Voice control method and device, electronic equipment and computer readable storage medium

Abstract

The invention provides a voice control method and device, and relates to the technical field of computers, in particular to the technical fields of voice processing, deep learning, image processing and the like. According to the specific implementation scheme, the method comprises the steps of receiving voice information of a first voice area in at least two voice areas; detecting whether the voice information is matched with the intention inheritance key information; in response to detection that the voice information is matched with intention inheritance key information, detecting whether inheritable intention information of a second voice region is provided or not, and the second voice region is a voice region different from the first voice region; and in response to the detected inheritable intention information, the controlled object in the first sound area is controlled to execute the instruction action, so that the user experience is improved.

Description

Technical Field

[0001] The present disclosure relates to the field of computers, specifically to technical fields such as speech processing, deep learning, and image processing, and in particular to a speech control method and device, an electronic device, a computer-readable storage medium, and a computer program product. Background Art

[0002] Currently, the mainstream cockpit voice systems on the market all support multi-zone interaction, such as dual-zone, four-zone, six-zone, etc. At the same time, there are many functions supporting the sound zones, such as zoned air conditioning, zoned vehicle control, system control, etc.

[0003] The functions supporting each sound zone are implemented independently. When users control the components of the corresponding sound zone, they need to wake up with keywords and then recognize fixed instructions. The control algorithm has a large amount of calculation and the user experience is poor. Summary of the invention

[0004] The present disclosure provides a voice control method and apparatus, an electronic device, a computer-readable storage medium, and a computer program product.

[0005] According to a first aspect, a voice control method is provided, the method comprising: receiving voice information of a first sound zone among at least two sound zones; detecting whether the voice information matches intention inheritance key information; in response to detecting that the voice information matches the intention inheritance key information, detecting whether there is inheritable intention information of a second sound zone among at least two sound zones, the second sound zone being a sound zone different from the first sound zone; in response to detecting that there is inheritable intention information, controlling a controlled object in the first sound zone to execute a command action.

[0006] According to a second aspect, a voice control device is provided, which includes: a receiving unit, configured to receive voice information of a first sound zone among at least two sound zones; a key detection unit, configured to detect whether the voice information matches intention inheritance key information; an intention detection unit, configured to detect whether there is inheritable intention information of a second sound zone among at least two sound zones in response to detecting that the voice information matches the intention inheritance key information, the second sound zone being a sound zone different from the first sound zone; and an execution unit, configured to control a controlled object in the first sound zone to execute an instruction action in response to detecting that there is inheritable intention information.

[0007] According to a third aspect, an electronic device is provided, comprising: at least one processor; and a memory communicatively connected to the at least one processor, wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can execute the method described in any implementation manner of the first aspect.

[0008] According to a fourth aspect, a non-transitory computer-readable storage medium storing computer instructions is provided, where the computer instructions are used to cause a computer to execute the method described in any implementation of the first aspect.

[0009] According to a fifth aspect, a computer program product is provided, comprising a computer program, which implements the method described in any implementation manner of the first aspect when executed by a processor.

[0010] The voice control method and device provided by the embodiments of the present disclosure firstly receive voice information of the first sound zone of at least two sound zones; secondly, detect whether the voice information matches the key information of intention inheritance; thirdly, in response to detecting that the voice information matches the key information of intention inheritance, detect whether there is inheritable intention information of the second sound zone, the second sound zone is a sound zone different from the first sound zone; finally, in response to detecting that there is inheritable intention information, control the controlled object of the first sound zone to execute the command action. Thus, when the voice information of the first sound zone matches the key information of intention inheritance, if the inheritable intention information of the second sound zone is detected, the inheritable intention information is directly executed, so that the user of the first sound zone does not need to issue a control instruction that matches the command action, which simplifies the user voice control process, saves the calculation amount of the control algorithm, and improves the user experience.

[0011] It should be understood that the content described in this section is not intended to identify the key or important features of the embodiments of the present disclosure, nor is it intended to limit the scope of the present disclosure. Other features of the present disclosure will become easily understood through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0012] The accompanying drawings are used to better understand the present solution and do not constitute a limitation of the present disclosure.

[0013] Figure 1 is a flow chart of an embodiment of the voice control method according to the present disclosure;

[0014] Figure 2 is a flowchart of the information processing of the first and second sound zones of the present disclosure;

[0015] Figure 3 is a structural schematic diagram of an embodiment of a voice control device according to the present disclosure;

[0016] Figure 4 It is a block diagram of an electronic device used to implement the voice control method of the embodiment of the present disclosure. DETAILED DESCRIPTION

[0018] The following is a description of exemplary embodiments of the present disclosure in conjunction with the accompanying drawings, including various details of the embodiments of the present disclosure to facilitate understanding, which should be considered as merely exemplary. Therefore, it should be recognized by those of ordinary skill in the art that various changes and modifications may be made to the embodiments described herein without departing from the scope and spirit of the present disclosure. Similarly, for the sake of clarity and conciseness, descriptions of well-known functions and structures are omitted in the following description.

[0019] At present, more and more vehicle cockpit projects have integrated voice modules to achieve better human-computer interaction functions. The voice modules include keyword detection (wake-up) sub-module, recognition detection (human voice detection + human voice recognition) sub-module, TTS broadcast (voice broadcast) sub-module, sound zone isolation sub-module, etc. Among them, the keyword detection (wake-up) sub-module is used to detect specific keywords to facilitate users to enable subsequent voice interaction functions, such as wake-up, global scenario command words, and keyword-free wake-up word technology.

[0020] At present, mainstream cockpit voice systems all support multi-zone voice control, that is, each zone is equipped with a voice module, so as to realize the human-computer interaction function of each zone. If you want to realize the intention inheritance scheme across zones, the existing scheme needs to be awakened by keywords and recognize fixed commands to realize the intention inheritance function. For example, after the driver wakes up and says "XX, open the window", if the co-driver also wants to open the same intention command, he must be awakened by the co-driver and then say "XX, open the window" to realize the same intention command as the driver. This operation is cumbersome and may interrupt the continuous output of the driver. That is, if passengers in other zones want to inherit the intention of a certain zone, the passengers in this zone need to be awakened again and output the same command. The function is cumbersome, and one more recognition engine will be opened at the same time, and the computing cost is high.

[0021] The present disclosure proposes a voice control method, which can realize the inheritance of voice control intentions between multiple voice zones, improve user experience, Figure 1 A process 100 according to an embodiment of the voice control method of the present disclosure is shown. The voice control method comprises the following steps:

[0022] Step 101: receiving voice information of a first audio zone among at least two audio zones.

[0023] In this embodiment, the sound zone is a sound receiving area. The sound zones are divided differently for different scenarios. For example, in a vehicle-mounted scenario, in order to accurately perform voice recognition and interaction, multiple sound zones are usually divided in the vehicle. First, identify in which sound zone the passenger who wakes up the vehicle terminal is located, and then recognize and interact with the passenger's voice, so as to reduce the impact of voice data from other sound zones on voice recognition.

[0024] In this embodiment, the first sound zone is one of the at least two sound zones. In order to effectively distinguish the sound zone currently being analyzed from the others of the at least two sound zones, the sound zone currently being analyzed is taken as the first sound zone. For example, in a vehicle-mounted scenario, the at least two sound zones include: a main driving sound zone and a co-driving sound zone. The first sound zone is a main driving sound zone, or the first sound zone is a co-driving sound zone.

