Speech recognition system and speech recognition device

Abstract

The utility model discloses a voice recognition system and voice recognition equipment. The voice recognition system comprises a human body induction circuit, a power supply control circuit and a voice recognition circuit. The human body induction circuit is connected with the power supply control circuit and is used for outputting an induction signal to the power supply control circuit in response to existence of a human body in a preset induction area. The power supply control circuit is connected with the input power supply and the voice recognition circuit. The power supply control circuit is used for establishing connection between the input power supply and the voice recognition circuit when receiving the induction signal. The voice recognition circuit is used for operating when connected with an input power supply so as to recognize a voice signal. Through the mode, the power consumption of voice recognition can be reduced, and the endurance time of the battery is prolonged.

Description

TECHNICAL FIELD

[0001] The embodiments of the present application relate to the technical field of speech recognition, in particular to a speech recognition system and a speech recognition device. BACKGROUND

[0002] A speech recognition device is a device capable of receiving and processing human speech input and converting it into text or executing corresponding instructions. Such devices are widely used in various scenarios, such as smart homes, smartphones, car-mounted systems, personal assistants (such as Siri, etc.), and professional fields such as medical records, customer service, etc.

[0003] Current speech recognition devices are usually powered by batteries and always remain in working condition to recognize speech signals in real time, resulting in high power consumption of the speech recognition device and short battery life. CONTENT OF THE UTILITY MODEL

[0004] The embodiments of the present application provide a speech recognition system and a speech recognition device, which can reduce the power consumption of the speech recognition device and prolong the battery life.

[0005] In a first aspect, the embodiments of the present application provide a speech recognition system, comprising:

[0006] a human body sensing circuit, a power supply control circuit, and a speech recognition circuit;

[0007] The human body sensing circuit is connected to the power supply control circuit, and the human body sensing circuit is configured to output a sensing signal to the power supply control circuit when a human body exists in a preset sensing area;

[0008] The power supply control circuit is connected to an input power source and the speech recognition circuit, and the power supply control circuit is configured to establish a connection between the input power source and the speech recognition circuit when the sensing signal is received;

[0009] The speech recognition circuit is configured to run when connected to the input power source to recognize speech signals.

[0010] In one or more embodiments, the human body sensing circuit is further configured to delay for a first time period when the human body leaves the sensing area, and stop outputting the sensing signal when the first time period ends;

[0011] The power supply control circuit is further configured to disconnect the connection between the input power source and the speech recognition circuit when the sensing signal is not received;

[0012] The speech recognition circuit is further configured to stop running when not connected to the input power source.

[0013] In one or more embodiments, the voice recognition system further includes a data processing circuit, which is connected to the human body sensing circuit and the power supply control circuit respectively.

[0014] The human body sensing circuit is also used to output a sensing signal to the data processing circuit when a human body is present in the preset sensing area.

[0015] The power supply control circuit is also used to establish a connection between the input power supply and the data processing circuit when the sensing signal is received;

[0016] The data processing circuit is configured to operate when connected to the input power supply, and to output a control signal to the power supply control circuit when it is operating and receiving the sensing signal, so that the power supply control circuit maintains the connection between the input power supply and the voice recognition circuit, and to configure the human body sensing circuit to configure the sensing area when it is operating.

[0017] In one or more embodiments, the speech recognition system further includes:

[0018] A voltage conversion circuit is connected between the input power supply and the human body sensing circuit to convert the voltage of the input power supply into a first voltage to power the human body sensing circuit.

[0019] In one or more embodiments, the human body sensing circuit includes a radar module, a first resistor, a second resistor, a third resistor, and a fourth resistor;

[0020] The voltage input terminal of the radar module is connected to the input power supply, the output terminal of the radar module is connected to the power supply control circuit and the data processing circuit, and the first communication terminal and the second communication terminal of the radar module are both connected to the data processing circuit.

[0021] In one or more embodiments, the power supply control circuit includes:

[0022] The first switch branch is connected to the human body sensing circuit and the data processing circuit respectively, and is used to turn on when the sensing signal and / or the control signal is received, and to turn off when the sensing signal and the control signal are not received.

[0023] The second switch branch is connected to the first switch branch, the input power supply, the data processing circuit, and the voice recognition circuit, respectively. It is used to turn on when the first switch branch is turned on, so as to establish the connection between the input power supply and the data processing circuit and the voice recognition circuit, respectively, and to turn off when the first switch branch is turned off.

