Sound control household appliance system, intelligent household equipment and method

Abstract

The invention discloses a sound control household appliance system, intelligent household equipment and a method. The sound-controlled household appliance system comprises a cloud end used for sending a corpus data packet to a smart home device, so that the smart home device receives the corpus data packet through a communication module in a voice integrated module, and configures and stores the corpus data packet through a voice processor; the voice processor is used for performing recognition through the corpus data packet according to the voice received by the audio circuit to obtain a first control instruction corresponding to the voice; the communication module is used for sending the first control instruction to the MCU; the MCU is used for executing a corresponding function according to the first control instruction; and the mobile terminal is used for determining whether to trigger an upgrading instruction according to the version number of the pre-acquired corpus data packet, or sending a second control instruction to the smart home equipment through the cloud, so that the smart home equipment executes a corresponding function according to the second control instruction. According to the invention, the technical effect of improving the configuration flexibility of the smart home equipment can be improved.

Description

Technical Field

[0001] The present invention relates to the application field of Internet of Things technology, and in particular to a voice-controlled home appliance system, smart home equipment and method. Background Art

[0002] With the development of Internet of Things technology and the popularization of intelligence, there are many categories of voice-controlled smart home appliances on the market, making home appliances move from ordinary to intelligent. Smart home appliances with voice functions have brought consumers a more convenient lifestyle.

[0003] However, in the field of product research and development of voice-controlled smart home appliances, the development method is basically to fully customize a single product, which generally has a long development cycle. If there are subsequent products (even products of the same category), it is difficult to continue to reuse them, especially in terms of software, which all need to be developed from scratch.

[0004] At present, most voice-controlled smart home appliances are not connected to the Internet, and only have local recognition and local control. Even if they are connected to the Internet, they cannot get upgrades or update voice commands from the cloud. For example, the product design supports the voice "25°", and then the product automatically sets it to 25° after recognition. If the user says "too hot", the product cannot recognize it. The voice command has been fixed at the beginning of the design.

[0005] Some products also support cloud-based voice recognition, which then returns recognition instructions to the product for execution. The disadvantage is that the network delay is long, and some products require waiting for 3-5 seconds or even longer before responding. This is especially true in remote areas with poor network environments, where this type of voice control is almost unacceptable.

[0006] Currently, no effective solution has been proposed to address the problem that smart home devices cannot be flexibly configured due to the rigidity of the development and / or settings of smart home devices in the prior art. Summary of the invention

[0007] In order to solve the above technical problems, the embodiments of the present invention hope to provide a voice-controlled home appliance system, a smart home device and a method, so as to at least solve the problem that the smart home devices cannot be flexibly configured due to the solidification of the development and / or setting of the smart home devices in the prior art.

[0008] The technical solution of the present invention is achieved in this way:

[0009] An embodiment of the present invention provides a voice-controlled home appliance system, comprising: a cloud, a smart home device and a mobile terminal, wherein the smart home device comprises: a voice integration module, an audio circuit and an MCU, and the voice integration module comprises a voice processor and a communication module; the cloud is connected to the smart home device and is used to send a corpus data packet to the smart home device, so that the smart home device receives the corpus data packet through the communication module in the voice integration module, and configures and stores the corpus data packet through the voice processor; the voice processor is connected to the audio circuit and is used to identify the voice received by the audio circuit through the corpus data packet to obtain a first control instruction corresponding to the voice; the communication module is connected to the voice processor and is used to send the first control instruction to the MCU; the MCU is used to execute a corresponding function according to the first control instruction; the mobile terminal is respectively connected to the cloud and the smart home device and is used to determine whether to trigger an upgrade instruction according to a version number of a pre-acquired corpus data packet, or to send a second control instruction to the smart home device through the cloud so that the smart home device executes a corresponding function according to the second control instruction.

[0010] Optionally, the voice processor includes: an audio codec and a voice processing engine, wherein the audio codec is respectively connected to the audio circuit and the voice processing engine, and is used to convert the voice from an analog signal to a digital signal when the collected voice is received through the audio circuit, and transmit the digital signal to the voice processing engine; when receiving the execution result returned by the voice processing engine based on the digital signal, the execution result is converted from a digital signal to an analog signal, and the analog signal is transmitted to the audio circuit; the voice processing engine is respectively connected to the audio codec and the communication module, and is used to receive the digital signal through the audio codec, and identify the wake-up word and / or command word from the digital signal through the corpus data packet, match the corresponding first control instruction according to the wake-up word and / or command word, and send the first control instruction to the communication module; or, receive the execution result returned by the communication module, and return the execution result to the audio codec.

[0011] Further, optionally, the voice processing engine is also used to receive the voice sent by the cloud and the first control instruction corresponding to the voice through the communication module, obtain the characteristic value of the voice, and generate characteristic values ​​for the voice with the same semantics as the voice, obtain the characteristic value sample space, correspond the characteristic value sample space with the first control instruction, obtain the corresponding relationship between the characteristic value sample space and the first control instruction, and transmit the corresponding relationship to the communication module; or, the voice processing engine is also used to receive the instruction of the learning mode triggered by the mobile terminal, collect the user voice through the audio circuit according to the instruction of the learning mode, and convert the user voice from an analog signal to a digital signal through an audio codec, obtain the characteristic value in the digital signal, correspond the characteristic value with the corresponding first control instruction, obtain the corresponding relationship, and transmit the corresponding relationship to the communication module.

