A pickup circuit and a robot

Abstract

The application provides a pickup circuit and a robot, comprising collecting a sound signal through a voice collection module, simultaneously collecting an audio signal specially for a noise component of the robot through a feedforward noise collection module, or collecting and obtaining a noise signal in combination with a motion state of the robot, the existence of the noise signal may cause the sound signal collected by the voice collection module to be distorted, and then the signal processing module may misjudge in the noise removal process, and the accuracy of the target voice signal is affected. The feedforward noise collection module can accurately obtain the noise component, and the signal processing module can remove the noise, which can effectively reduce the misjudgment risk caused by noise interference, and can maximize the removal of the noise generated by the robot itself in the sound signal, such as motor running sound, mechanical component friction sound and the like, so that a clearer and purer target voice signal is obtained, the quality of the voice signal is greatly improved, and the stability and reliability of the whole voice processing system are ensured.

Description

Technical Field

[0001] This application relates to the field of voice technology applications, and in particular to a sound pickup circuit and a robot. Background Technology

[0002] Voice pickup technology, as a key component of environmental perception and interaction, plays a vital role in fields such as human-computer interaction and environmental monitoring. With the ever-increasing demands for voice quality, achieving clear and reliable voice acquisition and enhancement in complex acoustic environments has become a pressing technical challenge.

[0003] It should be noted that the above introduction to the technical background is only for the purpose of providing a clear and complete explanation of the technical solutions of this application and facilitating understanding by those skilled in the art. It should not be assumed that these technical solutions are known to those skilled in the art simply because they have been described in the background section of this application. Utility Model Content

[0004] This application aims to at least partially address one of the technical problems in the related art.

[0005] Therefore, one objective of this application is to provide a sound pickup circuit for use in a robot, comprising a voice acquisition module, a feedforward noise acquisition module, and a signal processing module, wherein:

[0006] The voice acquisition module is used to acquire voice to obtain sound signals;

[0007] The feedforward noise acquisition module is used to acquire audio data from the noise components of the robot and / or acquire the motion state of the robot to obtain noise signals.

[0008] The signal processing module is connected to the speech acquisition module and the feedforward noise acquisition module, and is used to denoise the sound signal based on the noise signal to obtain the target speech signal.

[0009] According to one embodiment of the sound pickup circuit of this application, the noise component of the robot includes a vibration component and a robot motion component; the feedforward noise acquisition module includes at least one of an inertial sensor and a first microphone unit;

[0010] The inertial sensor is disposed on the robot's moving part and is used to collect the motion state of the robot's moving part;

[0011] The first microphone unit is disposed on the vibrating component and is used to collect the audio emitted by the vibrating component.

[0012] According to one embodiment of the sound pickup circuit of this application, the feedforward noise acquisition module further includes:

[0013] An inverting circuit, connected to the inertial sensor and the first microphone unit, is used to invert the output signal of the inertial sensor and the output signal of the first microphone unit to obtain two inverted signals.

[0014] A signal combining circuit, connected to the inverting circuit, is used to combine the two inverted signals to obtain the noise signal.

[0015] According to one embodiment of the sound pickup circuit of this application, the feedforward noise acquisition module further includes:

[0016] A weight allocation submodule is connected between the inverting circuit and the signal combining circuit, and is used to adjust the signal amplitude output by the inertial sensor and the first microphone unit.

[0017] According to one embodiment of the sound pickup circuit of this application, the feedforward noise acquisition module further includes:

[0018] A pose sensor, connected to the inertial sensor and the first microphone unit, is used to collect the robot's motion posture and synchronize the output signals of the inertial sensor and the first microphone unit according to the robot's motion posture.

[0019] According to one embodiment of the sound pickup circuit of this application, the voice acquisition module includes a second microphone unit, a first preprocessing circuit, an analog-to-digital converter, a second preprocessing circuit, a memory, and a control submodule, wherein:

[0020] The second microphone unit is used to acquire the raw sound signal;

[0021] The first preprocessing circuit, connected to the second microphone unit, is used to amplify and / or filter the original sound signal to obtain an output signal;

[0022] The analog-to-digital converter is connected to the first preprocessing circuit and is used to convert the output signal of the first preprocessing circuit into a digital signal by digital-to-analog conversion.

