 shows the unique spectrum in the case of three different sound converter devices 10-1, 10-2 and 10-3. The respective output sensitivity levels S of the sound converter devices 10-1, 10-2, and 10-3 are characterized by different resonant frequencies of 1 kHz, 1.5 kHz, and 2 kHz. The sound pattern has a typical spectrum with corresponding frequency peaks. The evaluation device preferably evaluates the electrical signals of the sound transducer devices 10-i whose resonant frequencies substantially correspond to the frequency peaks of the spectrum of the sound pattern. If the sound pattern has a frequency peak of, for example, 1 kHz and 2 kHz, the first sound converter device 10-1 and the third sound converter device 10-3 can be evaluated. If the signal strength of the evaluation electrical signal exceeds a predefined threshold (which may depend on the individual sound transducer device 11-i), the evaluation device 11 is designed to detect the sound pattern. The threshold is preferably adapted to the environmental noise level to prevent undesired sensor devices (specifically microphone devices) from being actuated, especially in noisy environments.

According to another specific example, the evaluation device is designed to evaluate telecommunications in at least one The time pattern of the signal strength of the number substantially corresponds to a predefined time feature, detecting the sound pattern.

According to another specific example, the evaluation device 11 detects the sound pattern when the signal strength ratio of the electrical signals of the various sound transducer devices is within a predefined range of values (depending on the sound pattern).

Figure 9 shows a schematic circuit block diagram of a speech recognition device 2 according to an embodiment of the invention having an actuating device 1, in particular an actuating device 1 according to one of the aforementioned specific examples. The evaluation device 11 of the actuating device 1 comprises a comparator 91 which compares the output voltages of the piezoelectric elements of the sound converter devices 10-1 to 10-n with respective voltage thresholds and transmits them to the corresponding signals over the corresponding signals A pattern recognition device 92 designed to recognize a predefined sound pattern. If the evaluation device 11 detects the sound pattern, the evaluation device 11 outputs an actuation signal to the microphone device 12. The microphone device 12 includes a microphone 93, and a signal processing unit 94 having an analog/digital converter 95, a digital signal processing unit 96, and a voice recognition module 97.

According to a specific example, the sound pattern can be defined by the user via user interface via the interface of the voice recognition device.

According to another specific example, the electrical signal of the sound converter device 10-i is output to the total contact pin (GPIO pin) of the evaluation device 11, wherein the evaluation device 11 outputs an actuation signal when the respective input voltages of the GPIO pins are exceeded.

Figure 10 shows a flow chart explaining a method for actuating the sensor device 12, wherein in a first method step S1, a plurality of sound signals are received by means of a sound converter device having respective resonance frequencies and respective sensitivity levels . The resonant frequency and/or sensitivity level of the sound transducer device is at least partially different from each other.

In a second method step S2, the sound signal is converted into a separate electrical signal and output.

In a further method step S3, the predefined sound pattern is detected by evaluating at least one of the output electrical signals.

If a sound pattern is detected, the sensor device is actuated in a fourth method step S4.

FIG. 11 illustrates a method for actuating a microphone device 12 in accordance with another embodiment. First, the microphone device 12 is placed in the standby mode or the sleep mode (S101). Subsequently, the external sound signal is detected in the above manner by means of the sound converter device 10-1, 10-2, 10-3 and the evaluation unit 11 to find the sound pattern (S102). Check if there is a sound pattern (S103). If this is not the case, the search is continued (S102), otherwise the microphone device 12 is actuated (S104). In a further method step S105, the speech recognition procedure can be carried out by means of the microphone device 12. Subsequently, it is checked whether the related sound event is still present (S106). If this is the case, the speech recognition process is continued, otherwise the microphone is placed in the standby mode again (S101).

Claims (10)

An actuating device (1) for actuating an external sensor device (12) having: a plurality of sound transducer devices (10-I; 10a; 10b) having a respective Resonating frequency and sensitivity, and designed to receive a sound signal, convert the received sound signal into an electrical signal and output the electrical signal, wherein the sound converter devices (10-i; 10a; 10b) The resonant frequencies and/or the sensitivity levels are at least partially different from each other; and an evaluation device (11) designed to detect a predefined sound pattern by evaluating at least one of the output electrical signals And in this case an output signal is output to the external sensor device (12) to actuate the external sensor device (12). The actuating device (1) of claim 1, wherein the evaluation device (11) is designed to evaluate the sound transducer devices whose resonant frequency substantially corresponds to a frequency peak of one of the sound patterns of the sound pattern ( 10-i; 10a; 10b) of the electrical signals. The actuating device (1) of claim 1 or 2, wherein the sound transducer device (10-i; 10a; 10b) comprises a piezoelectric element (22) disposed on the deflection element (21) The deflection elements (21) are deflectable by the acoustic signal, and the piezoelectric elements (22) are designed to output the electrical signal as a function of the deflection of the deflection elements (21). The actuating device (1) of claim 1 or 2, wherein the evaluation device (11) is designed such that when the signal strength of one of the at least one evaluation electrical signals exceeds a respective predefined threshold value, Detect this sound style. An actuating device (1) according to claim 1 or 2, wherein the evaluation device (11) is designed to substantially characterize a signal strength of one of the at least one evaluation electrical signals The sound pattern is detected when corresponding to a predefined time feature. A voice recognition device (2) having: a microphone device (12) for receiving sound; and an actuating device (1) according to any one of claims 1 to 5, The microphone device (12) is designed to actuate when the actuation device (1) detects a predefined sound pattern. The speech recognition device (2) of claim 6, wherein the sound pattern is defined by a user input. A method for actuating a sensor device (12) having the steps of receiving (S1) a plurality of sound signals by means of a sound converter device (10-i; 10a; 10b), the sound converters The device has respective resonant frequency and/or sensitivity levels that are at least partially different from each other; the sound signals are converted (S2) into respective electrical signals and outputted; and at least at least one of the output electrical signals is evaluated One detects (S3) a predefined sound pattern; if the sound pattern is detected, the sensor device (12) is actuated (S4). The method of claim 8, wherein the electrical signals of the sound transducer devices (10-i; 10a; 10b) whose resonant frequencies substantially correspond to frequency peaks of one of the sound patterns are evaluated. The method of claim 8 or claim 9, wherein the sound pattern is detected when a signal strength of one of the at least one evaluation electrical signals exceeds a respective predefined threshold.