WO2020216089A1 - Voice control system and method, and voice suite and voice apparatus - Google Patents

Abstract

The present application discloses a voice control system and method, and a voice suite and a voice apparatus. The system comprises a plurality of voice suites and a server in communication with the voice suites, wherein each voice suite comprises: a voice apparatus, configured to acquire voice input and send the acquired voice input to a processing apparatus; and a processing apparatus, configured to receive the voice input acquired by the voice apparatus, and send the voice input to a server. The server is configured to perform semantic recognition on the voice input sent by the processing apparatus, so as to generate and issue an operation command of a target device operation corresponding to the recognized semantic. Thus, using a discrete voice apparatus as a voice acquisition entrance, and implementing, by means of a processing apparatus and a server, semantic parsing and command issuing facilitates flexible voice control of various devices, especially various smart home devices and even legacy devices.

Description

Voice control system and method, voice kit and voice device

This application claims the priority of the Chinese patent application filed on April 25, 2019 with the application number 201910339913.9 and the invention title "voice control system and method, and voice kit and voice device", the entire content of which is incorporated into this application by reference in.

Technical field

This application relates to the field of information technology, and in particular to a voice control system and method, as well as a voice kit and a voice device.

Background technique

With the popularization and development of smart home technology, the use of voice to control various household appliances has become standard. In the existing technology, the voice control can be realized by a smart speaker as a control node in the home or an electrical appliance itself with a voice interaction function.

However, the volume and power consumption of existing smart speakers are relatively large, and it is usually difficult to achieve battery power supply. Therefore, the existence of the power cord, the large size, and the requirements for the external environment (for example, not suitable for placing in a bathroom with high humidity, etc.) restrict the flexible arrangement and use of smart speakers.

Devices with voice interaction functions are usually equipped with specialized voice processing modules. These voice processing modules can only be arranged inside the device. On the one hand, the cost of mold opening and separate debugging is high. On the other hand, due to the constraints of the equipment assembly, they cannot be placed flexibly and have poor adaptability.

For this reason, a more flexible and easy-to-implement voice solution is needed.

Summary of the invention

In order to solve at least one of the above-mentioned problems, this application proposes a separate voice device as an entrance to intelligent voice. The voice device collects voice input and sends it to the processing device, which implements semantic analysis and corresponding command issuance locally or in the cloud, thus facilitating various types of equipment, especially various smart home equipment and even traditional equipment. Flexible control.

According to one aspect of the present application, a voice control system is proposed, including a plurality of voice kits and a server communicating with the voice kit, wherein the voice kit includes: a voice device for collecting voice input and collecting The received voice input is sent to the processing device; the processing device is used to receive the voice input collected by the voice device and send the voice input to the server, and the server is used to send to the processing device Perform semantic recognition on the voice input to generate and issue operation commands for the operation of the target device corresponding to the recognition semantics. The voice kit may include a plurality of the voice devices arranged in different areas, and each processing device is communicatively connected with one or more of the voice devices within a communication range. Thus, through the convenient arrangement of the voice collection entrance, the flexibility and coverage of voice control are improved.

Preferably, the voice device and the processing device perform short-distance communication, for example, each includes a low-power short-distance communication module for local short-distance communication with each other, and the communication module includes at least one of the following: A Bluetooth communication module that communicates with the processing device using Bluetooth technology. An infrared communication module that communicates with the processing device based on infrared technology; and a Zigbee communication module that communicates with the processing device based on Zigbee technology.

The server can be implemented as a remote server, so the processing device can include a WiFi module for remote communication with the server.

Preferably, the voice device recognizes a wake-up word from the collected voice input, and sends the collected voice input to the processing device after the wake-up module recognizes the wake-up word.

Preferably, the voice device performs voice output on the content received from the processing device and/or the target device via its own speaker or a wired/wireless external speaker. The content of the voice output may include at least one of the following: the statement content of the execution command; and the interaction content with the user, for example, further obtaining the missing semantic elements.

Preferably, the voice device is also used to perform infrared output on the command received by the processing device, so as to realize the operation of the target device corresponding to the command.

Preferably, the voice device further includes an attachment mechanism for attaching to the wall, the processing device, the target device, or the surface of other facilities.

Preferably, the processing device is configured to perform semantic recognition for at least part of the voice input, and generate operation commands for recognizing the operation of the target device corresponding to the semantics.

In the voice control system of the present application, the target device may directly receive the issued operation command from the server and execute the operation corresponding to the operation command; and/or the processing device may receive the download from the server The issued operation command is issued to the target device by itself or via the voice device. Correspondingly, the target device that directly receives and executes the issued operation command from the server may include a networked smart home appliance; and the target device that obtains the operation command via the processing device or the voice device includes Traditional home appliances.

Preferably, the server may include: a semantic processing server for semantic recognition of uploaded voice input; a command generation server for generating operation commands for target recognition operations based on the recognized semantics; and issuing The command of the operation command is issued to the server.

As an alternative or supplement, the command generation and issuance for some target devices can be completed by an external server. Thus, the server may include a semantic processing server for semantic recognition of the uploaded voice input, and the server sends the recognized semantics to an external server, wherein the external server is based on The recognized semantics generates a command generation server for the operation command for the target recognition operation, and the command issuance server for issuing the operation command.

Preferably, the server may obtain in advance at least one piece of local device configuration information: the distribution and device information of the voice device, the processing device, and/or at least part of the target device; the voice device, the processing device The correspondence between at least two of the device and the target device.

Preferably, the server automatically fills in semantic elements that are missing in the recognition semantics for performing operations on the target device based on the local device configuration information.

Preferably, the server may include: a local server that communicates with the voice kit in close range, and the local server may be used to perform semantic recognition for at least part of the voice input, and generate operations for recognizing the operation of the target device corresponding to the semantics Command and issue the operation command.