[0023] The second preprocessing circuit, connected to the analog-to-digital converter, is used to perform voice enhancement on the digital signal to obtain the sound signal.

[0024] According to one embodiment of the sound pickup circuit of this application, the voice acquisition module further includes:

[0025] A memory, connected to the second preprocessing circuit, is used to store the sound signal.

[0026] According to one embodiment of the present application, the first preprocessing circuit includes a power amplifier and a filter;

[0027] The power amplifier is connected to the second microphone unit and is used to amplify the original sound signal.

[0028] The filter is connected to the power amplifier and is used to filter the original sound signal to obtain the output signal.

[0029] According to one embodiment of the present application, the second preprocessing circuit includes a sound enhancement sub-circuit, a background noise suppression sub-circuit, and an echo cancellation sub-circuit.

[0030] The sound enhancement sub-circuit is connected to the analog-to-digital converter and is used to enhance the digital signal with voice.

[0031] The background noise suppression sub-circuit is connected to the sound enhancement sub-circuit and is used to remove background noise from the digital signal;

[0032] The echo cancellation sub-circuit is connected to the background noise suppression sub-circuit and is used to eliminate the echo of the digital signal to obtain the sound signal.

[0033] According to one embodiment of the sound pickup circuit of this application, the signal processing module includes a third preprocessing circuit, an analysis submodule, and a noise reduction submodule;

[0034] The third preprocessing circuit is connected to the voice acquisition module and the feedforward noise acquisition module, and is used to resample the sound signal and the noise signal.

[0035] The analysis submodule is connected to the third preprocessing circuit and is used to analyze the correlation between the sound signal and the noise signal, and to obtain the noise component from the sound signal using the correlation.

[0036] The noise reduction submodule is connected to the analysis submodule and is used to eliminate the noise component from the sound signal to obtain the target speech signal.

[0037] Another object of this application is to provide a robot that includes the sound pickup circuit provided in one embodiment of this application.

[0038] In this application, a voice acquisition module collects sound signals, while a feedforward noise acquisition module is used to specifically collect audio from the robot's noisy components, or noise signals are acquired by combining the robot's motion state with the acquisition. The presence of noise signals may distort the sound signals collected by the voice acquisition module, leading to misjudgments by the signal processing module during denoising and affecting the accuracy of the target speech signal. The feedforward noise acquisition module can accurately acquire noise components, and the signal processing module can denoise them, effectively reducing the risk of misjudgments due to noise interference. It can remove noise generated by the robot itself, such as motor operation noise and mechanical component friction noise, from the sound signal to the greatest extent possible, resulting in a clearer and purer target speech signal. This greatly improves the quality of the speech signal and ensures the stability and reliability of the entire voice processing system. Attached Figure Description

[0039] Figure 1 This is a schematic diagram of a pickup circuit provided in an embodiment of this application;

[0040] Figure 2 This is a schematic diagram of a feedforward noise acquisition module according to an embodiment of this application;

[0041] Figure 3 This is a schematic diagram of another feedforward noise acquisition module provided according to an embodiment of this application;

[0042] Figure 4 This is a schematic diagram of another feedforward noise acquisition module provided according to an embodiment of this application;

[0043] Figure 5 This is a schematic diagram of the structure of a voice acquisition module according to an embodiment of this application;

[0044] Figure 6 This is a schematic diagram of another voice acquisition module provided according to an embodiment of this application;

[0045] Figure 7 This is a schematic diagram of the structure of a first preprocessing circuit according to an embodiment of this application;

[0046] Figure 8 This is a schematic diagram of the structure of a second preprocessing circuit according to an embodiment of this application;

[0047] Figure 9 This is a schematic diagram of the structure of a signal processing module provided according to an embodiment of this application. Detailed Implementation

[0048] The embodiments of this application are described in detail below. Examples of these embodiments are shown in the accompanying drawings, wherein the same or similar reference numerals denote the same or similar elements or elements having the same or similar functions throughout. The embodiments described below with reference to the accompanying drawings are exemplary and intended to explain this application, and should not be construed as limiting this application.