[0012] Optionally, a communication module is used to receive the correspondence sent by the voice processing engine, update the correspondence between the wake-up word or command word and the first control instruction according to the correspondence; and upload the characteristic value sample space in the correspondence to the cloud.

[0013] Optionally, the cloud is used to send the characteristic value sample space to the smart home device when the user adds a new smart home device or upgrades the smart home device, so that the smart home device is updated according to the characteristic value sample space.

[0014] Optionally, the sound-controlled home appliance system also includes: a sound collecting device and a sound playing device, wherein the sound collecting device is connected to the audio circuit, and is used to collect voice and / or ambient sound, and transmit the voice and / or ambient sound to the audio circuit; the sound playing device is connected to the audio circuit, and is used to play the analog signal converted by the audio circuit according to the execution result of the audio codec.

[0015] Optionally, the mobile terminal is further used to determine whether to trigger an upgrade instruction based on the version number of the pre-acquired corpus data packet. When the upgrade instruction is triggered, a request for obtaining the corpus data packet to be upgraded is sent to the cloud, so that the cloud obtains the corresponding corpus data packet based on the request for obtaining the corpus data packet to be upgraded, and sends the corpus data packet to the smart home device, and returns the upgrade result of the smart home device to the mobile terminal; or, the corpus data packet to be upgraded is sent to the smart home device, so that the smart home device is upgraded based on the corpus data packet to be upgraded, and the upgrade result is returned to the mobile terminal.

[0016] Optionally, the mobile terminal is further used to send a second control instruction to the smart home device through the cloud, so that the cloud sends a control command to the smart home device according to the second control instruction, and the smart home device executes a corresponding function according to the control command.

[0017] An embodiment of the present invention provides a smart home device, which is applied to a voice-controlled home appliance system, and includes: a voice integration module, an audio circuit and an MCU, wherein the voice integration module includes: a voice processor and a communication module; the communication module is used to receive a corpus data packet sent from the cloud; the voice processor is connected to the communication module and the audio circuit respectively, and is used to configure and store the corpus data packet; the voice received by the audio circuit is recognized through the corpus data packet to obtain a control instruction corresponding to the voice; the communication module is connected to the voice processor, and is used to send the control instruction to the MCU; the MCU is used to execute a corresponding function according to the control instruction.

[0018] An embodiment of the present invention provides a control method applied to a voice-controlled home appliance system, which is applied to the voice-controlled home appliance system, including: recognizing the collected voice according to the acquired corpus data packet to obtain a recognition result; allocating corresponding control instructions according to the recognition result; and executing corresponding functions according to the control instructions.

[0019] Optionally, the collected voice is recognized according to the acquired corpus data packet, and the recognition result includes: judging by the voice processor whether the voice contains wake-up words and / or command words based on the corpus data packet; if the judgment result is yes, determining the recognition result as the detection of the wake-up words and / or command words; if the judgment result is no, the state of the smart home device remains unchanged.

[0020] Further, optionally, allocating corresponding control instructions according to the recognition results includes: allocating corresponding control instructions according to the wake-up words and / or command words in the recognition results through the communication module.

[0021] Optionally, the method also includes: in the process of upgrading the corpus data packet, receiving the corpus data packet to be upgraded sent by the cloud according to the version number of the corpus data packet to be upgraded detected by the mobile terminal; installing the corpus data packet to be upgraded in the voice processor; and assigning corresponding control instructions to the wake-up words, command words and playback words in the upgraded corpus data packet through the communication module.

[0022] Optionally, the method also includes: receiving voice and control instructions corresponding to the voice sent by the cloud through the communication module, obtaining feature values ​​of the voice through the voice processing engine in the voice processor, and generating feature values ​​for voices with the same semantics as the voices to obtain feature value sample space, matching the feature value sample space with the control instructions, obtaining a corresponding relationship between the feature value sample space and the control instructions, and transmitting the corresponding relationship to the communication module so that the communication module is updated according to the corresponding relationship; or, receiving a learning mode instruction triggered by the mobile terminal, collecting the user voice through the audio circuit according to the learning mode instruction through the voice processing engine in the voice processor, and converting the user voice from an analog signal to a digital signal through an audio codec, obtaining feature values ​​in the digital signal, matching the feature values ​​with the corresponding control instructions, obtaining a corresponding relationship, and transmitting the corresponding relationship to the communication module so that the communication module is updated according to the corresponding relationship.