[0025] In this embodiment, the voice information of the first sound zone is the voice information emitted by the user in the first sound zone. The voices emitted by users in different sound zones of at least two sound zones carry position information. The position information carried in the voices of each user can be effectively used to determine which sound zone the voice belongs to. For the current scene, when the voice information of the first sound zone includes wake-up information for waking up the execution subject on which the voice control method runs and control information for controlling the controlled object of the current scene, the execution subject can be woken up to control the controlled object.

[0026] In the technical solution disclosed in this disclosure, the collection, storage, use, processing, transmission, provision and disclosure of voice information are carried out after authorization and in compliance with relevant laws and regulations. The information related to the user in the voice information is obtained after the user's permission, and the information related to the user is processed under confidentiality conditions.

[0027] Step 102: Detect whether the voice information matches the intended inheritance key information.

[0028] In this embodiment, the key information for intention inheritance is information that represents the need to inherit the inheritable intention information of the second sound zone among at least two sound zones. The key information for intention inheritance is information pre-set in order to inherit the intention of other sound zones among at least two sound zones. When the voice information matches the key information for intention inheritance, it is determined that the subject that sends the voice information has the intention to inherit the inheritable intention information of other sound zones among at least two sound zones. For example, the key information for intention inheritance is: I want it too, wherein the second sound zone is the sound zone other than the first sound zone among the at least two sound zones, and the inheritable intention information of the second sound zone is the intention to control the controlled object of the second sound zone. The intention can be information obtained by performing intent analysis on the instruction. The key information for intention inheritance can appear through multiple modal expressions. For example, the key information for intention inheritance can be an intention inheritance voice segment. The intention inheritance voice segment is a voice segment that includes the user's intention inheritance demand. The voice segment is the voice of "I want it too" issued by the user.

[0029] In this embodiment, the voice information matches the intention-inheritance key information means that: the voice information contains information related to the intention-inheritance key information, and when the intention-inheritance key information is an intention-inheritance voice segment, the above step 102 may include: in response to detecting that the voice information matches the intention-inheritance voice segment, determining that the voice information matches the intention-inheritance key information.

[0030] Step 103 , in response to detecting that the voice information matches the intention inheritance key information, detecting whether there is inheritable intention information of the second sound zone.

[0031] In this embodiment, the second sound zone is a sound zone different from the first sound zone. The inheritable intention information of the second sound zone means that the intention information of the second sound zone to control the controlled object of the sound zone can be inherited by the first sound zone, so that the inheritable intention information can be applied to the control of the same type of controlled objects in the first sound zone.

[0032] In this embodiment, the inheritable intent information of the second audio zone is the intent information generated when the user voice of the user of the second audio zone controls the controlled object of the second audio zone. The second audio zone may be one of the at least two audio zones, or may be multiple audio zones of the at least two audio zones. It should be noted that the content of the inheritable intent information of the second audio zone may be different for different scenarios. For example, in a car scenario, the inheritable intent information of the second audio zone includes: turning on the lights of the second audio zone; in a smart home scenario, the inheritable intent information of the second audio zone includes: turning on the TV in the second audio zone.

[0033] Step 104 : In response to detecting the inheritable intention information, controlling the controlled object in the first sound zone to execute the command action.

[0034] In this embodiment, the controlled object of the first sound zone is an object controlled by the voice of the first sound zone. The controlled object can be a device in the current scene, such as a car air conditioner in a car scene, or a TV in a smart home scene; the controlled object can also be a system in the current scene, such as a car computer system in a car scene.

[0035] In this embodiment, the command action is an action generated by the controlled object due to the command, and the command action is also an action related to the inheritable intention information. For example, the inheritable intention information of the second sound zone is to control the vehicle air conditioner in the second sound zone to turn on, then the controlled object of the first sound zone is the vehicle air conditioner in the first sound zone, and a command action that the vehicle air conditioner in the first sound zone needs to execute is to turn on the air conditioner. This command action is an action realized by the execution subject on which the voice control method runs sending a command to the air conditioner.

[0036] In this embodiment, the above step 104 includes: generating an action execution intention based on the inheritable intention information, generating an instruction for the controlled object based on the action execution intention, and outputting the instruction to the controlled object so that the controlled object in the first sound zone executes the instruction action corresponding to the instruction.

[0037] As shown in Table 1, the first sound zone is the sound zone corresponding to the co-pilot driver's seat, and the second sound zone is the sound zone corresponding to the driver's seat. The inheritable intention information of the second sound zone is generated by the voice information sent to the second sound zone, and the action execution intention generated by the inheritable intention information of the second sound zone is: open the driver's seat window; after the voice information "I want it too" in the first sound zone matches the intention inheritance key information (such as the intention inheritance key information is: I want it too), the action execution intention of the first sound zone is generated: open the co-pilot driver's seat window, and the controlled object in the first sound zone is controlled to open the related instruction action of the co-pilot driver's seat window.

[0038] Table 1

[0039] Main driver Co-pilot Action execution intention Hello XX, open the window Open the driver's window I want too Open the passenger window

[0040] In this embodiment, the inheritable intention information of the second sound zone is information about the command action executed on the controlled object of the second sound zone related to the inheritable intention information. After the first sound zone inherits the inheritable intention information of the second sound zone, the controlled object of the first sound zone related to the inheritable intention information can be controlled to execute the command action of the inheritable intention information based on the controlled object corresponding to the inheritable intention information.

[0041] The voice control method provided in this embodiment can realize the intention inheritance of most scenes by matching the intention inheritance key information, so that the control of the controlled object in the inherited sound zone can be realized without going through the cumbersome post-wake-up recognition process. Since the intention inheritance key information is matched, only the intention inheritance key information needs to be added, and there is no additional algorithm calculation amount, which reduces the algorithm calculation amount compared with the traditional controlled object control method.

[0042] The voice control method provided by the embodiment of the present disclosure first receives voice information of the first sound zone of at least two sound zones; secondly, detects whether the voice information matches the key information of intention inheritance; thirdly, in response to detecting that the voice information matches the key information of intention inheritance, detects whether there is inheritable intention information of the second sound zone, the second sound zone is a sound zone different from the first sound zone; finally, in response to detecting that there is inheritable intention information, controls the controlled object of the first sound zone to execute the command action. Thus, when the voice information of the first sound zone matches the key information of intention inheritance, if the inheritable intention information of the second sound zone is detected, the inheritable intention information is directly executed, so that the user of the first sound zone does not need to issue a control instruction that matches the command action, simplifies the user voice control process, saves the amount of calculation of the control algorithm, and improves the user experience.

[0043] Optionally, before detecting whether the voice information matches the key information of intention inheritance, the above-mentioned voice control method also includes: obtaining a converted text based on the voice information; detecting whether the converted text has a wake-up word; in response to not detecting that the converted text has a wake-up word, detecting whether the voice information matches the key information of intention inheritance.

[0044] In some optional implementations of the present disclosure, the above-mentioned intention inheritance key information includes: intention inheritance keywords, and the above-mentioned detection of whether the voice information matches the intention inheritance key information includes: obtaining a converted text based on the voice information; detecting whether the converted text contains the intention inheritance keywords; in response to detecting that the converted text contains the intention inheritance keywords, determining that the voice information matches the intention inheritance key information.

[0045] In this optional implementation, the above-mentioned intention inheritance keyword can be a keyword preset for inheriting the inheritable intention information of the second audio zone, for example, the intention inheritance keyword is: I want it too.

[0046] In this optional implementation, the converted text is the text obtained after text recognition of the voice information. The voice information can be converted into converted text by voice-to-text conversion, and the converted text is matched with the intent-inheritance keywords. If the converted text matches the intent-inheritance keywords, it is determined that the converted text contains the intent-inheritance keywords.