[0049] Sound pickup technology plays a crucial role in many fields such as human-computer interaction and environmental monitoring. It is not only a bridge to achieve natural and smooth communication between humans and machines, but also an important means to accurately capture environmental sound information and provide decision-making basis for various intelligent systems.

[0050] In intelligent voice assistant scenarios, sound pickup technology can capture users' voice commands. For example, when a user says commands such as "play music," "check the weather," or "set an alarm," the sound pickup device needs to quickly and accurately convert the sound signal into an electrical signal and transmit it to the subsequent voice recognition system. Only when the sound pickup process is accurate enough can the voice recognition system correctly understand the user's intent, thereby triggering the corresponding operation and providing users with a convenient and efficient service experience.

[0051] In intelligent customer service scenarios, when customers communicate with customer service robots via telephone or online voice, the audio pickup device is required to clearly record every question asked by the customer. Even when the customer speaks quickly, has a heavy accent, or is in a noisy environment, it should be able to accurately capture key information. This not only helps to improve the efficiency of solving customer problems, but also enhances customer satisfaction with the service.

[0052] In the field of environmental monitoring, acoustic imaging technology is an important tool for understanding environmental changes. For example, in wildlife conservation, researchers can use acoustic imaging devices to set up monitoring points in natural environments such as forests and grasslands to record animal sounds over long periods. By analyzing these sounds, researchers can understand information such as animal species, numbers, activity patterns, and reproduction status, providing strong evidence for protecting wildlife habitats and developing scientific conservation strategies.

[0053] In urban environmental monitoring, sound pickup technology can be used to monitor traffic noise, industrial noise, and other sources. By installing sound pickup devices in different areas of the city, noise data can be collected in real time. Combined with big data analysis and artificial intelligence algorithms, the status of urban noise pollution can be accurately assessed and predicted. This helps urban planners to formulate reasonable traffic plans, industrial layouts, and noise control measures, thereby improving the living environment of urban residents.

[0054] However, in the real world, complex acoustic environments are ubiquitous, inevitably resulting in a large amount of noise components mixed in with the sound signals collected by audio pickup devices. These noise components come from a wide range of sources and have different characteristics, including both stable, continuous noise and non-steady-state, sudden noise. This noise interference directly leads to a significant decline in speech quality, making originally clear and fluent speech unclear and intermittent, distorting the timbre, pitch, rhythm, and other characteristics of the speech, and causing it to lose its original realism and naturalness.

[0055] Therefore, how to achieve clear and reliable speech acquisition and enhancement in complex acoustic environments has become a key issue that urgently needs to be addressed in the field of acoustic technology.

[0056] The pickup circuit and robot of the present application embodiment are described below with reference to the accompanying drawings.

[0057] Figure 1 This is a schematic diagram of a pickup circuit provided in an embodiment of this application.

[0058] like Figure 1 As shown, the pickup circuit 100 in this embodiment includes: a voice acquisition module 101, a feedforward noise acquisition module 102, and a signal processing module 103, wherein:

[0059] The voice acquisition module 101 is used to acquire voice to obtain sound signals. The feedforward noise acquisition module 102 is used to acquire audio from the robot's noisy components and / or acquire the robot's motion state to obtain noise signals, wherein the noisy components generate noise based on the robot's operation. The signal processing module 103, connected to the voice acquisition module 101 and the feedforward noise acquisition module 102, is used to denoise the sound signals based on the noise signals to obtain the target voice signal.

[0060] It should be noted that this target voice signal provides reliable input data for subsequent applications such as speech recognition and speech synthesis. For example, when a robot in a smart home system receives a clear target voice signal, it can accurately identify the user's intent and control the corresponding home appliances to perform operations. For instance, the target voice signal can quickly turn on the air conditioner and adjust it to a comfortable temperature, eliminating the need for the user to manually operate various remote controls or switches, thereby improving the convenience and comfort of daily life.