[0023] The embodiment of the present invention provides a voice-controlled home appliance system, a smart home device and a method, which are connected to the smart home device through the cloud and are used to send a corpus data packet to the smart home device, so that the smart home device receives the corpus data packet through a communication module in a voice integration module, and configures and stores the corpus data packet through a voice processor; the voice processor is connected to an audio circuit and is used to identify the voice received by the audio circuit through the corpus data packet to obtain a first control instruction corresponding to the voice; the communication module is connected to the voice processor and is used to send the first control instruction to an MCU; the MCU is used to execute a corresponding function according to the first control instruction; a mobile terminal is respectively connected to the cloud and the smart home device and is used to determine whether to trigger an upgrade instruction according to a version number of a pre-acquired corpus data packet, or to send a second control instruction to the smart home device through the cloud so that the smart home device executes the corresponding function according to the second control instruction, thereby improving the technical effect of the flexibility of configuring the smart home device. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] The drawings described herein are used to provide a further understanding of the present invention and constitute a part of this application. The exemplary embodiments of the present invention and their descriptions are used to explain the present invention and do not constitute an improper limitation of the present invention. In the drawings:

[0025] Figure 1 A schematic diagram of a voice-controlled home appliance system provided by an embodiment of the present invention;

[0026] Figure 2 A schematic diagram of the structure of a corpus data packet in a voice-controlled home appliance system provided by an embodiment of the present invention;

[0027] Figure 3 A schematic diagram of a voice processor in a voice-controlled home appliance system provided by an embodiment of the present invention;

[0028] Figure 4 A schematic diagram of self-learning speech of a speech processing engine in a voice-controlled home appliance system provided by an embodiment of the present invention;

[0029] Figure 5 A schematic diagram of implementing voice library sharing in a voice-controlled home appliance system provided by an embodiment of the present invention;

[0030] Figure 6 A timing diagram of implementing voice library sharing in a voice-controlled home appliance system provided by an embodiment of the present invention;

[0031] Figure 7 The present invention provides a control method for a voice-controlled home appliance system. DETAILED DESCRIPTION

[0032] In order to enable those skilled in the art to better understand the scheme of the present invention, the technical scheme in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of the present invention.

[0033] It should be noted that the terms "first", "second" and the like in the specification, claims and drawings of the present invention are used to distinguish different objects rather than to limit a specific order.

[0034] It should also be noted that the following embodiments of the present invention may be executed separately, or in combination with each other, and the embodiments of the present invention do not impose any specific limitations on this.

[0035] An embodiment of the present invention provides a sound-controlled home appliance system. Figure 1 A schematic diagram of a sound-controlled home appliance system provided by an embodiment of the present invention; Figure 1 As shown, the voice-controlled home appliance system provided in the embodiment of the present application includes:

[0036] A cloud 12, a smart home device 14 and a mobile terminal 16, wherein the smart home device 14 includes: a voice integration module 141, an audio circuit 142 and an MCU 143, and the voice integration module 141 includes a voice processor 1411 and a communication module 1412; the cloud 12 is connected to the smart home device 14, and is used to send a corpus data packet to the smart home device 14, so that the smart home device 14 receives the corpus data packet through the communication module 1412 in the voice integration module 141, and configures and stores the corpus data packet through the voice processor 1411; the voice processor 1411 is connected to the audio circuit 142, and is used to recognize the voice received by the audio circuit 142 through the corpus data packet to obtain a first control instruction corresponding to the voice; the communication module 1412 is connected to the voice processor 1411, and is used to send the first control instruction to the MCU 143; the MCU 143, used to execute the corresponding function according to the first control instruction; the mobile terminal 16 is connected to the cloud 12 and the smart home device 14 respectively, and is used to determine whether to trigger the upgrade instruction according to the version number of the pre-acquired corpus data packet, or to send the second control instruction to the smart home device 14 through the cloud 12, so that the smart home device 14 executes the corresponding function according to the second control instruction.

[0037] Optionally, the sound-controlled home appliance system in the embodiment of the present application also includes: a sound collection device and a sound playback device, wherein the sound collection device is connected to the audio circuit 142, for collecting voice and / or ambient sound, and transmitting the voice and / or ambient sound to the audio circuit 142; the sound playback device is connected to the audio circuit 142, for playing the analog signal converted by the audio circuit 142 according to the execution result of the audio codec.

[0038] Specifically, Figure 1 As shown, the smart home device 14 includes: a voice integration module 141, an audio circuit 142 and an MCU 143. The voice integration module 141 and the MCU 143 communicate with each other through a serial port. The voice integration module 141 is connected to the audio circuit 142. The audio circuit 142 is connected to a sound collection device and a sound playback device respectively. In the embodiment of the present application, the sound collection device may be a mic (microphone), and the sound playback device may be a speaker (sound system);

[0039] In the embodiment of the present application, the voice-controlled home appliance system includes at least the following scenarios:

[0040] Scenario 1: R&D scenario:

[0041] During the research and development process, the manufacturer can set up the corpus data packets corresponding to each type of smart home device through the cloud. In the embodiment of the present application, the corpus data packets include: wake-up words, command words, broadcast words and corresponding control instructions;

[0042] In the assembly or upgrade scenario, for each type of smart home device, the corpus data packet can be sent to the smart home device 14 through the cloud 12, wherein the smart home device 14 receives the corpus data packet sent by the cloud 12 through the communication module 1412 in the voice integration module 141, and the voice processor 1411 in the voice integration module 141 is installed and stored according to the corpus data packet received by the communication module 1412.