According to another aspect of the present application, a voice kit is provided, including: a voice device for collecting voice input and sending the collected voice input to a processing device via local communication; and a processing device, including a voice device A communication unit connected in communication, the processing device receives voice data collected by the voice device via the communication unit, so as to realize the semantic recognition of the voice input and the target device operation corresponding to the recognition semantics.

Preferably, the voice device and the processing device may each include a low-power short-range communication module for performing short-range communication with each other, and the communication module includes at least one of the following: communicating with the processing device based on Bluetooth technology Bluetooth communication module; infrared communication module based on infrared technology to communicate with the processing device; and Zigbee communication module based on Zigbee technology to communicate with the processing device.

The processing device may include: a networking unit for uploading voice data from the user received from the voice device to a server, wherein the server performs semantic recognition for the voice input to generate and Issue operation commands that identify the operation of the target device corresponding to the semantics. Therefore, by changing the networked device to the processing device, the cloud can be used to realize intelligent AI voice processing while ensuring the miniaturization of the voice stickers.

Preferably, the target device can directly receive the issued operation command from the server and execute the operation corresponding to the operation command; and/or the processing device receives the issued operation command from the server via the networking unit The operation command is issued to the target device by itself or via the voice device. In this way, different execution methods of cloud-issued commands can be implemented according to different application scenarios: direct control of home appliances, AI module control, voice paste control, etc. Specifically, the target device that directly receives the issued operation command from the server and executes it may include its own networked smart home appliance; and the target device that obtains the operation command via the processing device or the voice device includes a traditional home appliance. .

Further, the networking unit may be further used for: receiving the interactive content with the user issued by the server, and the communication unit is also used for: sending the interactive content with the user to the voice device for Voice output. As a result, the voice entry function of voice stickers is further improved.

According to different application scenarios, the voice devices and processing devices in the kit can be freely combined. For example, the kit may include multiple voice devices arranged in different areas, and each processing device is connected to one or more devices in the communication range. The communication connection of miniaturized equipment.

As an alternative or supplement, the processing device may further include: a voice recognition unit for performing semantic recognition for the voice input; and an operation command generating unit for generating operation commands for recognizing the operation of the target device corresponding to the semantics. This enables fast local processing of simple commands.

According to one aspect of the present application, a voice device is proposed, which includes: a microphone for collecting voice input; a communication module for sending the collected voice input to a processing device, so as to realize the semantics of the voice input Identify and identify the target device operations corresponding to the semantics. As a result, by retaining only the simplest semantic collection and communication function voice stickers as the entrance of voice input, flexible arrangement of voice function modules and convenient voice collection are realized. Further, the device may include an attachment mechanism for attaching to a wall, a processing device, a target device, or the surface of other facilities, for example, in the form of a smart voice sticker.

Preferably, the device may further include: a wake-up module for recognizing a wake-up word from a voice input from a user, and the communication module is used for collecting the wake-up word after the wake-up module recognizes the wake-up word The voice input is sent to the processing device. The wake-up module can be implemented by an existing low-power miniaturized DSP, thereby providing a far-field wake-up function while ensuring low power consumption and miniaturization.

Preferably, the device may further include: a speaker for voice output of the content received by the communication module from the processing device. The content of the voice output by the speaker includes at least one of the following: statement content for executing the command; and interactive content with the user, for example, further obtaining missing semantic elements. Thus, the voice portal function of the above device is further completed.

Preferably, the device may further include: an external speaker connection module, so that the connected external speaker performs voice output on the content received by the communication module from the processing device. The external speaker connection module is at least one of the following: the communication module including a Bluetooth connection function to connect an external Bluetooth speaker; and an external speaker wired connection module including an audio jack. As a result, the device can be combined with existing Bluetooth or traditional speakers, so that the aforementioned speakers can become smart devices after being connected to voice stickers, which can realize voice control and cloud resource access.

Preferably, the device may further include: an infrared module, which is used to perform infrared output on the command received by the communication module from the processing device, so as to realize the operation of the target device corresponding to the command. Thus, the device can be multiplexed as a universal infrared remote control for traditional home appliances.

Preferably, the communication module is a low-power short-range communication module, such as a Bluetooth communication module that communicates with the processing device based on Bluetooth technology, or an infrared communication module that communicates with the processing device based on infrared technology, or It is a Zigbee communication module based on Zigbee technology to communicate with the processing device. This provides the possibility for miniaturization and low-power communication.

Preferably, the device may further include: a power supply module, the power supply module includes at least one of the following: a wireless charging assembly; a battery assembly; and a USB socket. The low power consumption of the device can eliminate the need for a power cord, thereby further improving its portability.

According to another aspect of the present application, a voice control method is proposed, which can be implemented by the above voice device, kit and system, and includes: the voice device collects voice input; the voice device sends the voice input to the processing device; The processing device realizes semantic recognition for the voice input and generation of corresponding target device operation commands.

Preferably, the voice device sending the voice input to the processing device includes: the voice device sending the voice input to the processing device via short-range communication.

Preferably, the processing device that implements semantic recognition for the voice input and the generation of the corresponding target device operation command includes: the processing device uploads the voice input to the server; the server performs the semantics of the voice input Recognition to obtain the operation command of the target device operation corresponding to the recognition semantics.

Preferably, the processing device uploading the voice input to the server includes: the processing device uploads the voice input to the local server via short-range communication.

Preferably, the semantic recognition of the voice input by the server to obtain the operation command of the target device operation corresponding to the recognition semantics includes: the local server performs semantic recognition for at least part of the voice input to generate recognition semantics The operation command corresponding to the operation of the target device is issued and the operation command is issued.

Preferably, the processing device that implements semantic recognition for the voice input and the generation of corresponding target device operation commands includes: the processing device performs semantic recognition for at least part of the voice input, and generates the target device operation corresponding to the recognition semantics The operation command.