[0061] For example, when an unusual situation occurs in a user's home, such as doors or windows being illegally opened or a smoke alarm being triggered, the robot equipped with the smart security system will issue an alarm and send a notification to the user's terminal device. The user can interact with the smart security system via voice and inquire about the specific situation. After receiving the target voice signal, the smart security system can accurately understand the user's question and provide a corresponding answer, allowing the user to stay informed about the situation at home.

[0062] For example, the voice assistant robot equipped in a smart car can engage in natural and fluent conversations with the driver and passengers. Drivers and passengers can inquire about weather conditions, news, traffic conditions, and other information. After receiving the target voice signal, the voice assistant robot will retrieve the relevant information from the in-vehicle internet and accurately broadcast it.

[0063] Optionally, the robot's noise component includes a vibration component and a robot motion component; the feedforward noise acquisition module includes at least one of an inertial sensor and a first microphone unit. In some scenarios, the feedforward noise acquisition module is disposed in the robot's noise component, wherein the inertial sensor is disposed in the robot motion component for acquiring the motion state of the robot motion component; and the first microphone unit is disposed in the vibration component for acquiring the audio emitted by the vibration component.

[0064] In some scenarios, the feedforward noise acquisition module is placed near the moving parts, such as the head, chassis / hull area, etc. If the feedforward noise acquisition module is placed in the head, since the moving parts that generate noise are the rotating joints and linkage structures inside the head, the feedforward noise acquisition module collects the noise generated by the rotating joints and linkage structures by being placed on the vibration isolation bracket of the head. A robust electret microphone is selected as the first microphone unit to counteract the effects of external wind noise and collision impacts.

[0065] If the feedforward noise acquisition module is set in the chassis / hub area, since the moving parts that generate noise are the bearings in the chassis / hub area, the feedforward noise acquisition module collects the noise generated by the bearings through the waterproof and mud-proof device set in the chassis / hub area.

[0066] In some embodiments, the inertial sensor includes an accelerometer, a gyroscope, etc. The inertial sensor is installed on the robot's moving parts (such as motor housings, joint axes, etc.) to collect the motion state of the robot's moving parts, such as acceleration and angular velocity.

[0067] It should be noted that inertial sensors utilize the piezoelectric effect to reflect the dynamic changes of robot moving parts in three axes, such as vibration frequency and amplitude, and usually have high-frequency properties.

[0068] In some embodiments, the first microphone unit is positioned close to the surface of the vibrating component or an aerodynamic noise source (such as an air vent) to collect audio emitted by the vibrating component (the audio is based on air disturbances radiated by the vibrating component).

[0069] It should be noted that the first microphone unit converts sound wave pressure into an electrical signal to reflect the aerodynamic noise generated by air disturbance, which is usually low-frequency.

[0070] Optionally, Figure 2This is a schematic diagram of a feedforward noise acquisition module provided according to an embodiment of this application.

[0071] like Figure 2 As shown, the feedforward noise acquisition module 102 further includes an inverting circuit 1023 and a signal combining circuit 1024. The inverting circuit 1023 is connected to the inertial sensor 1021 and the first microphone unit 1022, and is used to invert the output signals of the inertial sensor 1021 and the first microphone unit 1022 to obtain two inverted signals. The signal combining circuit 1024 is connected to the inverting circuit 1023, and is used to combine the two inverted signals to obtain a noise signal.

[0072] As an example, the inverting circuit 1023 is based on an inverting amplifier, which makes the phase difference between the output signal of the inertial sensor 1021 and the output signal of the first microphone unit 1022 180°.

[0073] As an example, the signal combining circuit 1024 is set up based on an adder circuit (such as a resistor network adder), which combines two inverse signals to obtain a noise signal.

[0074] Optionally, Figure 3 This is a schematic diagram of another feedforward noise acquisition module provided according to an embodiment of this application.

[0075] like Figure 3 As shown, the feedforward noise acquisition module 102 also includes a weight allocation submodule 1025, which is connected between the inverting circuit 1023 and the signal merging circuit 1024, and is used to adjust the signal amplitude output by the inertial sensor 1021 and the first microphone unit 1022.