[0043] Scenario 2: User usage scenario:

[0044] When the user uses it, the sound collection device in the smart home device 14 collects the user's voice, and the audio circuit 142 transmits the voice to the voice integration module 141. The voice processor 1411 in the voice integration module 141 recognizes the voice through the corpus data packet, obtains the first control instruction corresponding to the voice, and transmits the first control instruction to the MCU 143 through the communication module 1412. The MCU 143 executes the corresponding function according to the first control instruction.

[0045] in, Figure 2 A schematic diagram of the structure of a linguistic data packet in a voice-controlled home appliance system provided by an embodiment of the present invention; Figure 2 As shown, based on the structure of the corpus data packet, when the voice processor 1411 receives the voice, the voice processor 1411 recognizes the voice according to the corpus data packet, and the recognition can obtain the wake-up word and / or the command word, and then recognizes the protocol instruction corresponding to the wake-up word and / or the command word according to the corpus data packet (that is, the first control instruction in the embodiment of the present application), and sends the protocol instruction to the communication module 1412;

[0046] When the voice processor 1411 receives the execution result through the communication module 1412, the playback word is obtained through the corpus data packet recognition, and the corresponding protocol instructions and playback audio are configured according to the playback word, and played through the speaker.

[0047] The corpus data package in the embodiment of the present application is a corpus file defined based on product requirements in an actual product, and is generated using a script tool based on the format definition of the file.

[0048] Optionally, the voice processor 1411 includes: an audio codec and a voice processing engine, wherein the audio codec is respectively connected to the audio circuit 142 and the voice processing engine, and is used to convert the voice from an analog signal to a digital signal when the collected voice is received through the audio circuit 142, and transmit the digital signal to the voice processing engine; when receiving the execution result returned by the voice processing engine based on the digital signal, the execution result is converted from a digital signal to an analog signal, and the analog signal is transmitted to the audio circuit 142; the voice processing engine is respectively connected to the audio codec and the communication module 1412, and is used to receive the digital signal through the audio codec, and identify the wake-up word and / or command word from the digital signal through the corpus data packet, match the corresponding first control instruction according to the wake-up word and / or command word, and send the first control instruction to the communication module 1412; or, receive the execution result returned by the communication module 1412, and return the execution result to the audio codec.

[0049] Specifically, Figure 3 A schematic diagram of a voice processor in a voice-controlled home appliance system provided by an embodiment of the present invention; Figure 3 As shown, the speech processor 1411 includes: an audio codec and a speech processing engine; wherein the audio codec is recorded as CODEC;

[0050] The audio circuit 142 receives the collected voice through the microphone, and the audio codec converts the voice from an analog signal to a digital signal, and transmits the digital signal to the voice processing engine. The voice processing engine receives the digital signal through the audio codec, and identifies the wake-up word and / or command word from the digital signal through the corpus data packet, matches the corresponding first control instruction according to the wake-up word and / or command word, and sends the first control instruction to the communication module 1412; the communication module 1412 transmits the first control instruction to the MCU 143, and the MCU 143 executes the corresponding function according to the first control instruction;

[0051] The voice processing engine receives the execution result returned by the communication module 1412, and returns the execution result to the audio codec. The audio codec converts the execution result from a digital signal to an analog signal, and transmits the analog signal to the audio circuit 142. The audio circuit 142 controls the sound playback device to play according to the analog signal.

[0052] like Figure 3 As shown, the voice integration module 141 also includes: a voltage conversion module, wherein the voltage conversion module is a 5V to 3.3V DC conversion module, wherein 3.3V is used to power the voice processing engine; the voice processing engine and the communication module are connected via a UART / SPI serial port, and the communication module 1412 and the MCU 143 are connected via a UART serial port.

[0053] Further, optionally, the voice processing engine is also used to receive the voice and the first control instruction corresponding to the voice sent by the cloud 12 through the communication module 1412, obtain the characteristic value of the voice, and generate characteristic values ​​for the voice with the same semantics as the voice, obtain the characteristic value sample space, correspond the characteristic value sample space with the first control instruction, obtain the corresponding relationship between the characteristic value sample space and the first control instruction, and transmit the corresponding relationship to the communication module 1412; or, the voice processing engine is also used to receive the instruction of the learning mode triggered by the mobile terminal 16, collect the user voice through the audio circuit 142 according to the instruction of the learning mode, and convert the user voice from an analog signal to a digital signal through the audio codec, obtain the characteristic value in the digital signal, correspond the characteristic value with the corresponding first control instruction, obtain the corresponding relationship, and transmit the corresponding relationship to the communication module 1412.

[0054] Optionally, the communication module 1412 is used to receive the correspondence sent by the voice processing engine, update the correspondence between the wake-up word or command word and the first control instruction according to the correspondence; and upload the feature value sample space in the correspondence to the cloud 12.