Preferably, the method may further include: using the voice device and/or the processing device to obtain an operation command for identifying the operation of the target device corresponding to the semantics; and the voice device and/or the processing device send the voice device and/or the processing device to the target The device issues the operation command.

Preferably, the method may further include: the processing device obtains and delivers the voice output content; the processing device delivers the voice output content to the voice device; and the voice device outputs the voice output content by voice.

The external and discrete modular solutions of this application can be easily integrated into any required equipment. In addition, the kit of the present application is convenient to use, and the miniaturized and portable design of the voice module makes it easy to be placed on various surfaces and suitable for use in multiple scenarios. Furthermore, the smart voice sticker of the present application can also realize the function of a universal infrared remote control by adding an infrared module, thereby realizing intelligent control of traditional infrared devices.

Description of the drawings

Through a more detailed description of the exemplary embodiments of the present disclosure in conjunction with the accompanying drawings, the above and other objectives, features, and advantages of the present disclosure will become more apparent. Among them, in the exemplary embodiments of the present disclosure, the same reference numerals generally represent The same parts.

Fig. 1 shows a schematic diagram of the composition of a speech device according to an embodiment of the present application;

Figure 2 shows an example implemented as a smart voice sticker;

Figure 3 shows an example of the composition of an intelligent voice sticker in this application;

Figure 4 shows an example of the circuit processing flow of a voice device of the present application;

FIG. 5 shows a schematic diagram of the composition of a voice kit according to an embodiment of the present application;

Fig. 6 shows a schematic diagram of the composition of a voice control system according to an embodiment of the present application;

Fig. 7 shows a schematic flowchart of a voice control method according to an embodiment of the present application;

Figure 8 shows an example of voice control in this application.

Detailed ways

Hereinafter, preferred embodiments of the present disclosure will be described in more detail with reference to the accompanying drawings. Although the preferred embodiments of the present disclosure are shown in the drawings, it should be understood that the present disclosure can be implemented in various forms and should not be limited by the embodiments set forth herein. On the contrary, these embodiments are provided to make the present disclosure more thorough and complete, and to fully convey the scope of the present disclosure to those skilled in the art.

As mentioned above, the existing smart home voice interaction solutions (for example, smart speakers and devices with voice interaction functions) have problems such as low flexibility and limited use scenarios. For this reason, a separate voice device is proposed as the entrance of intelligent voice. The voice device collects voice input and sends it to the processing device via local communication. The latter implements semantic analysis and corresponding command issuance via local or cloud, thus facilitating various types of equipment, especially various smart home equipment and even Flexible control of traditional equipment.

Fig. 1 shows a schematic diagram of the composition of a speech device according to an embodiment of the present application. As shown in FIG. 1, the voice device 100 may include a microphone (MIC) 110 and a communication module 120. MIC 110 is used to collect voice input. The communication module 120 is used to send the collected voice input to the processing device, so as to realize the semantic recognition of the voice input and the target device operation corresponding to the semantic recognition based on the processing device. As described below, the target device can be a smart device or a traditional home appliance.

The voice device can be combined with the processing device to form a voice kit (as shown in Figure 5 below). Here, "kit" refers to a group of devices that assist in achieving specific functions. In this application, the voice device is used as an entry point for collecting voice signals from users in different areas due to its flexible and easy-to-arrange features, and the processing device summarizes the above-mentioned voice information collected by the voice device, and performs local and/or cloud semantic processing. The function realizes the semantic recognition of the voice input and the operation of the target device corresponding to the recognition semantics.

In the case of cloud semantic recognition and command issuance, the above voice kit can also be combined with the server to form a voice control system (as shown in Figure 6 below). The server can be a local server that communicates with the voice kit via a short distance, or it can communicate remotely with processing devices in multiple voice kits to provide them with cloud semantic recognition, command generation, and remote services respectively. End (for example, server farm).

Returning to Figure 1, MIC is a transducer that converts sound into electrical signals. As shown in the figure, the MIC 110 can collect the sound signal sent by the user, convert it into an electric signal containing the user's sound information, and send it to the communication module 120. The communication module 120 may send an electrical signal containing user information as the user's voice input data to the processing device for subsequent semantic recognition of voice input and target device operations corresponding to the semantic recognition. In other embodiments, the MIC may also collect sound signals emitted by non-users, for example, sound signals emitted by other smart devices.

Therefore, the voice device 100 can achieve low power consumption and miniaturization by retaining only the simplest semantic collection and communication functions, so as to facilitate its flexible installation.

In order to achieve miniaturization and low power consumption, the communication module 120 may be a short-range communication module with low power consumption. Here, "near-field communication" refers to short-distance wireless communication where the communication distance is usually within a few hundred meters. In one embodiment, the communication module 120 may be a Bluetooth (BT) communication module that communicates with the processing device based on Bluetooth technology, for example, a communication module based on a Bluetooth Mesh solution. In another embodiment, the communication module 120 may be an infrared (Infrared, IR) communication module that communicates with the processing device based on infrared technology, for example, a high-speed infrared transmission module. In another embodiment, the communication module 120 may be a Zigbee communication module that communicates with the processing device based on Zigbee technology. In other embodiments, the communication module 120 may also use a combination of BT and IR. It should be understood that the communication module 120 of the present application can also be implemented by, for example, a new low-power short-distance communication technology developed in the future, so as to create conditions for its flexible layout through the miniaturization and low power consumption of the device itself.

In other embodiments, the voice device 100 may also include a WiFi communication module that has relatively low power consumption and generally requires stronger processing functions, so as to communicate with the processing device using a local area network. Of course, the aforementioned WiFi communication module may also be used for short-distance communication in some embodiments.