[0076] It should be noted that the output signal amplitude of the inertial sensor 1021 may be relatively small, while the output signal amplitude of the first microphone unit 1022 is relatively large. Appropriate weight adjustment is required through the weight allocation submodule 1025 to match the amplitudes of the two signals, so as to avoid one signal drowning out the other signal during synthesis.

[0077] As an example, the weight allocation submodule 1025 is set up based on a resistor voltage divider network. By selecting resistors with different resistance values ​​to form a voltage divider network, the signal amplitude output by the inertial sensor 1021 and the first microphone unit 1022 is adjusted.

[0078] As an example, the weight allocation submodule 1025 is based on an operational amplifier gain adjustment circuit. By configuring the output gain of the operational amplifier gain adjustment circuit, the signal amplitude output by the inertial sensor 1021 and the first microphone unit 1022 is adjusted.

[0079] For example, when the robot is stationary, the amplitude of the output signal of the inertial sensor 1021 can be reduced and the amplitude of the output signal of the first microphone unit 1022 can be increased; when the robot is in motion, the amplitude of the output signal of the inertial sensor 1021 can be increased and the amplitude of the output signal of the first microphone unit 1022 can be decreased.

[0080] Optionally, Figure 4 This is a schematic diagram of another feedforward noise acquisition module provided according to an embodiment of this application.

[0081] The feedforward noise acquisition module 102 also includes a pose sensor 1026, which is connected to an inertial sensor 1021 and a first microphone unit 1022. The pose sensor 1026 is used to acquire the robot's motion posture and synchronize the output signals of the inertial sensor 1021 and the first microphone unit 1022 according to the robot's motion posture. By synchronizing the timing, the accuracy of voice interaction and noise suppression in complex acoustic environments is improved.

[0082] It should be noted that the feedforward noise acquisition module can also be configured using application-specific integrated circuits (ASICs, which are integrated circuits designed and manufactured for specific user requirements and specific systems; in this embodiment, the integrated circuit represents a control module), intellectual property cores (IP cores are mature designs of circuit modules with independent functions in chip or integrated circuit designs; these circuit designs can be applied to other chip or integrated circuit design projects that include the circuit module, thereby reducing the design workload, shortening the design cycle, and improving the success rate of chip or integrated circuit design. IP cores are classified into three levels: behavioral, structural, and physical, thus corresponding to three types of IP cores: soft cores designed with hardware description languages, solid cores that complete structural descriptions, and hard cores based on physical descriptions and verified by processes), etc. The specific configuration forms will not be elaborated here. As long as the noise components of the robot can be audio-acquired and / or the motion state of the robot can be acquired to obtain noise signals, any configuration form of the feedforward noise acquisition module is applicable and is not limited to this embodiment.

[0083] Optionally, Figure 5 This is a schematic diagram of the structure of a voice acquisition module provided according to an embodiment of this application.

[0084] like Figure 5As shown, the voice acquisition module 101 includes a second microphone unit 1011, a first preprocessing circuit 1012, an analog-to-digital converter 1013, and a second preprocessing circuit 1014, wherein:

[0085] The second microphone unit 1011 is used to acquire the original sound signal. It should be noted that the second microphone unit 1011 can be composed of multiple microphones arranged in a specific geometric shape to form an array. Through the spatial distribution of multiple microphones, the time delay and amplitude differences of the original sound signal are captured. Based on the time delay and amplitude differences, the angle and distance of the sound source from the array are determined, thereby realizing the tracking of the original sound signal.

[0086] The first preprocessing circuit 1012 is connected to the second microphone unit 1011 and is used to amplify and / or filter the original sound signal to obtain an output signal. The analog-to-digital converter 1013 is connected to the first preprocessing circuit 1012 and is used to convert the output signal of the first preprocessing circuit 1012 into a digital signal.

[0087] The second preprocessing circuit 1014 is connected to the analog-to-digital converter 1013 and is used to enhance the digital signal to obtain a sound signal.