[0055] Specifically, Figure 4 A schematic diagram of a voice processing engine self-learning voice in a voice-controlled home appliance system provided by an embodiment of the present invention; Figure 4As shown, the self-learning voice logic of the voice processing engine in the embodiment of the present application is: audio input or word input of a certain target function, obtaining feature values ​​from the audio and words by picking up sound, further obtaining feature values ​​through audio or words with the same meaning to construct a feature value sample space, and assigning corresponding protocol instructions to each type of voice according to the feature value sample space, wherein the protocol instruction can be the first control instruction in the embodiment of the present application.

[0056] In the embodiment of the present application, the method for the voice processing engine to learn voice commands includes two methods:

[0057] Method 1:

[0058] The APP in the mobile terminal selects the personalized command customization menu, selects the function to be operated, such as "increase 2 degrees", and records the personalized voice of "too cold" and uploads it to the cloud.

[0059] The cloud sends the voice and the protocol instruction corresponding to "increase by 2 degrees" to the voice integration module 141, and notifies the voice integration module 141 to enter the learning mode. The voice integration module 141 processes the received voice through the voice processing engine to obtain feature values. The feature values ​​generated for the voices with the same semantics become the feature value sample space and correspond to a certain protocol instruction.

[0060] The communication module obtains the binding of the eigenvalue sample space and the protocol instruction (ie, adds the new eigenvalue to the corresponding relationship of the protocol instruction).

[0061] Usually, the APP in the mobile terminal needs to issue 2-5 voice recordings of the same protocol command, and the voice integration module 141 also needs 2-5 learnings to obtain the minimum deviation of the characteristic value, and then the instruction learning process is completed. The communication module uploads the characteristic value sample space to the cloud;

[0062] The user can try to control the smart home device 14 by voice in front of the voice integration module 141. If the function is completed normally, the learning is successful; otherwise, re-learning is required.

[0063] Method 2:

[0064] The APP in the mobile terminal notifies the smart home device 14 to enter the command learning mode and select a corresponding function (such as turning on or off the air conditioner);

[0065] The voice integration module 141 enters a learning mode of a certain function, collects the user's personalized voice through a microphone, processes the personalized voice through a voice processing engine to obtain a feature value, and completes the mapping relationship between the feature value and the control instruction through the communication module.

[0066] Through multiple learning cycles, the voice processing engine obtains the minimum variance (best match) of the feature value, and completes the actual home appliance control under the control of the communication module, notifying the voice link broadcast.

[0067] The communication module uploads the eigenvalue sample space and protocol instructions to the cloud to complete the upgrade of the voice library.

[0068] Optionally, the cloud 12 is used to send the characteristic value sample space to the smart home device 14 when the user adds a new smart home device 14 or upgrades the smart home device 14, so that the smart home device 14 is updated according to the characteristic value sample space.

[0069] Specifically, for new products or home appliances that need to be upgraded, the cloud can send the feature value sample space and protocol instructions to the voice integration module 141 of each product to complete the sharing of the voice library.

[0070] like Figure 5 As shown, Figure 5 A schematic diagram of implementing voice library sharing in a voice-controlled home appliance system provided in an embodiment of the present invention; all voice-controlled smart home appliances (i.e., smart home devices in the embodiment of the present application) are configured and added to the same account and configured into a group. The number of devices in the group depends on the memory of the mobile phone (i.e., the mobile terminal in the embodiment of the present application) and is basically not limited. Under normal circumstances, 64 devices can meet the application scenarios of most home appliances; the APP end (i.e., the APP in the mobile terminal in the embodiment of the present application) obtains the version number of the product corpus data packet of each device; the APP initiates a voice corpus upgrade instruction, including the MAC of each device and the version number of the product corpus data packet; the APP initiates a voice corpus upgrade instruction, including the MAC of each device and the version number of the product corpus data packet; the cloud sends the product corpus data packet OTA to each device; after the OTA of the product corpus data of each device is completed, the version number is reported; the APP determines whether the OTA of each product is successful by querying the version number of the product corpus data packet of each device.

[0071] like Figure 6 As shown, Figure 6 A timing diagram of implementing voice library sharing in a voice-controlled home appliance system provided in an embodiment of the present invention; by updating the product corpus data packet and then updating the product functions, including newly added voice control, users can use more new functions while keeping the smart home appliance hardware unchanged, thereby improving the user experience.

[0072] Optionally, the mobile terminal 16 is further used to determine whether to trigger an upgrade instruction based on the version number of the pre-acquired corpus data packet. When the upgrade instruction is triggered, a request for obtaining the corpus data packet to be upgraded is sent to the cloud 12, so that the cloud 12 obtains the corresponding corpus data packet according to the request for obtaining the corpus data packet to be upgraded, and sends the corpus data packet to the smart home device 14, and returns the upgrade result of the smart home device 14 to the mobile terminal 16; or, the corpus data packet to be upgraded is sent to the smart home device 14, so that the smart home device 14 is upgraded according to the corpus data packet to be upgraded, and the upgrade result is returned to the mobile terminal 16.