In the voice device 100, an infrared module may be included as described above for short-distance low-power communication with the processing device. In other embodiments, it may further include an infrared module for applying operations to the target device, or multiplex the above-mentioned infrared communication module as a command application module. To this end, the infrared module may be used to perform infrared output on the command received by the communication module 120 from the processing device, so as to implement the operation of the target device corresponding to the command. The above-mentioned operations via infrared output are particularly suitable for traditional target devices, such as televisions and air conditioners controlled by infrared remote controllers. The infrared module used as an infrared remote control can generate and send an infrared signal corresponding to the receiving frequency band of the device based on the specific device targeted by the operation. Therefore, the voice device 100 can also be used as a universal infrared remote control. In other embodiments, the voice device can also implement control of the target device based on other technologies (for example, multiplexing Bluetooth communication devices).

In one embodiment, the voice device 100 further includes a power supply module, including but not limited to: a wireless charging component; a battery component; and a USB socket. Since the power consumption of the voice collection and transmission function of the voice device 100 is extremely small, the power consumption of the voice device 100 is relatively small, and it is suitable for adopting a power supply structure without a power cord. This greatly improves the portability and flexibility of the voice device 100 itself.

Since the voice device 100 of the present application is suitable to be implemented in a form that does not require a direct power cord, the voice device 100 may also include an attachment mechanism for attaching to a wall, a processing device, a target device, or other facility surfaces. Figure 2 shows an example implemented as a smart voice sticker. As shown in FIG. 2, the voice device 100 of the present application may include, for example, a magnet or an electrostatic adsorption surface as the attachment mechanism 200, so as to be easily adsorbed on the surface of other facilities, such as the appliance (for example, a microwave oven or oven) in the figure. Side wall.

In order to further reduce power consumption and avoid misoperation, the voice device 100 of the present application also preferably includes a remote wake-up function. Here, "remote wake-up" refers to the way that a voice device can be awakened by a specific voice wake-up word. For example, a commercially available Tmall wizard can be awakened with the wake word "Tmall wizard". Specifically, the voice device 100 may further include a wake-up module for recognizing the wake-up word from the voice input from the user. The communication module 120 can correspondingly send the collected voice input to the processing device after the wake-up module recognizes the wake-up word. For example, in the case where the wake-up word is only used to wake up and does not include other instructions, the communication module 120 may receive and transmit the voice input of the user after the wake-up word is spoken. In the case where the wake-up word also includes instructions, the communication module 120 can receive the user's wake-up word itself and subsequent voice input and transmit it. Since the wake-up module can be realized by a limited miniaturization and low-power DSP (digital signal processing) circuit, the addition of the far-field wake-up function will not substantially affect the miniaturization and low power consumption characteristics of the voice device 100.

In a preferred embodiment, the voice device 100 of the present application may further include a speaker for voice output of the content received by the communication module from the processing device. Therefore, through the introduction of speakers, it is possible to conduct further voice interaction with users. The content of the voice output by the speaker includes at least one of the following: the statement content of the execution command; and the interactive content, for example, the interactive content with the user, to further obtain the missing semantic elements. The voice interaction will be described in detail in the following description with reference to the voice control system.

In order to control the volume of the voice device 100, the speaker can be implemented with miniaturized integrated components. Further, the voice device 100 may also include an external speaker connection module, so that the connected external speaker performs voice output on the content received by the communication module from the processing device, thereby providing better quality than the built-in speaker. Audio output. In one embodiment, the external speaker connection module can be connected to the external speaker via a wired or wireless connection. The external speaker connection module may be a Bluetooth communication module. In other words, the Bluetooth communication module used to communicate with the processing device can also be reused as an external speaker connection module. As an alternative or supplement, the external speaker connection module may be a wired connection module including an audio interface, for example, a 3.5mm jack used to connect to a traditional speaker.

Figure 3 shows an example of the composition of an intelligent voice sticker in this application. As shown in Figure 3, the voice device implemented as a smart voice sticker 300 may include a MIC 310, a communication module implemented as a Bluetooth and/or infrared short-range communication module (BT/IR) 320, a battery 330, and a speaker (SPK) 340. The smart voice sticker 300 may also have an attachment structure suitable for attaching to any suitable attachment surface (as shown in FIG. 3), for example. In other embodiments, the communication module 320 may also include a Zigbee communication module.

Specifically, the MIC 310 converts the received user voice into an electric signal, and sends the above-mentioned electric signal carrying the user's voice information to the BT/IR module 320. The BT/IR module 320 sends the user's voice input data to the processing device, so as to use the processing device and the cloud to realize semantic recognition and corresponding operation command generation. Subsequently, the BT/IR module 320 may also obtain the content data that the cloud desired voice sticker 300 outputs to the user through the processing device, for example, data for further interaction with the user or reporting the operation result. The BT/IR module 320 may send the above-mentioned electric signal including the cloud content information to the SPK 340, and the latter performs an electro-acoustic conversion and voice report to the user.

In different embodiments, TTS (speech synthesis) can be implemented by different subjects. For example, the cloud can directly issue the data via the TTS, or the processing device or the smart voice sticker 300 can include the aforementioned TTS module. In one embodiment, for the consideration of transmission efficiency and low power consumption and miniaturization of the smart voice sticker, it is preferable that the processing device performs voice synthesis based on the content issued by the cloud, and then transmits the signal containing the above-mentioned voice synthesis to the voice sticker 300 , The BT/IR module 320 transmits the above-mentioned information as an electrical signal for the SPK to directly perform electro-acoustic conversion.

Figure 4 shows an example of the circuit processing flow of a voice device of the present application. As shown in FIG. 4, the microphone array receives external voice input (for example, the user's voice), and wake-up word recognition is performed by a DSP (digital signal processor) with AEC (Echo Cancellation) function. After the wake word is recognized, the collected voice is sent via the system-on-chip with WiFi function or its integrated infrared or Bluetooth (or BLE/mesh), and the subsequent voice output data (for example, command execution statement or interaction problem) . The received voice data is then processed by the codec (codec) for TTS processing, and then output through the speaker. During this period, the power supply supplies energy for the various functional modules of the voice device.