[0088] Exemplary illustration, Figure 7 This is a schematic diagram of the structure of a first preprocessing circuit according to an embodiment of this application, as shown below. Figure 7 As shown, the first preprocessing circuit 1012 includes a power amplifier and a filter. The power amplifier is connected to the second microphone unit 1011 and is used to amplify the original sound signal. The filter is connected to the power amplifier and is used to filter the original sound signal to obtain an output signal.

[0089] Exemplary illustration, Figure 8 This is a schematic diagram of a second preprocessing circuit according to an embodiment of this application, as shown below. Figure 8 As shown, the second preprocessing circuit 1014 includes a sound enhancement subcircuit, a background noise suppression subcircuit, and an echo cancellation subcircuit. The sound enhancement subcircuit is connected to the analog-to-digital converter 1013 and is used to enhance the digital signal for speech. For example, the sound enhancement subcircuit can achieve speech enhancement by filtering, noise reduction, and gain control of the digital signal. The background noise suppression subcircuit is connected to the sound enhancement subcircuit and is used to remove background noise from the digital signal. For example, the background noise suppression subcircuit can use Wiener filtering to remove background noise from the digital signal. The echo cancellation subcircuit is connected to the background noise suppression subcircuit and is used to eliminate the echo in the digital signal to obtain the sound signal. For example, the echo cancellation subcircuit can use adaptive filtering to eliminate the echo in the digital signal.

[0090] Optionally, Figure 6 This is a schematic diagram of another voice acquisition module provided according to an embodiment of this application.

[0091] like Figure 6 As shown, the voice acquisition module 101 also includes a memory 1015, which is connected to the second preprocessing circuit 1014 and is used to store sound signals.

[0092] For example, in a speech recognition scenario, the sound signals stored in memory 1015 can be used by the speech recognition engine for subsequent analysis, such as event playback and investigation. In an audio analysis scenario, the sound signals stored in memory 1015 can be used for audio analysis, editing, and synthesis.

[0093] It should be noted that the voice acquisition module 101 may also include a control submodule (not shown in the attached figures). This control submodule is connected to the first microphone unit, the first preprocessing circuit, the analog-to-digital converter, the second preprocessing circuit, and the memory. It is used to control the start and stop of the acquisition of the second microphone unit, adjust the gain and filtering parameters of the first preprocessing circuit, adjust the sampling frequency of the analog-to-digital converter, control the voice enhancement index of the second preprocessing circuit, and control the transmission mode of the memory.

[0094] It should be added that the voice acquisition module can also be configured using dedicated integrated circuits, IP cores, etc. The specific configuration methods will not be described in detail here. As long as voice can be acquired and sound signals can be obtained, any configuration method of the voice acquisition module is applicable and is not limited to this embodiment.

[0095] Optionally, Figure 9 This is a schematic diagram of the structure of a signal processing module according to an embodiment of this application.

[0096] like Figure 9 As shown, the signal processing module 103 includes a third preprocessing circuit 1031, an analysis submodule 1032, and a noise reduction submodule 1033, wherein:

[0097] The third preprocessing circuit 1031 is connected to the voice acquisition module 101 and the feedforward noise acquisition module 102, and is used to resample the sound signal and noise signal. It should be noted that resampling is used to adjust the number of sound signal and noise signal samples acquired per unit time and to match the sampling rate of the sound signal and noise signal, which facilitates subsequent signal processing.

[0098] The analysis submodule 1032 is connected to the third preprocessing circuit 1031 and is used to analyze the correlation between the sound signal and the noise signal, and to obtain the noise component from the sound signal using the correlation. It should be noted that the correlation is used to measure the similarity between the sound signal and the noise signal in the frequency domain; based on the similarity, the sound signal and the noise signal are weighted to suppress speech components weakly correlated with noise and retain noise components strongly correlated with noise; adaptive filtering tracks noise changes and obtains the noise component from the mixed signal of the sound signal and the noise signal.