[0047] In this optional implementation, the intention inheritance keyword is a keyword set to inherit the same wake-up control intention between any of at least two audio zones, wherein the wake-up control intention is the intention to send a control instruction to the controlled object in each audio zone after the wake-up procedure. For example, the audio zones on the vehicle include: the main driver audio zone and the co-driver audio zone. When the main driver sends: XX, please open the window, the wake-up control intention is: wake up the current execution subject through XX, and execute the control instruction to open the main driver window of the controlled object. At this time, the "I want it too" sent by the user in the co-driver audio zone is collected, and it is determined that the co-driver audio zone also needs to perform the same wake-up control intention.

[0048] The method for detecting the match between voice information and key information of intent inheritance provided by this optional implementation first converts the voice information into a translated text, and detects whether the voice information matches the key information of intent inheritance by determining whether the translated text contains the key words of intent inheritance, thereby providing a reliable means for matching the voice information with the key information of intent inheritance and improving the accuracy of matching the key information of intent inheritance.

[0049] In some optional implementations of the present disclosure, the above-mentioned intention inheritance key information includes: intention inheritance semantic information, and detecting whether the voice information matches the intention inheritance key information includes: extracting text semantic information based on the voice information; detecting whether the text semantic information matches the intention inheritance semantic information; in response to detecting that the text semantic information matches the intention inheritance semantic information, determining that the voice information matches the intention inheritance key information.

[0050] In this optional implementation, the intention inheritance semantic information is pre-set semantic information for inheriting the inheritable intention information of the second audio zone. For example, the inheritable intention information of the second audio zone is: air conditioning to increase the temperature, and the intention inheritance semantic information is: I also want to increase the temperature.

[0051] In this optional implementation, the above-mentioned extraction of text semantic information based on voice information includes: converting the voice information into converted text, extracting the semantic information of the converted text, and obtaining text semantic information, wherein the text semantic information is the semantic analysis content at the lexical level.

[0052] In this optional implementation, detecting whether the text semantic information matches the intention-inherited semantic information means: detecting whether the text semantic information is the same as the intention-inherited semantic information, and if they are the same, determining that the text semantic information matches the intention-inherited semantic information.

[0053] The method for detecting whether voice information matches intent-inherited key information provided by this optional implementation first extracts text semantic information of the voice information, and then detects whether the text semantic information matches the intent-inherited semantic information. If so, it is determined that the voice information matches the intent-inherited key information, thereby providing another reliable means for matching voice information with intent-inherited key information and improving the accuracy of matching intent-inherited key information.

[0054] In some optional implementations of the present disclosure, in response to detecting that the voice information matches the intention inheritance key information, detecting whether there is inheritable intention information for the second sound zone includes: in response to detecting that the voice information matches the intention inheritance key information, selecting the inheritable intention information with the shortest time difference with the voice information from the inheritable intention information of at least two sound zones stored locally; detecting whether the inheritable intention information with the shortest time difference with the voice information belongs to the first sound zone; in response to detecting that the inheritable intention information with the shortest time difference with the voice information does not belong to the first sound zone, determining that there is inheritable intention information for the second sound zone.

[0055] In this optional implementation, the locally stored inheritable intent information has a storage time, and the voice information has a receiving time, and selecting the inheritable intent information with the shortest time difference with the voice information from the inheritable intent information of at least two sound zones means: comparing the storage time of the inheritable intent information of at least two sound zones with the receiving time of the voice information, and selecting the inheritable intent information with a storage time closest to the receiving time of the voice information as the inheritable intent information with the shortest time difference with the voice information.

[0056] In this optional implementation, only the inheritable intention information having the shortest time difference with the voice information belongs to the second sound zone outside the first sound zone, and the inheritable intention information of the second sound zone among at least two sound zones is determined.

[0057] The method for detecting inheritable intention information of the second sound zone provided by this optional implementation method selects inheritable intention information with the shortest time difference with the voice information from the local area, and when the inheritable intention information does not belong to the first sound zone, determines the inheritable intention information of the second sound zone, thereby providing a reliable implementation method for determining the inheritable intention information of the second sound zone and improving the reliability of obtaining the inheritable intention information.

[0058] Optionally, after detecting that there is no inheritable intention information for the second sound zone in at least two sound zones, the above-mentioned voice control method also includes: detecting whether there is inheritable intention information for the first sound zone, and in response to detecting the inheritable intention information for the first sound zone, controlling the controlled object of the first sound zone to execute the command action.

[0059] In some optional implementations of the present disclosure, in response to detecting that the voice information matches the intention inheritance key information, detecting whether there is inheritable intention information for the second sound zone includes: in response to detecting that the voice information matches the intention inheritance key information, detecting whether the local database stores the action execution intention for the second sound zone; in response to detecting the action execution intention, and there is only one action execution intention; determining the action execution intention as inheritable intention information, and determining that the second sound zone has inheritable intention information.

[0060] In this optional implementation, the action execution intention is the intention information pre-stored for the current sound zone to control the controlled object of the current sound zone. The action execution intention can have multiple intention information. Different intention information corresponds to different action instructions. Each action instruction is used to control the controlled object to execute the instruction action.

[0061] The method for detecting inheritable intention information of the second sound zone provided by this optional implementation detects whether the local database stores the action execution intention of the second sound zone, and in response to detecting the action execution intention, and there is only one action execution intention; the action execution intention is determined as inheritable intention information, thereby improving the reliability of obtaining the inheritable intention information.

[0062] Optionally, in response to detecting that the voice information matches the intention inheritance key information, detecting whether there is inheritable intention information in the second sound zone also includes: in response to detecting an action execution intention, and there are multiple action execution intentions; determining the action execution intention among the multiple action execution intentions whose storage time is closest to the reception time of the voice information as inheritable intention information, and determining that the second sound zone has inheritable intention information.

[0063] In some optional implementations of the present disclosure, in response to detecting that the voice information matches the intention inheritance key information, detecting whether there is inheritable intention information for the second sound zone includes: in response to detecting that the voice information matches the intention inheritance key information, detecting whether the intention inheritance key information includes intention inheritance semantic information for the second sound zone; in response to detecting that the intention inheritance key information includes the intention inheritance semantic information for the second sound zone, determining that there is inheritable intention information for the second sound zone.

[0064] In this optional implementation, the intention inheritance semantic information of the second audio zone is pre-set semantic information used to inherit the inheritable intention information of the second audio zone. For example, the inheritable intention information of the second audio zone is: air conditioning cooling, and the intention inheritance semantic information of the second audio zone is: I also want to cool down.

[0065] In this optional implementation, when it is detected that the intention inheritance key information includes the intention inheritance semantic information of the second audio zone, it is determined that the inheritable intention information of the second audio zone must be included in the inheritable intention information of at least two audio zones.

[0066] The method for detecting inheritable intention information of the second sound zone provided by this optional implementation manner detects whether the intention inheritance key information includes the intention inheritance semantic information of the second sound zone; in response to detecting that the intention inheritance key information includes the intention inheritance semantic information of the second sound zone, the inheritable intention information of the second sound zone is determined, thereby providing a reliable implementation means for obtaining the inheritable intention information of the second sound zone and improving the reliability of obtaining the inheritable intention information.

[0067] It should be noted that the above-mentioned optional implementation methods for detecting whether there is inheritable intention information in the second sound zone are all methods of independently detecting the inheritable intention information in the second sound zone, and the execution subject on which the voice control method runs can detect the inheritable intention information in the second sound zone by executing any one of the above-mentioned optional implementation methods for detecting whether there is inheritable intention information in the second sound zone.