[0024] In one or more embodiments, the first switching branch includes a first diode, a second diode, a fifth resistor, a sixth resistor, and a first switching transistor;

[0025] The anode of the first diode is connected to the human body sensing circuit and the data processing circuit respectively. The cathode of the first diode is connected to the cathode of the second diode. The anode of the second diode is connected to the data processing circuit. The fifth resistor is connected between the cathode of the first diode and the first terminal of the first switching transistor. The sixth resistor is connected between the first terminal of the first switching transistor and ground. The second terminal of the first switching transistor is grounded. The third terminal of the first switching transistor is connected to the second switching branch.

[0026] In one or more embodiments, the second switching branch includes a seventh resistor, an eighth resistor, and a second switching transistor;

[0027] The seventh resistor is connected between the first switch branch and the first end of the second switch transistor, the eighth resistor is connected between the first and second ends of the second switch transistor, the second end of the second switch transistor is connected to the input power supply, and the third end of the second switch transistor is connected to the data processing circuit and the voice recognition circuit respectively.

[0028] In one or more embodiments, the voltage conversion circuit includes a first capacitor, a second capacitor, and a voltage regulator module;

[0029] The first capacitor is connected between the input power supply and ground. The voltage input terminal and the enable terminal of the voltage regulator module are both connected to the input power supply. The ground terminal of the voltage regulator module is grounded. The output terminal of the voltage regulator module is connected to the human body sensing circuit. The second capacitor is connected between the output terminal of the voltage regulator module and ground.

[0030] Secondly, embodiments of this application provide a voice recognition device, including the voice recognition system described above.

[0031] The beneficial effects of this application are as follows: The voice recognition system of this application includes a human body sensing circuit, a power supply control circuit, and a voice recognition circuit. The human body sensing circuit is connected to the power supply control circuit, which is connected to both the input power supply and the voice recognition circuit. In practical applications, the human body sensing circuit only outputs a sensing signal to the power supply control circuit when a human body enters the preset sensing area generated by the human body sensing circuit. Then, the power supply control circuit establishes a connection between the input power supply and the voice recognition circuit. Afterward, the voice recognition circuit operates to recognize voice signals. Therefore, for a voice recognition device whose input power is a battery and includes a voice recognition system, keeping the voice recognition circuit inactive before a human body enters the sensing area reduces the power consumption of the voice recognition device and thus extends battery life. Attached Figure Description

[0032] One or more embodiments are illustrated by way of example with reference to the accompanying drawings, which are not intended to limit the embodiments, and elements having the same reference numerals in the drawings are designated as similar elements.

[0033] Figure 1 This is a schematic diagram of the composition of the speech recognition system provided in the embodiments of this application. Figure 1 ;

[0034] Figure 2 This is a schematic diagram of the composition of the speech recognition system provided in the embodiments of this application. Figure 2 ;

[0035] Figure 3 This is a schematic diagram of the composition of the speech recognition system provided in the embodiments of this application. Figure 3 ;

[0036] Figure 4 This is a schematic diagram of the composition of the speech recognition system provided in the embodiments of this application. Figure 4 ;

[0037] Figure 5 Is with Figure 4 The circuit structure diagram corresponding to the block diagram shown is shown. Detailed Implementation

[0038] To make the objectives, technical solutions, and advantages of the embodiments of this application clearer, the technical solutions of the embodiments of this application will be clearly and thoroughly described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, not all embodiments. It should be understood that the specific embodiments described herein are only used to explain this application and are not intended to limit this application.

[0039] It should be noted that when an element is described as "connected" to another element, it can be directly connected to the other element, or there can be one or more intermediate elements between them.

[0040] Furthermore, the technical features involved in the various embodiments of this application described below can be combined with each other as long as they do not conflict with each other.

[0041] Please refer to Figure 1 , Figure 1 This is a schematic diagram illustrating the composition of a speech recognition system provided in an embodiment of this application. Figure 1 As shown, the voice recognition system 100 includes a human body sensing circuit 10, a power supply control circuit 20, and a voice recognition circuit 30.

[0042] The human body sensing circuit 10 is connected to the power supply control circuit 20. The human body sensing circuit 10 outputs a sensing signal to the power supply control circuit 20 when a human body is present in a preset sensing area. The power supply control circuit 20 is connected to both the input power supply VI N and the voice recognition circuit 30. The power supply control circuit 20 establishes a connection between the input power supply VI N and the voice recognition circuit 30 upon receiving the sensing signal. The voice recognition circuit 30 operates when connected to the input power supply VI N to recognize voice signals.