[0099] The noise reduction submodule 1033 is connected to the analysis submodule 1032 and is used to remove noise components from the sound signal to obtain the target speech signal. It should be noted that the power spectrum of the target speech signal is obtained by subtracting the power spectrum of the noise component from the power spectrum of the sound signal and the noise signal; the target speech signal is then obtained by reconstructing the phase information from the power spectrum of the target speech signal.

[0100] It should be added that the signal processing module can also be configured using a microcontroller unit (MCU), a digital signal processor (DSP), an application-specific integrated circuit (ASIC), an IP core, etc. The specific configuration will not be described in detail here. As long as the sound signal can be denoised based on the noise signal and the target speech signal can be obtained, any configuration of the signal processing module is applicable and is not limited to this embodiment.

[0101] Specifically, according to embodiments of this application, the aforementioned sound pickup circuit can be implemented as a robot. The voice acquisition module is responsible for the robot's voice acquisition actions, ensuring the robot can accurately capture environmental sound information. The feedforward noise acquisition module is responsible for the audio acquisition actions and / or motion state acquisition actions of the robot's noise components, thereby accurately acquiring noise components. The signal processing module is responsible for denoising the sound signal based on the noise components to obtain the desired target speech signal, providing reliable input data for subsequent applications such as speech recognition and speech synthesis.

[0102] As illustrated, the sound pickup circuit and robot provided in this application collect sound signals through a voice acquisition module, while simultaneously utilizing a feedforward noise acquisition module specifically for audio acquisition of the robot's noisy components, or combining noise signal acquisition with the robot's motion state. The presence of noise signals may distort the sound signals acquired by the voice acquisition module, leading to misjudgments by the signal processing module during denoising, thus affecting the accuracy of the target speech signal. The feedforward noise acquisition module can accurately acquire noise components, and the signal processing module can denoise them, effectively reducing the risk of misjudgments caused by noise interference. It can remove noise generated by the robot itself, such as motor operation noise and mechanical component friction noise, from the sound signal to the greatest extent possible, thereby obtaining a clearer and purer target speech signal, greatly improving the quality of the speech signal, and ensuring the stability and reliability of the entire voice processing system.

[0103] In the description of this application, it should be understood that the terms "center", "longitudinal", "lateral", "length", "width", "thickness", "upper", "lower", "front", "rear", "left", "right", "vertical", "horizontal", "top", "bottom", "inner", "outer", "clockwise", "counterclockwise", "axial", "radial", "circumferential", etc., indicating the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings, are only for the convenience of describing this application and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation, and therefore should not be construed as a limitation of this application.

[0104] Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of that feature. In the description of this application, "multiple" means two or more, unless otherwise explicitly specified.

[0105] In this application, unless otherwise expressly specified and limited, the terms "installation," "connection," "linking," and "fixing," etc., should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral part; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium; they can refer to the internal communication of two components or the interaction between two components. Those skilled in the art can understand the specific meaning of the above terms in this application according to the specific circumstances.

[0106] In this application, unless otherwise expressly specified and limited, "above" or "below" the second feature can mean that the first feature is in direct contact with the second feature, or that the first feature is in indirect contact with the second feature through an intermediate medium. Furthermore, "above," "on top of," and "over" the second feature can mean that the first feature is directly above or diagonally above the second feature, or simply that the first feature is at a higher horizontal level than the second feature. "Below," "below," and "under" the second feature can mean that the first feature is directly below or diagonally below the second feature, or simply that the first feature is at a lower horizontal level than the second feature.

[0107] In the description of this specification, the references to terms such as "one embodiment," "some embodiments," "example," "specific example," or "some examples," etc., refer to specific features, structures, materials, or characteristics described in connection with that embodiment or example, which are included in at least one embodiment or example of this application. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples. Moreover, without contradiction, those skilled in the art can combine and integrate the different embodiments or examples described in this specification, as well as the features of different embodiments or examples.

[0108] Although embodiments of this application have been shown and described above, it is understood that the above embodiments are exemplary and should not be construed as limiting this application. Those skilled in the art can make changes, modifications, substitutions and variations to the above embodiments within the scope of this application.