As mentioned above, the voice device of the present application can be combined with the processing device to obtain a voice kit for implementing voice collection and networking functions required for local operations. Figure 5 shows a schematic diagram of the composition of a voice kit according to an embodiment of the present application. As shown in FIG. 5, the voice kit 500 may include the voice device 510 and the processing device 520 described above with reference to FIGS. 1 to 4.

The processing device 520 includes a communication unit 521 communicatively connected to the voice device 510, and the processing device 520 receives the voice data collected by the voice device via its communication unit 521, so as to realize the semantic recognition of the voice input and the corresponding semantic recognition Operation of the target device. The communication unit 521 may be adapted to communicate in a communication mode corresponding to the communication module 511 of the small voice device 510, such as low-power short-distance communication. In an embodiment, the aforementioned communication unit 521 is adapted to communicate with the corresponding communication module 511 with Bluetooth, Zigbee and/or infrared technology.

Specifically, the above-mentioned semantic recognition and the generation and issuance of operation commands can be implemented in the cloud. To this end, the processing device 520 may include a networking unit 522 for uploading the voice data from the user received from the voice device 510 to a server in the cloud. The networking unit 522 is, for example, a module that uses WiFi and/or mobile communication technologies such as 4G and 5G to access the Internet. The server may perform semantic recognition for the voice input to generate and issue operation commands for the operation of the target device corresponding to the recognition semantics.

In some embodiments, the generation and issuance of the aforementioned semantic recognition and operation commands may be implemented locally. Therefore, the server may include: a local server that communicates with the voice kit in close range, the local server is used to perform semantic recognition for at least part of the voice input, and generate operation commands for recognizing the operation of the target device corresponding to the semantics And issue the operation command. For example, the local server can be a smart speaker used as a home smart processing terminal. As a result, the processing speed of voice commands is improved through the local server. In one embodiment, the above-mentioned local server can be connected to a cloud server to form the "server" in this application as a whole.

Based on different application scenarios, the kit may include multiple voice devices arranged in different areas, and each processing device is communicatively connected with one or more of the miniaturized devices within the communication range. For example, in a home scene, different voice stickers may be arranged in different rooms (for example, living room, bedroom, bathroom, and kitchen). The multiple voice stickers 510 may be connected to one processing device 520 within the Bluetooth communication range. And the only processing device 520 in the kit realizes networking with the cloud. In other embodiments, such as in a company scenario, multiple processing devices may be included in the kit to be responsible for a larger number of voice stickers. In other embodiments, the processing device 520 may also be used as an external module of the target device (for example, a smart home appliance) to perform networking operations for the target device itself.

According to different control scenarios, the user's voice input can control the target device based on different approaches. For example, in different embodiments, the target device may directly receive the issued operation command from the server and execute the operation corresponding to the operation command; and/or the processing device 520 receives the operation issued by the server via its networking unit 522 Commands are issued to the target device by itself or via the voice device 510. The target device that directly receives the issued operation command from the server and executes it may include its own networked smart home appliance. The target device that obtains the operation command via the processing device 520 or the voice device 510 may include traditional home appliances.

For example, all the smart home appliances running in the home have been connected to the control server. At this time, the operation command for the smart home appliance (for example, lowering the temperature of the refrigerator compartment) can be directly issued by the control server. For traditional home appliances that need to be controlled by corresponding infrared codes, the server can generate corresponding operation commands based on semantic recognition (for example, turn off the air conditioner), find the infrared operation code of the air conditioner, and send the above commands directly to the above-mentioned universal infrared remote control Voice device of the device. In other embodiments, the encoding of the above infrared operation may also be implemented locally, for example, implemented at a processing device or a voice device.

In the case that the voice device 510 includes a speaker for voice output, the processing device 520 can also use its networking unit 522 to receive the interactive content with the user issued by the server, and the communication unit 521 is also used to send the interactive content with the user. Give the voice device for voice output. The content of the above interaction can be the confirmation of the operation of the device (for example, "the light is turned on"), the acquisition of the necessary semantic elements (for example, the recognition of the user's "light on" voice input and there is more than one light in the range In this case, further ask "Which light to turn on"), or a combination of the above two (for example, "The TV is turned on, which channel needs to be watched), etc.

In an embodiment, the processing device 520 itself may have simple voice recognition and command generation and issuance functions. To this end, the processing device 520 may include: a voice recognition unit for performing semantic recognition for the voice input; and an operation command generating unit for generating operation commands for recognizing the operation of the target device corresponding to the semantics. As a result, the suite of the present application can not only realize the understanding of complex semantics by connecting to the server of the cloud, but also quickly respond to simple input.

In one embodiment, the processing device 520 itself may be a smart speaker connected to a cloud server, or another device that also has a voice collection function. Here, the voice device 510 can be used as an entrance to help the smart speaker to collect voices in different areas (for example, kitchen or bathroom areas where it is not convenient to directly perform voice operations on the smart speakers arranged in the living room).

Further, the above kit can be combined with the server to realize a voice control system. Fig. 6 shows a schematic diagram of the composition of a voice control system according to an embodiment of the present application. As shown in FIG. 6, the system 600 may include a plurality of voice packages 610 and server 620 as described above. Here, the server 620 may refer to a group of servers that provide specific functions.

Each voice kit 610 may include at least one voice device and at least one processing device, for example, the illustrated three voice devices and one processing device, and the voice device and the processing device are connected via short-distance low-power communication means (for example, , The illustrated BT) to communicate. Each voice device can be used for voice collection and transmission in different areas, for example.

