Multi-voice intelligent voice interaction method, electronic equipment and storage medium

Through the multi-person intelligent voice interaction method, the problem of insufficient intention recognition accuracy in multi-person scenarios is solved. By identifying and evaluating intentions, the user interaction experience and recognition accuracy are improved.

CN118982979APending Publication Date: 2024-11-19QINGDAO HAIER TECH +2
View PDF 0 Cites 4 Cited by

Patent Information

Application Number
CN202411109344.6
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-08-13
Publication Date
2024-11-19

AI Technical Summary

Technical Problem

The existing voice recognition system has insufficient intention recognition accuracy in scenarios where multiple people speak at the same time, which affects the user's interactive experience.

Method used

Using multi-voice intelligent voice interaction method, by obtaining voice information, we determine whether the voice of multiple users is included, the intent information is determined based on the voice information, and according to preset configuration rules, including intent logic judgment, priority evaluation and system resource allocation, Determine the task execution information to perform the corresponding task.

Benefits of technology

Accurately understand each user's intentions in a complex scenario where multiple people speak at the same time, and improve user interaction experience and recognition accuracy of multiple people and multiple intentions.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118982979A_ABST
    Figure CN118982979A_ABST
Patent Text Reader

Abstract

The invention discloses a multi-voice intelligent voice interaction method, electronic equipment and a storage medium, and relates to the technical field of smart home / smart home, and the multi-voice intelligent voice interaction method comprises the steps: obtaining voice information; judging whether the voice information contains voices of a plurality of users or not; determining at least one piece of intention information based on the voice information under the condition that the voice information contains the voices of the plurality of users; task execution information is determined based on a preset configuration rule and the intention information, and the preset configuration rule comprises at least one of intention logic judgment, priority evaluation and system resource allocation; and executing a corresponding task based on the task execution information. According to the method and the device, the interaction scene of simultaneous speaking of multiple persons can be effectively processed, the intention of each user can be accurately identified, the intention can be configured based on the preset configuration rule, the task execution information is determined, then the corresponding task is executed, and the interaction experience of the users is improved.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present application relates to the field of smart home / smart family, and more specifically, to a multi-voice intelligent voice interaction method, an electronic device and a storage medium. Background Art

[0002] With the development of artificial intelligence technology and the increasing popularity of smart devices, voice interaction has become one of the important ways of human-computer interaction. However, the current voice recognition systems on the market mostly focus on single-person use scenarios. When faced with complex environments where multiple people speak at the same time or their intentions are intertwined, their recognition accuracy drops significantly, which in turn affects the overall quality of the user interaction experience.

[0003] Accordingly, the art needs a new intelligent voice interaction solution to solve the above problems. Summary of the invention

[0004] In order to overcome the above-mentioned defects, the present application is proposed to solve or at least partially solve the technical problem that the existing speech recognition system has insufficient accuracy in intent recognition in a scenario where multiple people speak at the same time.

[0005] In a first aspect, a multi-voice intelligent voice interaction method is provided, the method comprising: acquiring voice information; determining whether the voice information contains the voices of multiple users; in a case where the voice information contains the voices of multiple users, determining at least one intention information based on the voice information; determining task execution information based on preset configuration rules and the intention information, the preset configuration rules comprising at least one of intention logic judgment, priority evaluation and system resource allocation; and executing a corresponding task based on the task execution information.

[0006] In a technical solution of the above-mentioned multi-voice intelligent voice interaction method, when the voice information contains the voices of multiple users, at least one intention information is determined based on the voice information, including: processing the voice information based on sound features to obtain voice information corresponding to each user; inputting the voice information corresponding to each user into a preset voice recognition model to obtain text information of the voice information corresponding to each user; performing intent recognition on each of the text information respectively to determine the intention information corresponding to each of the text information.

[0007] In a technical solution of the above-mentioned multi-voice intelligent voice interaction method, the voice information is processed based on the sound features to obtain the voice information corresponding to each user, including: preprocessing the voice information to obtain processed voice information; splitting the processed voice information based on the sound features to obtain multiple audio clips; and synthesizing the audio clips with the same sound features to obtain the voice information corresponding to each user.