[0043] As can be seen, in practical applications, the human body sensing circuit 10 only outputs a sensing signal to the power supply control circuit 20 when a human body enters the preset sensing area generated by the human body sensing circuit 10. Then, the power supply control circuit 20 establishes a connection between the input power supply VI N and the voice recognition circuit 30. Afterward, the voice recognition circuit 30 is powered on and operates, and can recognize voice signals. Thus, for a voice recognition device whose input power supply VI N is a battery and includes the voice recognition system 100, the voice recognition circuit 30 can remain inactive until a human body enters the sensing area, thereby reducing the power consumption of the voice recognition device 30 and extending battery life.

[0044] In some embodiments, the input power source VI N is a battery. The battery can be a lithium-ion battery, lithium metal battery, lead-acid battery, nickel-cadmium battery, nickel-metal hydride battery, lithium-sulfur battery, lithium-air battery, or sodium-ion battery, etc., and is not limited thereto. In terms of scale, the battery in the embodiments of this application can be a single cell, a battery module composed of multiple cells connected in series and / or in parallel, or a battery pack composed of multiple battery modules connected in series and / or in parallel.

[0045] In some embodiments, the speech recognition circuit 30 includes a chip with speech recognition capabilities, which can process speech input and convert it into understandable text or instructions. The speech recognition chip typically integrates functions such as an analog front-end (AFE), digital signal processing (DSP), and a machine learning accelerator. In some embodiments, the speech recognition chip also has a speech activity detection (VAD) function, also known as sound detection or voice detection, a technique used to distinguish between speech and non-speech (such as background noise or silence). However, in related technologies, the VAD function is frequently triggered by ambient noise, leading to a series of subsequent actions, such as audio acquisition, audio noise reduction filtering, audio signal feature extraction, and algorithm recognition, which continuously consume power, resulting in high power consumption of the speech recognition device, including the speech recognition system 100. The voice recognition system 100 provided in this application embodiment can reduce or even eliminate the extra power consumption caused by the VAD function by only activating the voice recognition circuit 30 when a human body enters the sensing area. Therefore, compared with related technologies, this application can reduce the power consumption of the voice recognition device 30, which is beneficial to extending the battery life.

[0046] In some embodiments, the human body sensing circuit 10 is further configured to delay for a first duration when the human body leaves the sensing area, and to stop outputting the sensing signal at the end of the first duration; the power supply control circuit 20 is further configured to disconnect the connection between the input power supply VI N and the voice recognition circuit 30 when no sensing signal is received; the voice recognition circuit 30 is further configured to stop operating when not connected to the input power supply VI N.

[0047] The first duration is a preset duration, which can be set based on the actual application scenario, and this application embodiment does not impose specific limitations on it. Specifically, on the one hand, by setting the output of the sensing signal to stop only after the human body has left the sensing area for more than the first duration, the probability of false judgment can be reduced, that is, the user can be accurately determined to have left the sensing area; on the other hand, after determining that the human body has left the sensing area, the power supply control circuit 20 disconnects the connection between the input power supply VI N and the voice recognition circuit 30, so that the voice recognition circuit 30 stops operating, which can reduce the power consumption of the voice recognition circuit 30 and help extend the battery life.

[0048] In some embodiments, such as Figure 2 As shown, the voice recognition system 100 also includes a data processing circuit 40, which is connected to the human body sensing circuit 10 and the power supply control circuit 20 respectively.

[0049] The human body sensing circuit 10 is further configured to output a sensing signal to the data processing circuit 40 when a human body is present in the preset sensing area (i.e., the sensing signal output by the human body sensing circuit 10 is simultaneously input to the power supply control circuit 10 and the data processing circuit 40). The power supply control circuit 10 is also configured to establish a connection between the input power supply VI N and the data processing circuit 40 upon receiving the sensing signal. The data processing circuit 40 is configured to operate when connected to the input power supply VI N, and to output a control signal to the power supply control circuit 20 when it is operating and receiving the sensing signal, so that the power supply control circuit 20 maintains the connection between the input power supply VI N and the voice recognition circuit 30, and to configure the human body sensing circuit 10 to configure the sensing area during its operation.