[0068] In some optional implementations of the present disclosure, in response to detecting inheritable intention information, controlling the controlled object of the first sound zone to execute a command action includes: generating and sending intention inquiry information based on the inheritable intention information; in response to receiving reply information to the intention inquiry information, controlling the controlled object of the first sound zone to execute the command action corresponding to the inheritable intention information and the reply information.

[0069] In this optional implementation, the intent inquiry information is question information that confirms or supplements the inheritable intent information, and the intent inquiry information can be expressed in various forms, such as voice, text, etc. The intent inquiry information can be question information returned to the sender of the voice message, for example, the voice message is: I want it too, and the intent inquiry information is: How many degrees do you want to adjust the temperature to.

[0070] In this optional implementation, the above-mentioned generation and sending of intention inquiry information based on inheritable intention information includes: determining the control type of controlling the controlled object based on the inheritable intention information; in response to the control type being used to characterize the adjustment of the controlled object to a preset range parameter range, generating and sending intention inquiry information based on the control type, as shown in Table 2, the first sound zone is the co-pilot sound zone, and the second sound zone is the main driving sound zone. For the voice information of the main driving sound zone "Hello XX, adjust the air conditioning temperature to 20 degrees", the execution subject and the feedback information are: OK, and the action execution intention is obtained: control the main driving air conditioning to 20 degrees, and save the action execution intention as inheritable intention information in the local database; for the voice information of the co-pilot zone "I want it too", it is determined that it matches the intention inheritance key information, obtains the inheritable intention information of the main driving sound zone, and generates the intention inquiry information "How many degrees do you want to adjust the temperature to", and after obtaining the reply information, generates the action execution intention of controlling the controlled device of the first sound zone, obtains the instruction based on the action execution intention, and sends the instruction to the controlled device of the first sound zone, thereby realizing the execution of the instruction action by the controlled device of the first sound zone.

[0071] Table 2

[0072]

[0073] In this optional implementation, the reply information is information that responds to the intended inquiry information, such as the "23 degrees" replied by the main body of the co-pilot audio zone in Table 2 is a kind of reply information.

[0074] Optionally, after receiving the voice information of the second sound zone among the at least two sound zones, the execution subject on which the voice control method runs can directly feedback feedback information, such as the feedback information "OK" that appears for the first time in Table 2, so that the sender of the voice information in the second sound zone can get a positive response.

[0075] Optionally, after receiving the reply information of the first sound zone among the at least two sound zones, the execution subject on which the voice control method runs can directly feedback feedback information, such as the feedback information "OK" appearing for the second time in Table 2, so that the sender of the voice information in the first sound zone can get a positive reply.

[0076] In this optional implementation, the command action corresponding to the inheritable intention information and the reply information refers to the command action generated by the inheritable intention information and the reply information, wherein the command action is the action generated by the controlled object due to the command.

[0077] In this optional implementation, the above-mentioned control of the controlled object of the first sound zone to execute the instruction action corresponding to the inheritable intention information and the reply information includes: splicing the inheritable intention information and the reply information together and inputting them into the trained intention recognition model to obtain the action execution intention of the inheritable intention information and the reply information; based on the action execution intention of the inheritable intention information and the reply information, through table lookup and template filling, obtain the instruction to control the controlled object of the first sound zone, send the instruction to the controlled object of the first sound zone related to the instruction, so that the controlled object of the first sound zone executes the instruction action corresponding to the inheritable intention information and the reply information. Figure 3 The action execution intention that can inherit the intention information and the reply information is: control the air conditioner of the co-pilot to 23 degrees. Through the action execution intention, the command to control the controlled object (air conditioner) of the co-pilot is obtained, so that the air conditioner of the co-pilot is adjusted to 23 degrees.

[0078] The method for controlling the controlled object of the first sound zone provided by this optional implementation generates and sends intention inquiry information based on inheritable intention information; in response to receiving reply information to the intention inquiry information, the controlled object of the first sound zone is controlled to execute the instruction action corresponding to the inheritable intention information and the reply information, thereby improving the accuracy of control over the controlled object of the first sound zone.

[0079] Optionally, in response to detecting inheritable intention information, controlling the controlled object of the first sound zone to execute a command action includes: based on the inheritable intention information, determining that the controlled object of the first sound zone does not need to generate intention query information, and controlling the controlled object of the first sound zone to execute the command action corresponding to the inheritable intention information.

[0080] In some embodiments of the present disclosure, the above-mentioned voice control method also includes: in response to detecting that the voice information does not match the intention-inherited key information, determining whether the voice information includes wake-up control information; in response to the judgment result that the voice information includes wake-up control information, based on the voice information, identifying the action execution intention; and controlling the controlled object in the first sound zone to perform the instruction action of the action execution intention.

[0081] In this embodiment, the wake-up control information is voice data including wake-up information and control information, wherein the wake-up information is information for waking up the control execution subject, and the control information is information for enabling the execution subject to control the controlled object. When the voice information includes the wake-up control information, it indicates that the control object can currently be controlled to execute the command action.

[0082] In this embodiment, the action execution intention is the intention information for controlling the controlled object of the current sound zone. Different intention information corresponds to different action instructions, and each action instruction is used to control the controlled object to execute the instruction action.

[0083] In the present embodiment, in response to the judgment result that the voice information includes wake-up control information, identifying the action execution intention based on the voice information includes: in response to the judgment result that the voice information includes wake-up control information, detecting whether the voice information contains control keywords for controlling the controlled object of the first sound zone; in response to the voice information containing control keywords, taking the control keywords and the controlled object of the first sound zone as the action execution intention.

[0084] The voice control method provided in this embodiment, when the voice information does not match the intention inheritance key information, obtains the action execution intention by judging whether the voice information contains wake-up control information, and controls the controlled object in the first sound zone to execute the instruction action of the action execution intention, thereby improving the accuracy of controlling the controlled object in the first sound zone.

[0085] Figure 2 A process 200 according to an embodiment of the voice control method of the present disclosure is shown. The voice control method comprises the following steps:

[0086] Step 201 , receiving voice information of a first audio zone among at least two audio zones, and then executing step 202 .

[0087] Step 202, detect whether the voice information matches the intended inheritance key information; if it is detected that the voice information matches the intended inheritance key information, execute step 203; if it is detected that the voice information does not match the intended inheritance key information, execute step 206.

[0088] Step 203 , detecting whether there is inheritable intention information of the second audio zone; if inheritable intention information of the second audio zone is detected, then executing step 204 .

[0089] In this embodiment, the second sound zone is a sound zone different from the first sound zone.

[0090] Step 204 , controlling the controlled object in the first sound zone to execute the command action, and then executing step 205 .

[0091] In this embodiment, it should be understood that the operations and features in the above steps 201 to 204 correspond to the operations and features in steps 101 to 104, respectively. Therefore, the above descriptions of the operations and features in steps 101 to 104 are also applicable to steps 201 to 204 and will not be repeated here.

[0092] Step 205, end.

[0093] Step 206 , determining whether the voice information includes the wake-up control information; if the determination result is that the voice information includes the wake-up control information, then executing step 207 .

[0094] Optionally, if the judgment result is that the voice information does not include the wake-up control information, it is judged whether there is wake-up information after the voice information. If the judgment result is that the voice information includes the wake-up information, a preset answer message is issued, such as the answer message: I am here. The wake-up information is information for waking up the execution subject on which the current voice control method is running. Through the wake-up information, the execution subject can be awakened from the sleep state to the working state.

[0095] Step 207, based on the voice information, identify the action execution intention, and then execute step 208.