[0073] Optionally, the mobile terminal 16 is further used to send a second control instruction to the smart home device 14 through the cloud 12, so that the cloud 12 sends a control command to the smart home device 14 according to the second control instruction, and the smart home device 14 performs a corresponding function according to the control command.

[0074] Specifically, in the embodiment of the present application, the mobile terminal 16 can send the custom voice to the smart home device 14 or send it to the smart home device 14 through the cloud through the custom voice; so that the smart home device 14 performs self-learning according to the custom voice through the voice processing engine according to artificial intelligence;

[0075] And in remote control, the mobile terminal 16 sends a second control instruction to the smart home device 14 through the cloud 12, so that the smart home device 14 performs a corresponding function according to the control command sent by the cloud.

[0076] The voice-controlled home appliance system provided in the embodiment of the present invention is mainly used to quickly customize the project, and a development mode is determined, based on the voice integration module 141 of the entire voice-controlled smart home device 14, to determine the generalized and configured parts, wherein the voice integration module 141 is connected to the electrical control MCU 143 of the home appliance through the serial port;

[0077] During the configuration process, the corpus data packet includes: wake-up words, command words, audio data corresponding to the command words and mapped instructions, play words, audio data corresponding to the play words and mapped instructions. For different categories of smart home appliances, corresponding customized processing can be done.

[0078] In terms of generalization, for serial port communication with home appliance MCU, generalization can be defined; if the home appliance with existing electronic control is to be upgraded to be intelligent, the instructions related to voice control and playback can be extracted based on the electronic control of the home appliance, and mapped to the product corpus data package.

[0079] Based on the above configuration and generalization, rapid docking of voice-controlled smart home appliances can be achieved, shortening the product development cycle and facilitating subsequent functional upgrades and expansions.

[0080] New categories of smart home appliances: extract the supported functions of the product, define welcome words, command words and play words, record or synthesize the corresponding audio, extract the corresponding mapping protocol instructions based on the MCU communication protocol of the product, and then generate the corpus data packet of the product, connect to the electronic control MCU, follow the general interaction process of the electronic control MCU adaptation module, and then extract the instructions and add them to the corpus data packet.

[0081] For smart home appliances of the same category, new functions can be realized by expanding the corpus data package.

[0082] The embodiment of the present invention provides a voice-controlled home appliance system, which is connected to a smart home device through a cloud, and is used to send a corpus data packet to the smart home device, so that the smart home device receives the corpus data packet through a communication module in a voice integration module, and configures and stores the corpus data packet through a voice processor; the voice processor is connected to an audio circuit, and is used to identify the voice received by the audio circuit through the corpus data packet to obtain a first control instruction corresponding to the voice; the communication module is connected to the voice processor, and is used to send the first control instruction to an MCU; the MCU is used to execute a corresponding function according to the first control instruction; a mobile terminal is respectively connected to the cloud and the smart home device, and is used to determine whether to trigger an upgrade instruction according to a version number of a pre-acquired corpus data packet, or to send a second control instruction to the smart home device through the cloud, so that the smart home device executes the corresponding function according to the second control instruction, thereby improving the technical effect of the flexibility of configuring the smart home device.

[0083] The embodiment of the present invention provides a smart home device, which is applied to Figures 1 to 6 The voice-controlled home appliance system shown in the embodiment of the present application includes: a voice integration module, an audio circuit and an MCU, wherein the voice integration module includes: a voice processor and a communication module; the communication module is used to receive corpus data packets sent from the cloud; the voice processor is connected to the communication module and the audio circuit respectively, and is used to configure and store corpus data packets; the voice received by the audio circuit is recognized through the corpus data packets to obtain control instructions corresponding to the voice; the communication module is connected to the voice processor, and is used to send the control instructions to the MCU; the MCU is used to execute corresponding functions according to the control instructions.

[0084] The embodiment of the present invention provides a smart home device, which is used to receive a corpus data packet sent from the cloud through a communication module; a voice processor is connected to the communication module and an audio circuit respectively, and is used to configure and store the corpus data packet; the voice received by the audio circuit is recognized through the corpus data packet to obtain a control instruction corresponding to the voice; the communication module is connected to the voice processor and is used to send the control instruction to the MCU; the MCU is used to execute a corresponding function according to the control instruction, thereby achieving a technical effect of improving the flexibility of the configuration of the smart home device.

[0085] The embodiment of the present invention provides a control method for a sound-controlled home appliance system, which is applied to Figures 1 to 6 The audio-controlled home appliance system shown in Figure 7 The present invention provides a control method for a sound-controlled home appliance system. Figure 7 As shown, the control method for a voice-controlled home appliance system provided in an embodiment of the present application includes:

[0086] Step S702, recognizing the collected speech according to the acquired corpus data packet to obtain a recognition result;

[0087] Step S704, assigning corresponding control instructions according to the recognition result;

[0088] Step S706, executing a corresponding function according to the control instruction.