[0050] Specifically, when a human body enters the sensing area, the human body sensing circuit 10 outputs a sensing signal. On one hand, the sensing signal is input to the power supply control circuit 10, and then the input power supply VI N can be powered by the power supply control circuit 10 to supply power to the data processing circuit 40 and the voice recognition circuit 30 respectively, and both the data processing circuit 40 and the voice recognition circuit 30 operate. On the other hand, the sensing signal is input to the data processing circuit 40, and after the data processing circuit 40 operates, the data processing circuit 40 outputs a control signal to the power supply control circuit 20 so that the power supply control circuit 20 maintains the connection between the input power supply VI N and the voice recognition circuit 30. Therefore, in application scenarios where the human body has left the sensing area and the human body sensing circuit 10 has stopped outputting sensing signals, but the process of the voice recognition circuit 30 recognizing the voice signal has not yet been completed (it can be understood that it takes a certain amount of time from the time the voice recognition circuit 30 receives the voice signal to the time it completes the voice signal recognition process; if the human body leaves the sensing area for more than a first time period during the voice signal recognition process, this is the aforementioned application scenario), if the data processing circuit 40 is not configured to output a control signal to the power supply control circuit 20, then because the human body sensing circuit 10 has stopped outputting sensing signals, the connection between the input power supply VIN and the voice recognition circuit 30 is disconnected, and the voice recognition circuit 30 stops operating due to power loss. Thus, the voice recognition circuit 30 cannot continue the process of recognizing the voice signal, resulting in an abnormality in the voice recognition function. Therefore, this embodiment of the application sets the data processing circuit 40 to output a control signal to the power supply control circuit 20, thereby maintaining the connection between the input power supply VIN and the voice recognition circuit 30. This ensures that the voice recognition circuit 30 remains powered and can continue the process of recognizing the voice signal. After the voice recognition circuit 30 completes the process of recognizing the voice signal, the data processing circuit 40 stops outputting control signals, which causes the power supply control circuit 20 to disconnect the connection between the input power supply VI N and the voice recognition circuit 30, so that the voice recognition circuit 30 stops operating. This reduces the power consumption of the voice recognition circuit 30 and helps to extend the battery life.

[0051] In some embodiments, the data processing circuit 40 performs at least logical operations, algorithm implementation, and wireless communication (such as Wi-Fi communication). In some embodiments, the data processing circuit 40 can receive the output result of the speech signal recognized by the speech recognition circuit 30, and can transmit the output result of the speech signal recognized by the speech recognition circuit 30 to the cloud by communicating with the cloud.

[0052] In some embodiments, such as Figure 3 As shown, the voice recognition system 100 also includes a voltage conversion circuit 50.

[0053] The voltage conversion circuit 50 is connected between the input power supply VI N and the human body sensing circuit 10. The voltage conversion circuit 50 is used to convert the voltage of the input power supply VI N into a first voltage to power the human body sensing circuit 10. The first voltage is a voltage suitable for powering the human body sensing circuit 10, and it can be set based on the actual application scenario. This application embodiment does not impose specific limitations on this.

[0054] In some embodiments, such as Figure 4 As shown, the power supply control circuit 20 includes a first switch branch 21 and a second switch branch 22.

[0055] The first switch branch 21 is connected to the human body sensing circuit 10 and the data processing circuit 40, respectively. The first switch branch 21 is used to turn on when a sensing signal and / or control signal is received, and to turn off when no sensing signal or control signal is received. The second switch branch 22 is connected to the first switch branch 21, the input power supply VI N, the data processing circuit 40, and the voice recognition circuit 30, respectively. The second switch branch 22 is used to turn on when the first switch branch 21 is turned on, so as to establish the connection between the input power supply VI N and the data processing circuit 40 and the voice recognition circuit 30, respectively, and to turn off when the first switch branch 21 is turned off.

[0056] Specifically, when the first switch branch 21 receives either the induction signal or the control signal, the first switch branch 21 is turned on. Subsequently, the second switch branch 22 is turned on. The input power supply VI N supplies power to the data processing circuit 40 and the voice recognition circuit 30 through the second switch branch 22, so that the data processing circuit 40 and the voice recognition circuit 30 can operate.

[0057] When the first switch branch 21 receives neither a sensing signal nor a control signal, it turns off. Subsequently, the second switch branch 22 also turns off. The connection between the input power supply VI N and the data processing circuit 40 and the voice recognition circuit 30 is broken, and both the data processing circuit 40 and the voice recognition circuit 30 lose power and stop operating.