[0096] In this embodiment, the above step 207 can perform intent recognition on the voice information to obtain the action execution intention of the voice information. Among them, performing intent recognition on voice data is a mature technology and will not be repeated here.

[0097] Step 208, detect whether the action execution intention is the inheritable intention information of the first sound zone; if it is detected that the action execution intention is the inheritable intention information of the first sound zone, then execute step 209; if it is detected that the action execution intention is not the inheritable intention information of the first sound zone, then execute step 210.

[0098] In this embodiment, the inheritable intent information is a kind of intent information inherited by other sound zones except the current sound zone among at least two sound zones, wherein, when the current sound zone is the first sound zone, the other sound zones except the current sound zone among the at least two sound zones are the second sound zone.

[0099] In this embodiment, the action execution intention is a kind of intention information. By detecting whether the action execution intention is inheritable intention information, the execution subject can detect whether to save the action execution intention. If the action execution intention is inheritable intention information, the action execution intention is saved as inheritable intention information; otherwise, the action execution intention is not saved.

[0100] In this embodiment, whether the action execution intention is inheritable intention information can be detected by extracting the key information of the action execution intention. Specifically, a correspondence table of inheritable intention information and key information is pre-set. After extracting the key information of the action execution intention, the extracted key information is compared one by one with the key information in the above correspondence table. If the key information in the correspondence table is the same as the extracted key information, it is determined that the action execution intention is inheritable intention information.

[0101] In this embodiment, the key information may be the intended implementation object or the object after the verb. For example, if an action execution intention is: turn on the car air conditioner, then the intended implementation object or the object after the verb is "car air conditioner".

[0102] Step 209 , storing the action execution intention as inheritable intention information of the first sound zone in the local database, and then executing step 210 .

[0103] In this embodiment, the local database is a pre-set database for storing inheritable intent information. The inheritable intent information in the local database can be stored according to the tone zone, so that the inheritable intent information of each tone zone can be effectively divided; optionally, the inheritable information in the local database may not be stored according to the tone zone. In order to effectively determine the tone zone to which the inheritable intent information belongs, the tone zone marked in the inheritable intent information can be obtained.

[0104] In this embodiment, when the voice information of the second sound zone other than the first sound zone in at least two sound zones matches the intention inheritance key information, the intention inheritance key information of the second sound zone in the local database can be directly inherited to control the controlled object of the second sound zone to execute the command action.

[0105] Step 210 , controlling the controlled object in the first sound zone to execute the command action of the action execution intention, and then executing step 205 .

[0106] In this embodiment, the action execution intention is intention information. In order to facilitate the execution subject's control over the controlled object, the action execution intention can be converted into an instruction to control the controlled object and the instruction is sent to the controlled object. After the controlled object receives the instruction, it executes the instruction action corresponding to the instruction.

[0107] In this embodiment, the instructions for controlling controlled objects in different sound zones are different. In order to effectively control the controlled objects in each sound zone, the execution subject can pre-generate instructions for the controlled objects in each sound zone. After obtaining the action execution intention, the action execution intention is matched to the instructions of the corresponding sound zone, and the instructions are sent to the controlled objects, so that the controlled objects complete the instruction actions.

[0108] The voice control method provided in this embodiment determines whether the voice information includes wake-up control information when the voice information does not match the key information of intention inheritance; if the judgment result is that the voice information includes the wake-up control information, the action execution intention is identified based on the voice information; it is determined whether the action execution intention is inheritable intention information, and if the action execution intention is inheritable intention information, the action execution intention is stored in the local data as the inheritable intention information of the first sound zone, and the instruction action of the action intention is executed. Therefore, when the voice information does not match the key information of intention inheritance, the wake-up control information in the voice information is detected, the action execution intention is obtained based on the wake-up control information, the action execution intention is stored in the local database as the inheritable intention information of the first sound zone, and the instruction action of the action execution intention is executed, which can effectively feedback the user's voice control instructions and improve the user experience; based on the voice information, the inheritable intention information of the first sound zone is generated and stored, and the intention inheritance corpus is provided for the intention inheritance of other sound zones in at least two sound zones, which can simplify the user's voice control process and save the control algorithm calculation amount.

[0109] In a specific example of the present disclosure, the above-mentioned voice control method includes: when the execution subject performs voice interaction processing with the subject of one of the at least two sound zones, if the voice information of the subject does not have the key information of intention inheritance, it is determined whether there is the intention information that can be inherited from other sound zones in the voice information, and the inheritable intention information is cached. When there are multiple other sound zones that also want to execute the same inheritable intention information at the same time, the voice information including the key information of intention inheritance is directly output, such as "I also want" using keyword detection technology to match the voice information to see if it hits the key information of intention inheritance. If the key information of intention inheritance is hit, it is determined whether there is the inheritable intention information locally. If it exists, the inheritable intention information is directly inherited, and the instruction action corresponding to the inheritable intention information under the sound zone is executed. If it is determined that the intention needs to turn on active inquiry to accurately locate the intention, the specific intention can be accurately located by turning on active inquiry through voice. If it does not exist, no response will be given.

[0110] In some optional implementations of the present disclosure, the above-mentioned detection of whether the action execution intention is inheritable intention information of the first sound zone includes: determining the controlled object in the current scene based on the action execution intention; detecting whether the controlled object belongs to a preset vertical category; in response to detecting that the controlled object belongs to the preset vertical category, determining that the action execution intention is inheritable intention information of the first sound zone.

[0111] In this optional implementation, vertical categories refer to providing specialized products or services in a specific field or market segment for a group of users with similar needs, hobbies, and other characteristics. Preset vertical categories are used to divide inheritable intent information. At least one specific vertical category is preset according to the usage scenario. For example, in the vehicle scenario, the preset vertical category can be high-frequency vertical categories such as vehicle control, system settings, air conditioning, and multimedia.

[0112] The voice control method provided in this embodiment determines the controlled object in the current scene based on the action execution intention; detects whether the controlled object belongs to a preset vertical category, and when the controlled object belongs to the preset vertical category, determines that the action intention is inheritable intention information of the first sound zone, and divides the action execution intention into inheritable intention information or non-inheritable intention information through the preset vertical category, thereby improving the accuracy and reliability of obtaining the inheritable intention information.

[0113] Optionally, the above-mentioned detection of whether the action execution intention is inheritable intention information of the first sound zone also includes: in response to detecting that the controlled object does not belong to a preset vertical category, determining that the action execution intention is not inheritable intention information of the first sound zone, and there is no need to store the action execution intention.

[0114] In some optional implementations of the present disclosure, the above-mentioned detection of whether the action execution intention is inheritable intention information of the first sound zone includes: determining the controlled object in the current scene based on the action execution intention; detecting whether the second sound zone has an object of the same type as the controlled object; in response to detecting that the second sound zone has an object of the same type as the controlled object, determining that the action execution intention is inheritable intention information of the first sound zone.

[0115] In this optional implementation, the controlled object refers to a device or system controlled by the execution subject, and the above-mentioned object of the same type as the controlled object refers to a controlled object of the same type or model. Each of the at least two sound zones may have a corresponding controlled object, and the execution subject sends instructions to the corresponding controlled objects in each sound zone, so that the controlled object can execute the instruction action corresponding to the instruction.

[0116] The voice control method provided by this optional implementation method determines the controlled object in the current scene based on the action execution intention; detects whether the second sound zone has an object of the same type as the controlled object; in response to detecting that the second sound zone has an object of the same type as the controlled object, determines that the action execution intention is inheritable intention information of the first sound zone, and divides the action execution intention into inheritable intention information or non-inheritable intention information according to the type of the controlled object, thereby improving the accuracy and reliability of obtaining the inheritable intention information.