[0089] Optionally, in step S702, the collected voice is recognized according to the acquired corpus data packet, and the recognition result includes: judging by the voice processor whether the voice contains the wake-up word and / or command word based on the corpus data packet; if the judgment result is yes, determining that the recognition result is that the wake-up word and / or command word is detected; if the judgment result is no, the state of the smart home device remains unchanged.

[0090] Further, optionally, assigning corresponding control instructions according to the recognition results in step S704 includes: assigning corresponding control instructions according to the wake-up word and / or command word in the recognition results through the communication module.

[0091] Optionally, the control method for a voice-controlled home appliance system provided by an embodiment of the present invention also includes: in the process of upgrading the corpus data packet, receiving the corpus data packet to be upgraded sent by the cloud according to the version number of the corpus data packet to be upgraded detected by the mobile terminal; installing the corpus data packet to be upgraded in the voice processor; and assigning corresponding control instructions to the wake-up words, command words and playback words in the upgraded corpus data packet through the communication module.

[0092] Optionally, the control method for a voice-controlled home appliance system provided by an embodiment of the present invention also includes: receiving voice and control instructions corresponding to the voice sent from the cloud through a communication module, obtaining feature values ​​of the voice through a voice processing engine in a voice processor, generating feature values ​​for voices with the same semantics as the voices, obtaining feature value sample space, matching the feature value sample space with control instructions, obtaining a corresponding relationship between the feature value sample space and the control instructions, and transmitting the corresponding relationship to the communication module so that the communication module is updated according to the corresponding relationship; or, receiving a learning mode instruction triggered by a mobile terminal, collecting user voice through an audio circuit according to the learning mode instruction through the voice processing engine in the voice processor, converting the user voice from an analog signal to a digital signal through an audio codec, obtaining feature values ​​in the digital signal, matching the feature values ​​with corresponding control instructions, obtaining a corresponding relationship, and transmitting the corresponding relationship to the communication module so that the communication module is updated according to the corresponding relationship.

[0093] An embodiment of the present invention provides a control method for a voice-controlled home appliance system provided by an embodiment of the present invention, which obtains a recognition result by recognizing the collected voice according to an acquired corpus data packet; assigning a corresponding control instruction according to the recognition result; and executing a corresponding function according to the control instruction, thereby achieving a technical effect of improving the flexibility of the configuration of smart home devices.

[0094] The above are only preferred embodiments of the present invention and are not intended to limit the protection scope of the present invention.

Claims

1. A sound-controlled home appliance system, characterized in that: include: Cloud, smart home devices and mobile terminals, among which, The smart home device includes: a voice integration module, an audio circuit and an MCU, wherein the voice integration module includes a voice processor and a communication module; The cloud is connected to the smart home device and is used to send the corpus data packet to the smart home device, so that the smart home device receives the corpus data packet through the communication module in the voice integration module, and configures and stores the corpus data packet through the voice processor; The speech processor is connected to the audio circuit and is used to identify the speech received by the audio circuit through the corpus data packet to obtain a first control instruction corresponding to the speech; The communication module is connected to the voice processor and is used to send the first control instruction to the MCU; The MCU is used to execute a corresponding function according to the first control instruction; The mobile terminal is connected to the cloud and the smart home device respectively, and is used to determine whether to trigger an upgrade instruction according to the version number of the pre-acquired corpus data packet, or to send a second control instruction to the smart home device through the cloud, so that the smart home device performs a corresponding function according to the second control instruction.

2. The sound-controlled home appliance system according to claim 1, characterized in that: The speech processor includes: an audio codec and a speech processing engine, wherein: The audio codec is connected to the audio circuit and the voice processing engine respectively, and is used to convert the voice from an analog signal to a digital signal when the collected voice is received through the audio circuit, and transmit the digital signal to the voice processing engine; when receiving the execution result returned by the voice processing engine according to the digital signal, convert the execution result from a digital signal to an analog signal, and transmit the analog signal to the audio circuit; The speech processing engine is connected to the audio codec and the communication module, respectively, and is used to receive the digital signal through the audio codec, and identify the wake-up word and / or command word from the digital signal through the corpus data packet, match the corresponding first control instruction according to the wake-up word and / or the command word, and send the first control instruction to the communication module; or, receive the execution result returned by the communication module, and return the execution result to the audio codec.

3. The sound-controlled home appliance system according to claim 2, characterized in that: The speech processing engine is further used to receive the speech sent by the cloud and the first control instruction corresponding to the speech through the communication module, obtain the feature value of the speech, generate feature values ​​for speech with the same semantics as the speech, obtain a feature value sample space, correspond the feature value sample space with the first control instruction, obtain a corresponding relationship between the feature value sample space and the first control instruction, and transmit the corresponding relationship to the communication module; or, The voice processing engine is also used to receive instructions for the learning mode triggered by the mobile terminal, collect user voice through the audio circuit according to the instructions for the learning mode, convert the user voice from an analog signal to a digital signal through the audio codec, obtain characteristic values ​​in the digital signal, match the characteristic values ​​with the corresponding first control instructions, obtain a corresponding relationship, and transmit the corresponding relationship to the communication module.