[0058] Please refer to Figure 5 , Figure 5 An example is shown with Figure 4 The structure shown corresponds to a circuit structure. For example... Figure 5 As shown, the human body sensing circuit 10 includes a radar module U1, a first resistor R1, a second resistor R2, a third resistor R3, and a fourth resistor R4.

[0059] The radar module U1 has a voltage input terminal (pin 1 of radar module U1) that receives a first voltage V1, an output terminal (pin 3 of radar module U1) that is connected to the power supply control circuit 20 and the data processing circuit 40, and a first communication terminal (pin 4 of radar module U1) and a second communication terminal (pin 5 of radar module U1) that are both connected to the data processing circuit 40.

[0060] Specifically, in this embodiment, a radar module of model MS58 is used as an example. In other embodiments, since there are different types of radar modules, the specific pin definitions may differ when using other types of radar modules, but the functions and signal definitions are the same. Therefore, if other types of radar modules are used, they can be configured in a similar manner to the above embodiment, which is readily understood by those skilled in the art and will not be elaborated further here.

[0061] When a human body enters the sensing area, the output terminal of radar module U1 outputs a sensing signal. Data processing circuit 40 configures the parameters of radar module U1 through the first and second communication terminals to configure the size of the sensing area. The first resistor R1 and the second resistor R2 are current-limiting resistors, and the third resistor R3 and the fourth resistor R4 are pull-up resistors.

[0062] In this embodiment, the first switching branch 21 includes a first diode D1, a second diode D2, a fifth resistor R5, a sixth resistor R6, and a first switching transistor Q1.

[0063] In this circuit, the anode of the first diode D1 is connected to the human body sensing circuit 10 and the data processing circuit 40, the cathode of the first diode D1 is connected to the cathode of the second diode D2, the anode of the second diode D2 is connected to the data processing circuit 40, the fifth resistor R5 is connected between the cathode of the first diode D1 and the first terminal of the first switching transistor Q1, the sixth resistor R6 is connected between the first terminal of the first switching transistor Q1 and ground GND, the second terminal of the first switching transistor Q1 is grounded GND, and the third terminal of the first switching transistor Q1 is connected to the second switching branch 22.

[0064] Specifically, the first diode D1 and the second diode D2 are used to prevent reverse current flow. The fifth resistor R5 and the sixth resistor R6 divide the voltage of the sensing signal or control signal. The voltage division of the sensing signal or control signal across the sixth resistor R6 drives the first switch Q1 to conduct (corresponding to the conduction of the first switch branch 21). The sixth resistor R6 also serves to discharge the discharge when the first switch Q1 is turned off (corresponding to the turn-off of the first switch branch 21), ensuring that the first switch Q1 is reliably turned off.

[0065] In this embodiment, the first switch Q1 is an NMOS transistor. The gate of the NMOS transistor is the first terminal of the first switch Q1, the source of the NMOS transistor is the second terminal of the first switch Q1, and the drain of the NMOS transistor is the third terminal of the first switch Q1.

[0066] In addition, the first switch Q1 can be any controllable switch, such as an insulated gate bipolar transistor (IGBT) device, an integrated gate commutated thyristor (IGCT) device, a gate turn-off thyristor (GTO) device, a silicon controlled rectifier (SCR) device, a junction gate field-effect transistor (JFET) device, a MOS controlled thyristor (MCT) device, etc.

[0067] In this embodiment, the second switch branch 22 includes a seventh resistor R7, an eighth resistor R8, and a second switch Q2.

[0068] Among them, the seventh resistor R7 is connected between the first switch branch 21 and the first terminal of the second switch transistor Q2, the eighth resistor R8 is connected between the first and second terminals of the second switch transistor Q2, the second terminal of the second switch transistor Q2 is connected to the input power supply V1N, and the third terminal of the second switch transistor Q2 is connected to the data processing circuit 40 and the voice recognition circuit 30 respectively.

[0069] Specifically, when the first switch Q1 is turned on, the seventh resistor R7 and the eighth resistor R8 divide the voltage of the input power supply VI N. The voltage division of the input power supply VI N across the eighth resistor R8 drives the second switch Q2 to turn on (corresponding to the second switch branch 22 being turned on). The eighth resistor R8 also serves to discharge the discharge when the second switch Q2 is turned off (corresponding to the second switch branch 22 being turned off), ensuring that the second switch Q2 is reliably turned off.