[0117] Optionally, the above-mentioned detection of whether the action execution intention is inheritable intention information of the first sound zone also includes: in response to detecting that the second sound zone does not have an object of the same type as the controlled object, determining that the action execution intention is not inheritable intention information of the first sound zone, and there is no need to store the action execution intention.

[0118] In some optional implementations of the present disclosure, in response to the judgment result that the voice information includes wake-up control information, identifying the action execution intention based on the voice information includes: in response to the judgment result that the voice information includes wake-up control information, obtaining a converted text based on the voice information; inputting the converted text into a trained intention recognition model to obtain the action execution intention output by the intention recognition model.

[0119] In this optional implementation, speech-to-text conversion can be used to convert speech information into translated text; the intention recognition model is a model for performing intent analysis on the translated text, and the action execution intention is the intention information for controlling the controlled object.

[0120] In this optional implementation, intent recognition is an important task in natural language processing, which aims to determine the intent or purpose expressed in the converted text. The intent recognition model is used to understand the semantics of the converted text and obtain the action execution intention in order to better answer the user's questions or provide related services. The intent recognition model understands the user's needs, performs the corresponding operation of the action execution intention or provides corresponding information, thereby improving the user experience and service quality.

[0121] The method for identifying action execution intentions provided by this optional implementation converts voice information into converted text when the voice information includes wake-up control information, and inputs the converted text into a trained intention recognition model to obtain the action execution intention output by the intention recognition model, and uses the intention recognition model to process the converted text to improve the accuracy of obtaining the action execution intention.

[0122] Optionally, in response to the judgment result that the voice information includes wake-up control information, identifying the action execution intention based on the voice information includes: in response to the judgment result that the voice information includes wake-up control information, obtaining a converted text based on the voice information; detecting whether the converted text includes a wake-up word; in response to detecting that the converted text includes the wake-up word, inputting the text after the wake-up word in the converted text into a trained intention recognition model to obtain the action execution intention output by the intention recognition model.

[0123] Further references Figure 3 As an implementation of the methods shown in the above figures, the present disclosure provides an embodiment of a voice control device, which is similar to Figure 1 Corresponding to the method embodiment shown, the device can be specifically implemented in various electronic devices.

[0124] like Figure 3 As shown, the voice control device 300 provided in this embodiment includes: a receiving unit 301, a key detection unit 302, an intention detection unit 303, and an execution unit 304. Among them, the above-mentioned receiving unit 301 can be configured to receive voice information of the first sound zone of at least two sound zones. The above-mentioned key detection unit 302 can be configured to detect whether the voice information matches the intention inheritance key information. The above-mentioned intention detection unit 303 can be configured to detect whether there is inheritable intention information of the second sound zone in response to detecting that the voice information matches the intention inheritance key information, and the second sound zone is a sound zone different from the first sound zone; the above-mentioned execution unit 304 can be configured to control the controlled object of the first sound zone to execute the instruction action in response to detecting that there is inheritable intention information.

[0125] In this embodiment, the specific processing of the receiving unit 301, the key detection unit 302, the intention detection unit 303, and the execution unit 304 and the technical effects thereof can be referred to in detail. Figure 1 The relevant descriptions of step 101, step 102, step 103, and step 104 in the corresponding embodiment are not repeated here.

[0126] In some optional implementations of the present embodiment, the above-mentioned intention inheritance key information includes: intention inheritance keywords, and the above-mentioned key detection unit 302 is configured to: obtain a converted text based on voice information; detect whether the converted text contains intention inheritance keywords; in response to detecting that the converted text contains intention inheritance keywords, determine that the voice information matches the intention inheritance key information.

[0127] In some optional implementations of the present embodiment, the above-mentioned intention inheritance key information includes: intention inheritance semantic information, and the above-mentioned key detection unit 302 is configured to: extract text semantic information based on voice information; detect whether the text semantic information matches the intention inheritance semantic information; in response to detecting that the text semantic information matches the intention inheritance semantic information, determine that the voice information matches the intention inheritance key information.

[0128] In some optional implementations of the present embodiment, the above-mentioned intention detection unit 303 is configured to: in response to detecting that the voice information matches the intention inheritance key information, select the inheritable intention information with the shortest time difference with the voice information from the inheritable intention information of at least two sound zones stored locally; detect whether the inheritable intention information with the shortest time difference with the voice information belongs to the first sound zone; in response to detecting that the inheritable intention information with the shortest time difference with the voice information does not belong to the first sound zone, determine the inheritable intention information with the second sound zone.

[0129] In some optional implementations of the present embodiment, the intention detection unit 303 is configured to: in response to detecting that the voice information matches the intention inheritance key information, detect whether the local database stores the action execution intention of the second sound zone; in response to detecting an action execution intention, and there is only one action execution intention; determine the action execution intention as inheritable intention information, and determine that the second sound zone has inheritable intention information.

[0130] In some optional implementations of the present embodiment, the above-mentioned intention detection unit 303 is configured to: in response to detecting that the voice information matches the intention inheritance key information, detect whether the intention inheritance key information includes the intention inheritance semantic information of the second sound zone; in response to detecting that the intention inheritance key information includes the intention inheritance semantic information of the second sound zone, determine the inheritable intention information having the second sound zone.

[0131] In some optional implementations of this embodiment, the above-mentioned execution unit 304 is configured to: generate and send intention inquiry information based on inheritable intention information; in response to receiving reply information to the intention inquiry information, control the controlled object of the first sound zone to execute the instruction action corresponding to the inheritable intention information and the reply information.

[0132] In some optional implementations of this embodiment, the above-mentioned device also includes: a first implementation unit (not shown in the figure), and the above-mentioned first implementation unit is configured to: in response to detecting that the voice information does not match the intention-inheritance key information, determine whether the voice information includes wake-up control information; in response to the judgment result that the voice information includes wake-up control information, based on the voice information, identify the action execution intention; control the controlled object in the first sound zone to perform the instruction action of the action execution intention.

[0133] In some optional implementations of the present embodiment, the above-mentioned device 300 also includes: a second implementation unit (not shown in the figure), and the above-mentioned second implementation unit is configured to: detect whether the action execution intention is inheritable intention information of the first sound zone; in response to detecting that the action execution intention is inheritable intention information of the first sound zone, store the action execution intention as inheritable intention information of the first sound zone in a local database.

[0134] In some optional implementations of this embodiment, the above-mentioned second implementation unit is configured to: determine the controlled object in the current scene based on the action execution intention; detect whether the controlled object belongs to a preset vertical category; in response to detecting that the controlled object belongs to the preset vertical category, determine that the action execution intention is inheritable intention information of the first sound zone.

[0135] In some optional implementations of the present disclosure, the second implementation unit is configured to: determine the controlled object in the current scene based on the action execution intention; detect whether the second sound zone has an object of the same type as the controlled object; and in response to detecting that the second sound zone has an object of the same type as the controlled object, determine that the action execution intention is inheritable intention information of the first sound zone.

[0136] In some optional implementations of the present disclosure, the above-mentioned first implementation unit is configured to: in response to a judgment result that the voice information includes wake-up control information, obtain a converted text based on the voice information; input the converted text into a trained intention recognition model to obtain the action execution intention output by the intention recognition model.