Each voice kit 610 is connected to the server 620 through the networking function of the processing device (for example, a WiFi module). The server 620 may perform semantic recognition on the voice input uploaded by the processing device to generate and issue operation commands for recognizing the operation of the target device corresponding to the semantics.

In one embodiment, the server itself can implement all operations such as semantic recognition, operation command generation and issuance, etc. Thus, the server 620 may include: a semantic processing server for semantically recognizing the uploaded voice input; a command generating server for generating operation commands for target recognition operations based on the recognized semantics; and a server for issuing operation commands The command is issued to the server.

In another embodiment, the server 620 may be used only for semantic recognition, or for generating and issuing operation commands for some target devices. At least the control of the target device in this application can be implemented by an external server. This is especially applicable when a service provider of a certain brand provides remote control functions for their smart devices. Thus, the server 620 may include a semantic processing server for semantic recognition of the uploaded voice input, and the server sends the recognized semantics to an external server, wherein the external server is based on The recognized semantics generates a command generation server for the operation command for the target recognition operation, and the command issuance server for issuing the operation command.

In an embodiment, the server 620 may obtain in advance at least one piece of local device configuration information as follows. The local device configuration information may include: the distribution and device information of the voice device, the processing device and/or at least part of the target device itself; and the correspondence between at least two of the voice device, the processing device and the target device. In this way, the server 620 can also automatically complete the semantic elements that are missing in the recognition semantics for performing operations on the target device based on the local device configuration information. For example, based on the location of the bathroom where the voice sticker is located, it is clear that "turning on the light" is the only ceiling light in the bathroom.

In addition, this application can also be implemented as a voice control method. Fig. 7 shows a schematic flowchart of a voice control method according to an embodiment of the present application. The method can be implemented by the above voice device, kit and system.

In step S710, the voice device (for example, a miniaturized voice sticker) collects voice input. In step S720, the voice device sends the voice input to the processing device. In an embodiment, the above-mentioned transmission may be, for example, short-distance communication based on infrared, Bluetooth and/or Zigbee.

In step S730, the processing device implements semantic recognition for the voice input and generation of corresponding target device operation commands. In different embodiments, the foregoing operations for semantic recognition and command generation can be completed by different objects, for example, can be completed by the processing device itself, a local server, a remote server, or any combination thereof.

Therefore, in one embodiment, in step S730, the processing device may send the voice input to the server, and use the semantic recognition of the voice input by the server to obtain the operation command of the target device operation corresponding to the semantic recognition.

In another embodiment, in step S730, the server may include a local server, and the local server may perform semantic recognition for at least part of the voice input, generate operation commands for identifying the operation of the target device corresponding to the semantics, and issue all The operation command.

In another embodiment, in step S730, the processing device performs semantic recognition for at least part of the voice input, and generates an operation command for recognizing the operation of the target device corresponding to the semantics.

In an embodiment, the voice control method of the present application may further include: using the voice device and/or the processing device to obtain an operation command for identifying the operation of the target device corresponding to the semantics; and the voice device and/or the processing device The processing device issues the operation command to the target device.

In one embodiment, the voice control method of the present application may further include: the processing device obtains and delivers the voice output content; the processing device delivers the voice output content to the voice device; and the voice device voice Output the voice output content.

In an embodiment, the voice control method of the present application may further include: the voice device recognizes the wake-up word from the voice input from the user. Therefore, step S720 may include: sending the collected voice input to the processing device after the wake word is recognized.

In an embodiment, the voice control method of the present application may further include: a server and/or an external server that has acquired the semantic recognition generates and issues operations of the target device operation corresponding to the semantic recognition based on the semantic recognition command.

In an embodiment, the voice control method of the present application may further include: the target device receives an operation command issued by the server and/or the external server and executes an operation corresponding to the operation command. Correspondingly, the target device receiving the operation command issued by the server and/or the external server and performing the operation corresponding to the operation command may include at least one of the following: the target device receives the operation command from the service Terminal or the external server directly receives the issued operation command; the processing device receives the operation command issued by the server via the networking unit, and issues the issued operation command to the target device by itself or via the voice device The operation command.

In one embodiment, the voice control method of the present application may further include: the server obtains the execution result of the operation command; the server generates the statement content of the execution command based on the execution result and sends it to The processing device; the processing device issues the statement content of the execution command to the voice device; and the voice device outputs the statement content of the execution command by voice.

In one embodiment, the voice control method of the present application may further include: the server generates content interacting with the user based on the semantic recognition and sends it to the processing device; and the processing device sends the content to the voice The device delivers the content of interaction with the user; and the voice device vocally outputs the content of interaction with the user.

In an embodiment, the voice control method of the present application may further include: the server automatically fills in the semantic elements that are missing in the operation on the target device in the recognition semantics based on the local device configuration information obtained in advance.

In one embodiment, the server involved in the voice control method of the present application may be a local server, so the local server may perform semantic recognition for at least part of the voice input to generate target device operations corresponding to the recognition semantics And issue the operation command.

Figure 8 shows an example of voice control in this application. As shown in Figure 8, in the local collection and upload stage, the voice device in the voice kit (and, in some embodiments, the voice collection module in the processing device) detects voice command input from the user or smart device, and the processing device Preliminary processing can be performed on the collected voice, for example, ASR (speech recognition) pickup, and the picked up voice command is transmitted to the cloud. In the cloud processing stage, the server can perform subsequent processing on the picked-up voice commands, such as NIP (Natural Voice Processing) and NIU (Natural Voice Understanding), and perform command analysis and TTS output according to the processing results. In the local processing stage, the parsed command can be directly transmitted to the target device for execution (for example, the smart device directly executes the command parsed by the cloud), or it can be executed by the target device after conversion by a processing device or a voice device, for example Commands (for example, infrared commands issued by a voice device for traditional household appliances). In addition, when there is audio output, the voice device can perform voice output through its own speaker or externally connected Bluetooth or traditional speakers.