[0070] In this embodiment, the second switch Q2 is a PMOS transistor. The gate of the PMOS transistor is the first terminal of the second switch Q2, the source of the PMOS transistor is the second terminal of the second switch Q2, and the drain of the PMOS transistor is the third terminal of the second switch Q2.

[0071] In addition, the second switch Q2 can be any controllable switch, such as an insulated gate bipolar transistor (IGBT) device, an integrated gate commutated thyristor (IGCT) device, a gate turn-off thyristor (GTO) device, a silicon controlled rectifier (SCR) device, a junction gate field-effect transistor (JFET) device, a MOS controlled thyristor (MCT) device, etc.

[0072] In this embodiment, the voltage conversion circuit 50 includes a first capacitor C1, a second capacitor C2, and a voltage regulator module U2.

[0073] The first capacitor C1 is connected between the input power supply VI N and ground GND. The voltage input terminal (i.e., pin 1 of the voltage regulator module U2) and the enable terminal (i.e., pin 3 of the voltage regulator module U2) of the voltage regulator module U2 are both connected to the input power supply VI N. The ground terminal (i.e., pin 2 of the voltage regulator module U2) of the voltage regulator module U2 is grounded to GND. The output terminal (i.e., pin 5 of the voltage regulator module U2) of the voltage regulator module U2 is connected to the human body sensing circuit 10. The second capacitor C2 is connected between the output terminal of the voltage regulator module U2 and ground GND.

[0074] Specifically, the first capacitor C1 and the second capacitor C2 are used for filtering, and the voltage across the second capacitor C2 is the first voltage V1. In some embodiments, the voltage regulator module U2 is a low-dropout linear regulator (LDO) of model ME6230C33M5G. In different embodiments, since the voltage regulator module U2 has different types, the specific pin definitions may differ when using other types of voltage regulator module U2, but the functions and signal definitions are the same. If other types of voltage regulator module U2 are used, they can be set up in a similar manner to the above embodiments, which is easily understood by those skilled in the art and will not be elaborated here.

[0075] The following are Figure 5 The principle of the circuit structure shown will be explained again.

[0076] When a human body is not in the sensing area, only the input power supply VI N, after passing through the voltage conversion circuit 50, inputs the first voltage V1 to power the radar module U1. At this time, the main power-consuming devices are the radar module U1 and the voltage regulator module U2, and the operating current in actual applications is approximately 40μA.

[0077] When a human body enters the sensing area, the radar module U1 outputs a sensing signal. This signal is input to the first terminal of the first switching transistor Q1 via the first diode D1 and the fifth resistor R5, causing Q1 to conduct and Q2 to conduct as well. The input power supply VI N powers the voice recognition circuit 30 and the data processing circuit 40 via the second switching transistor Q2, enabling them to operate. On one hand, the voice recognition circuit 30 can recognize voice signals; on the other hand, the data processing circuit 40 outputs a control signal, which is input to the first terminal of the first switching transistor Q1 via the second diode D2 and the fifth resistor R5, keeping both Q1 and Q2 conducting. The voice recognition circuit 30 and the data processing circuit 40 can maintain stable operation.

[0078] When a human body leaves the sensing area, radar module U1 starts timing. After the timing reaches the first duration, radar module U1 stops outputting sensing signals. At this time, since data processing circuit 40 continues to output control signals, input power supply VI N can continue to supply power to voice recognition circuit 30, enabling voice recognition circuit 30 to complete the current voice recognition process. After voice recognition circuit 30 completes the current voice recognition process, data processing circuit 40 stops outputting control signals. Subsequently, both first switch Q1 and second switch Q2 are turned off, and the current in the entire system returns to approximately 40μA. In related technologies, since voice recognition circuit 30 remains running, the current of the entire system is approximately 80mA, approximately 2000 times that of the embodiment of this application. Therefore, compared with related technologies, the voice recognition system provided in this application has lower power consumption, which is beneficial for extending battery life.

[0079] This application also provides a speech recognition device. The speech recognition device includes the speech recognition system 100 in any embodiment of this application.

[0080] The above description is merely an embodiment of this application and does not limit the patent scope of this application. Any equivalent structural or procedural transformations made using the content of this application's specification and drawings, or direct or indirect applications in other related technical fields, are similarly included within the patent protection scope of this application.