[0137] The voice control device provided by the embodiment of the present disclosure, first, the receiving unit 301 receives the voice information of the first sound zone of at least two sound zones; secondly, the key detection unit 302 detects whether the voice information matches the intention inheritance key information; thirdly, the intention detection unit 303 detects whether there is inheritable intention information of the second sound zone in response to detecting that the voice information matches the intention inheritance key information, and the second sound zone is a sound zone different from the first sound zone; finally, the execution unit 304 controls the controlled object of the first sound zone to execute the command action in response to detecting that there is inheritable intention information. Therefore, when the voice information of the first sound zone matches the intention inheritance key information, if the inheritable intention information of the second sound zone is detected, the inheritable intention information is directly executed, so that the user of the first sound zone does not need to issue a control instruction matching the command action, which simplifies the user voice control process, saves the calculation amount of the control algorithm, and improves the user experience.

[0138] In the technical solution of the present disclosure, the collection, storage, use, processing, transmission, provision and disclosure of user personal information involved are in compliance with the provisions of relevant laws and regulations and do not violate public order and good morals.

[0139] According to an embodiment of the present disclosure, the present disclosure also provides an electronic device, a readable storage medium and a computer program product.

[0140] Figure 4A schematic block diagram of an example electronic device 400 that can be used to implement an embodiment of the present disclosure is shown. The electronic device is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device can also represent various forms of mobile devices, such as personal digital processing, cellular phones, smart phones, wearable devices, and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely examples and are not intended to limit the implementation of the present disclosure described and / or required herein.

[0141] like Figure 4 As shown, the device 400 includes a computing unit 401, which can perform various appropriate actions and processes according to a computer program stored in a read-only memory (ROM) 402 or a computer program loaded from a storage unit 408 into a random access memory (RAM) 403. In the RAM 403, various programs and data required for the operation of the device 400 can also be stored. The computing unit 401, the ROM 402, and the RAM 403 are connected to each other via a bus 404. An input / output (I / O) interface 405 is also connected to the bus 404.

[0142] A number of components in the device 400 are connected to the I / O interface 405, including: an input unit 406, such as a keyboard, a mouse, etc.; an output unit 407, such as various types of displays, speakers, etc.; a storage unit 408, such as a disk, an optical disk, etc.; and a communication unit 409, such as a network card, a modem, a wireless communication transceiver, etc. The communication unit 409 allows the device 400 to exchange information / data with other devices through a computer network such as the Internet and / or various telecommunication networks.

[0143] The computing unit 401 may be a variety of general and / or special processing components with processing and computing capabilities. Some examples of the computing unit 401 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various dedicated artificial intelligence (AI) computing chips, various computing units running machine learning model algorithms, digital signal processors (DSPs), and any appropriate processors, controllers, microcontrollers, etc. The computing unit 401 performs the various methods and processes described above, such as the voice control method. For example, in some embodiments, the voice control method may be implemented as a computer software program, which is tangibly contained in a machine-readable medium, such as a storage unit 408. In some embodiments, part or all of the computer program may be loaded and / or installed on the device 400 via ROM 402 and / or communication unit 409. When the computer program is loaded into RAM 403 and executed by the computing unit 401, one or more steps of the voice control method described above may be performed. Alternatively, in other embodiments, the computing unit 401 may be configured to perform the voice control method in any other appropriate manner (e.g., by means of firmware).

[0144] Various implementations of the systems and techniques described above herein can be implemented in digital electronic circuit systems, integrated circuit systems, field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard products (ASSPs), systems on chips (SOCs), load programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various implementations can include: being implemented in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which can be a special purpose or general purpose programmable processor that can receive data and instructions from a storage system, at least one input device, and at least one output device, and transmit data and instructions to the storage system, the at least one input device, and the at least one output device.

[0145] The program code for implementing the method of the present disclosure may be written in any combination of one or more programming languages. These program codes may be provided to a processor or controller of a general-purpose computer, a special-purpose computer, or other programmable voice control device, so that the program code, when executed by the processor or controller, implements the functions / operations specified in the flow chart and / or block diagram. The program code may be executed entirely on the machine, partially on the machine, partially on the machine and partially on a remote machine as a stand-alone software package, or entirely on a remote machine or server.

[0146] In the context of the present disclosure, a machine-readable medium may be a tangible medium that may contain or store a program for use by or in conjunction with an instruction execution system, device, or equipment. A machine-readable medium may be a machine-readable signal medium or a machine-readable storage medium. A machine-readable medium may include, but is not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, device, or equipment, or any suitable combination of the foregoing. A more specific example of a machine-readable storage medium may include an electrical connection based on one or more lines, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

[0147] To provide interaction with a user, the systems and techniques described herein can be implemented on a computer having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user; and a keyboard and pointing device (e.g., a mouse or trackball) through which the user can provide input to the computer. Other types of devices can also be used to provide interaction with the user; for example, the feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including acoustic input, voice input, or tactile input).

[0148] The systems and techniques described herein may be implemented in a computing system that includes back-end components (e.g., as a data server), or a computing system that includes middleware components (e.g., an information server), or a computing system that includes front-end components (e.g., a user computer with a graphical user interface or a web browser through which a user can interact with implementations of the systems and techniques described herein), or a computing system that includes any combination of such back-end components, middleware components, or front-end components. The components of the system may be interconnected by any form or medium of digital data communication (e.g., a communications network). Examples of communications networks include: a local area network (LAN), a wide area network (WAN), and the Internet.

[0149] A computer system may include a client and a server. The client and the server are generally remote from each other and usually interact through a communication network. The relationship of client and server is generated by computer programs running on respective computers and having a client-server relationship with each other. The server may be a cloud server, a server of a distributed system, or a server combined with a blockchain.

[0150] It should be understood that the various forms of processes shown above can be used to reorder, add or delete steps. For example, the steps recorded in this disclosure can be executed in parallel, sequentially or in different orders, as long as the desired results of the technical solutions disclosed in this disclosure can be achieved, and this document does not limit this.

[0151] The above specific implementations do not constitute a limitation on the protection scope of the present disclosure. It should be understood by those skilled in the art that various modifications, combinations, sub-combinations and substitutions can be made according to design requirements and other factors. Any modification, equivalent substitution and improvement made within the spirit and principle of the present disclosure shall be included in the protection scope of the present disclosure.

Claims

1. A voice control method, the method comprising: receiving voice information of a first sound zone of at least two sound zones; Detecting whether the voice information matches the intended inheritance key information; In response to detecting that the voice information matches the intention inheritance key information, detecting whether there is inheritable intention information of the second sound zone, the second sound zone being a sound zone different from the first sound zone; In response to detecting the inheritable intention information, the controlled object of the first sound zone is controlled to perform a command action.

2. The method according to claim 1, wherein: The intention inheritance key information includes: intention inheritance keywords, and the detecting whether the voice information matches the intention inheritance key information includes: Based on the voice information, obtaining a converted text; Detecting whether the conversion text contains the intended inheritance keyword; In response to detecting that the intention-inheritance keyword is present in the converted text, it is determined that the voice information matches the intention-inheritance key information.

3. The method according to claim 1, wherein: The intention inheritance key information includes: intention inheritance semantic information, and the detecting whether the voice information matches the intention inheritance key information includes: Extracting text semantic information based on the voice information; Detecting whether the text semantic information matches the intention inheritance semantic information; In response to detecting that the text semantic information matches the intention-inheritance semantic information, it is determined that the voice information matches the intention-inheritance key information.

4. The method according to claim 1, wherein: In response to detecting that the voice information matches the intention inheritance key information, detecting whether there is inheritable intention information of the second sound zone includes: In response to detecting that the voice information matches the intention inheritance key information, selecting the inheritable intention information with the shortest time difference with the voice information from the inheritable intention information of at least two sound zones stored locally; Detecting whether the inheritable intention information with the shortest time difference from the voice information belongs to the first sound zone; In response to detecting that the inheritable intention information having the shortest time difference with the voice information does not belong to the first sound zone, determining the inheritable intention information having the second sound zone.