4. The sound-controlled home appliance system according to claim 3, characterized in that: The communication module is used to receive the corresponding relationship sent by the voice processing engine, and update the corresponding relationship between the wake-up word or the command word and the first control instruction according to the corresponding relationship; And upload the characteristic value sample space in the corresponding relationship to the cloud.

5. The sound-controlled home appliance system according to claim 4, characterized in that: The cloud is used to send the characteristic value sample space to the smart home device when the user adds a new smart home device or upgrades the smart home device, so that the smart home device is updated according to the characteristic value sample space.

6. The sound-controlled home appliance system according to claim 2, characterized in that: The sound-controlled home appliance system further includes: a sound collection device and a sound playback device, wherein: The sound collection device is connected to the audio circuit and is used to collect voice and / or ambient sound, and transmit the voice and / or ambient sound to the audio circuit; The sound playing device is connected to the audio circuit and is used to play the analog signal converted by the audio circuit according to the audio codec based on the execution result.

7. The sound-controlled home appliance system according to claim 1, characterized in that: The mobile terminal is further used to determine whether to trigger an upgrade instruction based on the pre-acquired version number of the corpus data packet. When the upgrade instruction is triggered, a request for obtaining the corpus data packet to be upgraded is sent to the cloud, so that the cloud obtains the corresponding corpus data packet according to the request for obtaining the corpus data packet to be upgraded, and sends the corpus data packet to the smart home device, and returns the upgrade result of the smart home device to the mobile terminal; or, sends the corpus data packet to be upgraded to the smart home device, so that the smart home device is upgraded according to the corpus data packet to be upgraded, and returns the upgrade result to the mobile terminal.

8. The sound-controlled home appliance system according to claim 1, characterized in that: The mobile terminal is also used to send a second control instruction to the smart home device through the cloud, so that the cloud sends a control command to the smart home device according to the second control instruction, and the smart home device performs a corresponding function according to the control command.

9. A smart home device, characterized in that: Applied to audio-controlled home appliance systems, including: Voice integrated module, audio circuit and MCU, among which, The voice integration module includes: a voice processor and a communication module; The communication module is used to receive the corpus data packets sent from the cloud; The voice processor is connected to the communication module and the audio circuit respectively, and is used to configure and store the linguistic data packet; the voice received by the audio circuit is recognized through the linguistic data packet to obtain a control instruction corresponding to the voice; The communication module is connected to the voice processor and is used to send the control instruction to the MCU; The MCU is used to execute corresponding functions according to the control instructions.

10. A control method applied to a sound-controlled home appliance system, characterized in that: Applied to audio-controlled home appliance systems, including: Recognize the collected speech according to the acquired corpus data packet to obtain a recognition result; Allocating corresponding control instructions according to the recognition result; Execute corresponding functions according to the control instructions.

11. The control method for a sound-controlled home appliance system according to claim 10, characterized in that: The step of recognizing the collected speech according to the acquired corpus data packet to obtain the recognition result includes: Determining, by a voice processor, whether the voice contains a wake-up word and / or a command word according to the corpus data packet; If the judgment result is yes, determining that the recognition result is detection of a wake-up word and / or a command word; When the judgment result is negative, the state of the smart home device remains unchanged.

12. The control method for a sound-controlled home appliance system according to claim 11, characterized in that: The assigning of corresponding control instructions according to the recognition result comprises: The communication module allocates corresponding control instructions according to the wake-up word and / or the command word in the recognition result.

13. The control method for a sound-controlled home appliance system according to claim 10, characterized in that: The method further comprises: In the process of upgrading the corpus data packet, receiving the corpus data packet to be upgraded sent by the cloud according to the version number of the corpus data packet to be upgraded detected by the mobile terminal; Installing the corpus data packet to be upgraded in the speech processor; Corresponding control instructions are assigned to the wake-up words, command words and playback words in the upgraded corpus data packet through the communication module.

14. The control method for a sound-controlled home appliance system according to claim 12, characterized in that: The method further comprises: Receiving the voice and the control instruction corresponding to the voice sent by the cloud through the communication module, obtaining the feature value of the voice through the voice processing engine in the voice processor, generating feature values ​​for the voice with the same semantics as the voice, obtaining the feature value sample space, making the feature value sample space correspond to the control instruction, obtaining the correspondence between the feature value sample space and the control instruction, and transmitting the correspondence to the communication module, so that the communication module is updated according to the correspondence; or, Receive the instruction of the learning mode triggered by the mobile terminal, collect the user voice through the audio circuit according to the instruction of the learning mode through the voice processing engine in the voice processor, convert the user voice from an analog signal to a digital signal through the audio codec, obtain the characteristic value in the digital signal, correspond the characteristic value with the corresponding control instruction to obtain the corresponding relationship, and transmit the corresponding relationship to the communication module so that the communication module is updated according to the corresponding relationship.

Images