[0081] The above embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them. Within the framework of this application, the technical features of the above embodiments or different embodiments can also be combined, and the steps can be implemented in any order. Those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of this application.

Claims

1. A speech recognition system, characterized by The voice recognition system comprises: a human body sensing circuit, a power supply control circuit and a voice recognition circuit; the human body sensing circuit is connected with the power supply control circuit, and is configured to output a sensing signal to the power supply control circuit when a human body exists in a preset sensing area; the power supply control circuit is connected with an input power supply and the voice recognition circuit, and is configured to establish a connection between the input power supply and the voice recognition circuit when the sensing signal is received; the voice recognition circuit is configured to run when connected with the input power supply, so as to recognize a voice signal.

2. The speech recognition system of claim 1, wherein, the human body sensing circuit is further configured to delay for a first time length when the human body leaves the sensing area, and stop outputting the sensing signal when the first time length ends; the power supply control circuit is further configured to disconnect the connection between the input power supply and the voice recognition circuit when the sensing signal is not received; the voice recognition circuit is further configured to stop running when not connected with the input power supply.

3. The speech recognition system of claim 1 or 2, wherein, The voice recognition system further comprises a data processing circuit, which is connected with the human body sensing circuit and the power supply control circuit; the human body sensing circuit is further configured to output a sensing signal to the data processing circuit when a human body exists in a preset sensing area; the power supply control circuit is further configured to establish a connection between the input power supply and the data processing circuit when the sensing signal is received; the data processing circuit is configured to run when connected with the input power supply, and output a control signal to the power supply control circuit when running and receiving the sensing signal, so as to make the power supply control circuit maintain the connection between the input power supply and the voice recognition circuit, and configure the human body sensing circuit when running, so as to configure the sensing area.

4. The speech recognition system of claim 1 or 2, wherein, The voice recognition system further comprises: a voltage conversion circuit connected between the input power supply and the human body sensing circuit, configured to convert the voltage of the input power supply into a first voltage, so as to supply power to the human body sensing circuit.

5. The speech recognition system of claim 3, wherein, The human body sensing circuit comprises a radar module; a voltage input end of the radar module is connected with the input power supply, an output end of the radar module is connected with the power supply control circuit and the data processing circuit, and a first communication end and a second communication end of the radar module are both connected with the data processing circuit.

6. The speech recognition system of claim 3, wherein, The power supply control circuit comprises: a first switch branch connected with the human body sensing circuit and the data processing circuit, configured to be turned on when the sensing signal and / or the control signal is received, and turned off when the sensing signal and the control signal is not received; a second switch branch connected with the first switch branch, the input power supply, the data processing circuit and the voice recognition circuit, configured to be turned on when the first switch branch is turned on, so as to establish a connection between the input power supply and the data processing circuit and the voice recognition circuit, and turned off when the first switch branch is turned off.

7. The speech recognition system of claim 6, wherein, The first switch branch comprises a first diode, a second diode, a fifth resistor, a sixth resistor and a first switch tube; An anode of the first diode is connected with the human body induction circuit and the data processing circuit respectively, a cathode of the first diode is connected with a cathode of the second diode, an anode of the second diode is connected with the data processing circuit, the fifth resistor is connected between the cathode of the first diode and a first end of the first switch tube, the sixth resistor is connected between the first end of the first switch tube and the ground, a second end of the first switch tube is grounded, and a third end of the first switch tube is connected with the second switch branch.

8. The speech recognition system of claim 6, wherein, The second switch branch comprises a seventh resistor, an eighth resistor and a second switch tube. The seventh resistor is connected between the first switch branch and a first end of the second switch tube, the eighth resistor is connected between the first end and a second end of the second switch tube, a second end of the second switch tube is connected with an input power supply, and a third end of the second switch tube is connected with the data processing circuit and the voice recognition circuit respectively.

9. The speech recognition system of claim 4, wherein, The voltage conversion circuit comprises a first capacitor, a second capacitor and a voltage stabilizing module. The first capacitor is connected between the input power supply and the ground, a voltage input end and an enable end of the voltage stabilizing module are connected with the input power supply, a grounding end of the voltage stabilizing module is grounded, an output end of the voltage stabilizing module is connected with the human body induction circuit, and the second capacitor is connected between the output end of the voltage stabilizing module and the ground.

10. A speech recognition device, characterized by The voice recognition system comprises the voice recognition system according to any one of claims 1-9.

Images