5. The method according to claim 1, wherein: In response to detecting that the voice information matches the intention inheritance key information, detecting whether there is inheritable intention information of the second sound zone includes: In response to detecting that the voice information matches the intention inheritance key information, detecting whether the local database stores the action execution intention of the second sound zone; In response to detecting the action execution intention, and the action execution intention is only one; The action execution intention is determined as inheritable intention information, and the second sound zone is determined to have inheritable intention information.

6. The method according to claim 1, wherein: In response to detecting that the voice information matches the intention inheritance key information, detecting whether there is inheritable intention information of the second sound zone includes: In response to detecting that the speech information matches the intention inheritance key information, detecting whether the intention inheritance key information includes the intention inheritance semantic information of the second sound zone; In response to detecting that the intention inheritance key information includes the intention inheritance semantic information of the second audio zone, inheritable intention information having the second audio zone is determined.

7. The method according to claim 1, wherein: In response to detecting the inheritable intention information, controlling the controlled object in the first sound zone to execute the command action includes: Based on the inheritable intent information, generate and send intent query information; In response to receiving reply information to the intention inquiry information, the controlled object of the first sound zone is controlled to execute the instruction action corresponding to the inheritable intention information and the reply information.

8. The method according to any one of claims 1 to 7, further comprising: In response to detecting that the voice information does not match the intended inheritance key information, determining whether the voice information includes wake-up control information; In response to a determination that the voice information includes wake-up control information, identifying an action execution intention based on the voice information; The controlled object of the first sound zone is controlled to execute the instruction action of the action execution intention.

9. The method according to claim 8, further comprising: Detecting whether the action execution intention is inheritable intention information of the first sound zone; In response to detecting that the action execution intention is inheritable intention information of the first music zone, the action execution intention is stored in a local database as inheritable intention information of the first music zone.

10. The method according to claim 9, wherein: The step of detecting whether the action execution intention is inheritable intention information of the first sound zone includes: Based on the action execution intention, determining the controlled object in the current scene; Detecting whether the controlled object belongs to a preset vertical category; In response to detecting that the controlled object belongs to a preset vertical category, determining that the action execution intention is inheritable intention information of the first sound zone.

11. The method according to claim 9, wherein: The step of detecting whether the action execution intention is inheritable intention information of the first sound zone includes: Based on the action execution intention, determining the controlled object in the current scene; detecting whether the second sound zone has an object of the same type as the controlled object; In response to detecting that the second sound zone has an object of the same type as the controlled object, determining the action execution intention as inheritable intention information of the first sound zone.

12. The method according to claim 8, wherein: In response to the judgment result that the voice information includes the wake-up control information, based on the voice information, identifying the action execution intention includes: In response to the judgment result that the voice information includes the wake-up control information, obtaining a converted text based on the voice information; The converted text is input into the trained intention recognition model to obtain the action execution intention output by the intention recognition model.

13. A voice control device, comprising: A receiving unit, configured to receive voice information of a first sound zone of at least two sound zones; A key detection unit, configured to detect whether the voice information matches the intended inheritance key information; an intention detection unit, configured to detect whether there is inheritable intention information of the second sound zone in response to detecting that the voice information matches the intention inheritance key information, the second sound zone being a sound zone different from the first sound zone; The execution unit is configured to control the controlled object of the first sound zone to execute the command action in response to detecting the inheritable intention information.

14. The device according to claim 13, wherein: The intention inheritance key information includes: intention inheritance keywords, and the key detection unit is configured to: obtain a converted text based on the voice information; detect whether the converted text contains the intention inheritance keywords; in response to detecting that the converted text contains the intention inheritance keywords, determine that the voice information matches the intention inheritance key information.

15. The device according to claim 13, wherein: The intention inheritance key information includes: intention inheritance semantic information, and the key detection unit is configured to: extract text semantic information based on the voice information; detect whether the text semantic information matches the intention inheritance semantic information; in response to detecting that the text semantic information matches the intention inheritance semantic information, determine that the voice information matches the intention inheritance key information.

16. The device according to claim 13, wherein: The intention detection unit is configured to: in response to detecting that the voice information matches the intention inheritance key information, select the inheritable intention information with the shortest time difference with the voice information from the inheritable intention information of at least two sound zones stored locally; detect whether the inheritable intention information with the shortest time difference with the voice information belongs to the first sound zone; in response to detecting that the inheritable intention information with the shortest time difference with the voice information does not belong to the first sound zone, determine the inheritable intention information of the second sound zone.

17. The device according to claim 13, wherein: The intention detection unit is configured to: in response to detecting that the voice information matches the intention inheritance key information, detect whether the local database stores the action execution intention of the second sound zone; in response to detecting the action execution intention, and there is only one action execution intention; determine the action execution intention as inheritable intention information, and determine that the second sound zone has inheritable intention information.

18. The device according to claim 13, wherein: The intention detection unit is configured to: in response to detecting that the voice information matches the intention inheritance key information, detect whether the intention inheritance key information includes the intention inheritance semantic information of the second sound zone; in response to detecting that the intention inheritance key information includes the intention inheritance semantic information of the second sound zone, determine the inheritable intention information of the second sound zone.

19. According to the device of claim 13, the execution unit is configured to: generate and send intention inquiry information based on the inheritable intention information; in response to receiving reply information to the intention inquiry information, control the controlled object of the first sound zone to execute the instruction action corresponding to the inheritable intention information and the reply information.

20. The device according to any one of claims 13 to 19, further comprising: The first implementation unit is configured to: in response to detecting that the voice information does not match the intention inheritance key information, determine whether the voice information includes wake-up control information; in response to the judgment result that the voice information includes wake-up control information, identify the action execution intention based on the voice information; and control the controlled object of the first sound zone to execute the instruction action of the action execution intention.

21. The apparatus according to claim 20, further comprising: The second implementation unit is configured to: detect whether the action execution intention is inheritable intention information of the first sound zone; In response to detecting that the action execution intention is inheritable intention information of the first music zone, the action execution intention is stored in a local database as inheritable intention information of the first music zone.

22. The device according to claim 21, wherein The second implementation unit is configured to: determine a controlled object in a current scene based on the action execution intention; detect whether the controlled object belongs to a preset vertical category; In response to detecting that the controlled object belongs to a preset vertical category, determining that the action execution intention is inheritable intention information of the first sound zone.

23. The device according to claim 21, wherein The second implementation unit is configured to: determine the controlled object in the current scene based on the action execution intention; detect whether the second sound zone has an object of the same type as the controlled object; and in response to detecting that the second sound zone has an object of the same type as the controlled object, determine that the action execution intention is inheritable intention information of the first sound zone.

24. The device according to claim 20, wherein: The first implementation unit is configured to: in response to a judgment result that the voice information includes wake-up control information, obtain a conversion text based on the voice information; input the conversion text into a trained intention recognition model to obtain an action execution intention output by the intention recognition model.

25. An electronic device, characterized in that: include: at least one processor; as well as a memory communicatively connected to the at least one processor; wherein, The memory stores instructions that can be executed by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to perform the method according to any one of claims 1 to 12.

26. A non-transitory computer-readable storage medium storing computer instructions, characterized in that: The computer instructions are used to enable the computer to execute the method according to any one of claims 1 to 12.

27. A computer program product, comprising a computer program, which, when executed by a processor, implements the method of any one of claims 1 to 12.

Images