Claims

1. A pickup circuit, characterized in that, The sound pickup circuit is applied to the robot and includes a voice acquisition module, a feedforward noise acquisition module, and a signal processing module, wherein: The voice acquisition module is used to acquire voice to obtain sound signals; The feedforward noise acquisition module is used to acquire audio data from the noise components of the robot and / or acquire the motion state of the robot to obtain noise signals. The signal processing module is connected to the speech acquisition module and the feedforward noise acquisition module, and is used to denoise the sound signal based on the noise signal to obtain the target speech signal.

2. The pickup circuit according to claim 1, characterized in that, The robot's noise components include vibration components and robot motion components; the feedforward noise acquisition module includes at least one of an inertial sensor and a first microphone unit; The inertial sensor is disposed on the robot's moving part and is used to collect the motion state of the robot's moving part; The first microphone unit is disposed on the vibrating component and is used to collect the audio emitted by the vibrating component.

3. The pickup circuit according to claim 2, characterized in that, The feedforward noise acquisition module further includes: An inverting circuit, connected to the inertial sensor and the first microphone unit, is used to invert the output signal of the inertial sensor and the output signal of the first microphone unit to obtain two inverted signals. A signal combining circuit, connected to the inverting circuit, is used to combine the two inverted signals to obtain the noise signal.

4. The pickup circuit according to claim 3, characterized in that, The feedforward noise acquisition module further includes: A weight allocation submodule is connected between the inverting circuit and the signal combining circuit, and is used to adjust the signal amplitude output by the inertial sensor and the first microphone unit.

5. The pickup circuit according to claim 3, characterized in that, The feedforward noise acquisition module further includes: A pose sensor, connected to the inertial sensor and the first microphone unit, is used to collect the robot's motion posture and synchronize the output signals of the inertial sensor and the first microphone unit according to the robot's motion posture.

6. The pickup circuit according to claim 1, characterized in that, The voice acquisition module includes a second microphone unit, a first preprocessing circuit, an analog-to-digital converter, and a second preprocessing circuit, wherein: The second microphone unit is used to acquire the raw sound signal; The first preprocessing circuit, connected to the second microphone unit, is used to amplify and / or filter the original sound signal to obtain an output signal; The analog-to-digital converter is connected to the first preprocessing circuit and is used to convert the output signal of the first preprocessing circuit into a digital signal by digital-to-analog conversion. The second preprocessing circuit, connected to the analog-to-digital converter, is used to perform voice enhancement on the digital signal to obtain the sound signal.

7. The pickup circuit according to claim 6, characterized in that, The voice acquisition module also includes: A memory, connected to the second preprocessing circuit, is used to store the sound signal.

8. The pickup circuit according to claim 6, characterized in that, The first preprocessing circuit includes a power amplifier and a filter; The power amplifier is connected to the second microphone unit and is used to amplify the original sound signal. The filter is connected to the power amplifier and is used to filter the original sound signal to obtain the output signal.

9. The pickup circuit according to claim 6, characterized in that, The second preprocessing circuit includes a sound enhancement sub-circuit, a background noise suppression sub-circuit, and an echo cancellation sub-circuit; The sound enhancement sub-circuit is connected to the analog-to-digital converter and is used to enhance the digital signal with voice. The background noise suppression sub-circuit is connected to the sound enhancement sub-circuit and is used to remove background noise from the digital signal; The echo cancellation sub-circuit is connected to the background noise suppression sub-circuit and is used to eliminate the echo of the digital signal to obtain the sound signal.

10. The pickup circuit according to claim 1, characterized in that, The signal processing module includes a third preprocessing circuit, an analysis submodule, and a noise reduction submodule; The third preprocessing circuit is connected to the voice acquisition module and the feedforward noise acquisition module, and is used to resample the sound signal and the noise signal. The analysis submodule is connected to the third preprocessing circuit and is used to analyze the correlation between the sound signal and the noise signal, and to obtain the noise component from the sound signal using the correlation. The noise reduction submodule is connected to the analysis submodule and is used to eliminate the noise component from the sound signal to obtain the target speech signal.

11. A robot, characterized in that, The robot includes a pickup circuit as described in any one of claims 1-10.

Images