The voice device, voice kit, and voice control system according to the present application have been described in detail above with reference to the accompanying drawings. The existing technology that incorporates the voice module into the device has a long development cycle and high cost, and requires acoustic design and debugging for each device. In contrast, the modular solution of the present application can be easily integrated into any required equipment. In addition, the kit of the present application is convenient to use, and the miniaturized and portable design of the voice module makes it easy to be placed on various surfaces and suitable for use in multiple scenarios. Furthermore, the smart voice sticker of the present application can also realize the function of a universal infrared remote control by adding an infrared module, thereby realizing intelligent control of traditional infrared devices.

In addition, the method according to the present application can also be implemented as a computer program or computer program product. The computer program or computer program product includes computer program code instructions for executing the above-mentioned steps defined in the above-mentioned method of the present application.

Alternatively, this application can also be implemented as a non-transitory machine-readable storage medium (or computer-readable storage medium, or machine-readable storage medium), on which executable code (or computer program, or computer instruction code) is stored ), when the executable code (or computer program, or computer instruction code) is executed by the processor of the electronic device (or computing device, server, etc.), the processor is caused to execute each of the above methods according to the application step.

Those skilled in the art will also understand that the various exemplary logic blocks, modules, circuits, and algorithm steps described in conjunction with the disclosure herein can be implemented as electronic hardware, computer software, or a combination of both.

The flowcharts and block diagrams in the accompanying drawings show the possible implementation architecture, functions, and operations of the system and method according to multiple embodiments of the present application. In this regard, each block in the flowchart or block diagram may represent a module, program segment, or part of the code, and the module, program segment, or part of the code contains one or more functions for realizing the specified logical function. Executable instructions. It should also be noted that in some alternative implementations, the functions marked in the block may also occur in a different order than marked in the drawings. For example, two consecutive blocks can actually be executed in parallel, or they can sometimes be executed in the reverse order, depending on the functions involved. It should also be noted that each block in the block diagram and/or flowchart, and the combination of the blocks in the block diagram and/or flowchart, can be implemented by a dedicated hardware-based system that performs the specified functions or operations Or it can be realized by a combination of dedicated hardware and computer instructions.

The embodiments of the present application have been described above, and the above description is exemplary and not exhaustive, and is not limited to the disclosed embodiments. Without departing from the scope and spirit of the described embodiments, many modifications and changes are obvious to those of ordinary skill in the art. The choice of terms used herein is intended to best explain the principles, practical applications, or improvements to technologies in the market of the embodiments, or to enable other ordinary skilled in the art to understand the embodiments disclosed herein.

Claims (43)

1. A voice control system includes a plurality of voice kits and a server communicating with the voice kit, wherein the voice kit includes:

   Voice device for collecting voice input and sending the collected voice input to the processing device;

   The processing device is used to receive the voice input collected by the voice device, and upload the voice input to the server, and

   The server end is used to perform semantic recognition on the voice input sent by the processing device, so as to generate and issue operation commands for the operation of the target device corresponding to the recognition semantics.
2. 8. The system of claim 1, wherein the voice device sends the collected voice input to the processing device via short-range communication.
3. The system of claim 2, wherein the voice kit includes a plurality of the voice devices arranged in different areas, and each processing device communicates with one or more of the voice devices within the short-range communication range connection.
4. 3. The system of claim 2, wherein the voice device and the processing device each comprise a low-power short-range communication module for performing short-range communication with each other, and the communication module includes at least one of the following:

   A Bluetooth communication module that communicates with the processing device based on Bluetooth technology;

   An infrared communication module based on infrared technology to communicate with the processing device;

   A Zigbee communication module based on Zigbee technology to communicate with the processing device.
5. 3. The system of claim 1, wherein the processing device uploads the voice input to the server via remote communication.
6. 7. The system of claim 5, wherein the processing device further comprises a WiFi module for remote communication with the server.
7. The system of claim 1, wherein the voice device recognizes a wake-up word from the collected voice input, and sends the collected voice input to the processing device after the wake-up module recognizes the wake-up word .
8. The system according to claim 1, wherein the voice device outputs the content received from the processing device and/or the target device via its own speaker or a wired/wireless external speaker.
9. The system according to claim 8, wherein the content for voice output includes at least one of the following:

   The statement of the execution order; and

   Interaction content with the user.
10. The system according to claim 1, wherein the voice device is also used to perform infrared output on the command received by the processing device, so as to realize the operation of the target device corresponding to the command.
11. The system of claim 1, wherein the voice device further comprises an attachment mechanism for attaching to a wall surface, a processing device, a target device, or a surface of other facilities.
12. The system according to claim 1, wherein the processing device is configured to perform semantic recognition for at least part of the voice input, and generate operation commands for identifying the operation of the target device corresponding to the semantics.
13. The system of claim 1, wherein:

    The target device directly receives the issued operation command from the server and executes the operation corresponding to the operation command; and/or

    The processing device receives the operation command issued by the server, and issues the operation command to the target device by itself or via the voice device.
14. The system of claim 13, wherein:

    The target device that directly receives and executes the issued operation command from the server includes a networked smart home appliance; and

    The target device that obtains the operation command via the processing device or the voice device includes a traditional home appliance.
15. The system of claim 1, wherein the server includes:

    Semantic processing server for semantic recognition of uploaded voice input;

    Based on the recognized semantics, a command generation server for generating operation commands for target recognition operations; and

    The command for issuing the operation command is issued to the server.
16. The system of claim 1, wherein the server includes:

    A local server for short-range communication with the voice kit, the local server is used to perform semantic recognition for at least part of the voice input, generate operation commands for recognizing the operation of the target device corresponding to the semantics, and issue the operation commands .
17. The system of claim 1, wherein the server includes:

    Semantic processing server for semantic recognition of uploaded voice input, and

    The server sends the identified semantics to the external server, where

    The external server generates a command generation server for an operation command for the target recognition operation based on the recognized semantics, and a command issuance server for issuing the operation command.
18. The system according to claim 1, wherein the server obtains in advance at least one of the following local device configuration information:

    The distribution and device information of the speech device, the processing device and/or at least part of the target device itself;

    Correspondence between at least two of the speech device, the processing device and the target device.
19. The system of claim 18, wherein the server automatically fills in the semantic elements that are missing in the recognition semantics for performing operations on the target device based on the local device configuration information.
20. A voice kit including:

    Voice device for collecting voice input and sending the collected voice input to the processing device;

    The processing device includes a communication unit communicatively connected with the voice device, and the processing device receives the voice data collected by the voice device via the communication unit, so as to realize the semantic recognition of the voice input and the corresponding semantic recognition Operation of the target device.
21. 22. The kit of claim 20, wherein the voice device and the processing device each include a low-power short-range communication module for performing short-range communication with each other, and the communication module includes at least one of the following:

    A Bluetooth communication module that communicates with the processing device based on Bluetooth technology;

    An infrared communication module based on infrared technology to communicate with the processing device; and

    A Zigbee communication module based on Zigbee technology to communicate with the processing device.
22. The kit of claim 20, wherein the processing device further comprises:

    The networking unit is configured to upload the voice data from the user received from the voice device to the server, where the server performs semantic recognition for the voice input to generate and issue the corresponding semantic recognition Operation command for target device operation.
23. The kit of claim 20, wherein the processing device comprises:

    A voice recognition unit for performing semantic recognition for the voice input; and

    The operation command generation unit is used to generate operation commands for identifying the operation of the target device corresponding to the semantics.
24. The kit according to claim 20, wherein the processing device receives the operation command issued by the server via the networking unit, and issues the operation command to the target device by itself or via the voice device.
25. The kit according to claim 24, wherein the networking unit is further used for:

    Receiving the interactive content issued by the server, and

    The communication unit is also used for:

    The interactive content is sent to the voice device for voice output.
26. The kit according to claim 20, wherein the voice kit includes a plurality of the voice devices arranged in different areas, and each processing device is communicatively connected with one or more of the voice devices within a communication range.
27. A voice device, including:

    Microphone, used to collect voice input;

    The communication module is used to send the collected voice input to the processing device, so as to realize the semantic recognition of the voice input and the operation of the target device corresponding to the semantic recognition through the processing device.
28. The speech device according to claim 27, further comprising:

    Wake-up module, used to recognize wake-up words from voice input, and

    The communication module is used for sending the collected voice input to the processing device after the wake-up module recognizes the wake-up word.
29. The speech device according to claim 27, further comprising:

    The speaker is used for voice output of the content received by the communication module from the processing device and/or the target device.
30. The device according to claim 29, wherein the content of the voice output includes at least one of the following:

    The statement of the execution order; and

    Interaction content with the user.
31. The speech device according to claim 27, further comprising:

    An external speaker connection module connected to an external speaker via a short-range wireless connection and/or a wired connection, so that the external speaker performs processing on the content received by the communication module from the processing device and/or the target device Voice output.
32. The speech device according to claim 27, further comprising:

    The infrared module is used to perform infrared output on the command received by the communication module from the processing device, so as to realize the operation of the target device corresponding to the command.
33. The voice device according to claim 27, wherein the communication module is a low-power short-distance communication module and includes at least one of the following:

    A Bluetooth communication module that communicates with the processing device based on Bluetooth technology;

    An infrared communication module based on infrared technology to communicate with the processing device; and

    A Zigbee communication module based on Zigbee technology to communicate with the processing device.
34. The speech device according to claim 27, further comprising:

    A power supply module, the power supply module includes at least one of the following:

    Wireless charging components;

    Battery pack

    USB socket.
35. The speech device according to claim 27, further comprising:

    An attachment mechanism for attaching to the wall, processing device, target equipment, or other facility surface.
36. A voice control method, including:

    Voice device collects voice input;

    The voice device sends the voice input to the processing device;

    The processing device implements semantic recognition for the voice input and generation of corresponding target device operation commands.
37. The method of claim 36, wherein the voice device sending the voice input to the processing device comprises:

    The voice device sends the voice input to the processing device via short-range communication.
38. 36. The method according to claim 36, wherein the processing means for realizing the semantic recognition of the voice input and generating the corresponding target device operation command comprises:

    The processing device uploads the voice input to the server;

    The server performs semantic recognition of the voice input to obtain operation commands for the operation of the target device corresponding to the recognition semantics.
39. The method of claim 38, wherein the processing device uploading the voice input to the server comprises:

    The processing device uploads the voice input to a local server via short-range communication.
40. The method of claim 39, wherein the semantic recognition of the voice input by the server to obtain the operation command of the target device operation corresponding to the recognition semantics comprises:

    The local server performs semantic recognition for at least part of the voice input, generates an operation command for identifying the operation of the target device corresponding to the semantics, and issues the operation command.
41. 36. The method according to claim 36, wherein the processing means for realizing the semantic recognition of the voice input and generating the corresponding target device operation command comprises:

    The processing device performs semantic recognition for at least part of the voice input, and generates an operation command to recognize the operation of the target device corresponding to the semantics.
42. The method of claim 36, further comprising:

    Use the voice device and/or the processing device to obtain the operation command for the operation of the target device corresponding to the recognition semantic; and

    The voice device and/or the processing device issue the operation command to the target device.
43. The method of claim 36, further comprising:

    The processing device acquires and delivers voice output content;

    The processing device delivers the voice output content to the voice device; and

    The voice device voice outputs the voice output content.

Images