[0008] In a technical solution of the above-mentioned multi-voice intelligent speech interaction method, the intent recognition is performed on each of the text messages respectively to determine the intent information corresponding to each of the text messages, including: inputting each of the text messages into a preset intent recognition model, performing lexical analysis, syntactic analysis, semantic understanding and classification processing on each of the text messages, and determining the intent information corresponding to each of the text messages.

[0009] In a technical solution of the above-mentioned multi-voice intelligent speech interaction method, the task execution information is determined based on the preset configuration rules and the intention information, including: performing intent logic judgment on each of the intention information to determine the intention information that meets the judgment conditions; performing priority evaluation on the intention information that meets the judgment conditions to determine the priority order of all intention information; obtaining system real-time resource information; based on the system real-time resource information and the priority order of all intent information, allocating system resources to the intent information to determine the task execution information.

[0010] In a technical solution of the above-mentioned multi-voice intelligent voice interaction method, the intention logic judgment is performed on each of the intention information to determine the intention information that meets the judgment conditions, including: performing intention logic judgment on each of the intention information based on preset judgment dimensions to determine the intention information that meets the judgment conditions, and the preset judgment dimensions include at least one of timestamp judgment, logical relationship judgment, and conflict judgment.

[0011] In a technical solution of the above-mentioned multi-voice intelligent speech interaction method, the priority evaluation of the intent information that meets the judgment conditions is performed to determine the priority order of all intent information, including: based on preset priority evaluation rules, priority evaluation of each of the intent information is performed to determine the priority order of all intent information, and the preset priority evaluation rules include at least one of urgency, importance and user authority.

[0012] In a technical solution of the above-mentioned multi-voice intelligent speech interaction method, the system real-time resource information includes at least one of real-time computing power information, real-time memory information, and real-time network bandwidth information; the system resources are allocated to the intent information based on the priority order of the system real-time resource information and all intent information, and the task execution information is determined, including: determining the execution unit information and execution thread information corresponding to each of the intent information based on the system real-time resource information and the priority order of all intent information; determining the task execution information based on the execution unit information and execution thread information corresponding to all the intent information.

[0013] In a second aspect, an electronic device is provided, comprising at least one memory and at least one processor, wherein the memory is suitable for storing multiple program codes, and the program codes are suitable for being loaded and run by the processor to execute the method described in any one of the technical solutions of the above-mentioned multi-voice intelligent voice interaction method.

[0014] In a third aspect, a computer-readable storage medium is provided, wherein the computer-readable storage medium includes a stored program, wherein when the program is run, the method described in any one of the technical solutions of the above-mentioned multi-voice intelligent voice interaction method is executed.

[0015] The above one or more technical solutions of the present application have at least one or more of the following beneficial effects:

[0016] The multi-voice intelligent voice interaction method of the present application includes: obtaining voice information; determining whether the voice information contains the voices of multiple users; in the case where the voice information contains the voices of multiple users, determining at least one intention information based on the voice information; determining task execution information based on preset configuration rules and the intention information, the preset configuration rules including at least one of intention logic judgment, priority evaluation and system resource allocation; executing corresponding tasks based on the task execution information. The present application can perform intent recognition for complex voice scenarios where multiple people speak at the same time and have different intentions, accurately understand the intentions of each user, and then configure each intention according to the preset configuration rules, obtain the task execution information corresponding to all intentions, and make appropriate responses based on the task execution information, effectively improving the user's interactive experience and the recognition accuracy of multiple people and multiple intentions. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the present application and, together with the description, serve to explain the principles of the present application.

[0018] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, for ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative labor.

[0019] Figure 1 This is a flowchart of the main steps of a multi-voice intelligent speech interaction method according to an embodiment of the present application;

[0020] Figure 2 is a detailed step flow chart of a multi-voice intelligent speech interaction method according to another embodiment of the present application;

[0021] Figure 3 It is a schematic diagram of the main structure of an electronic device according to an embodiment of the present application. Reference numerals:

[0022] 11: processor; 12: memory. DETAILED DESCRIPTION

[0023] In order to enable those skilled in the art to better understand the solution of the present application, the technical solution in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of the present application.

[0024] It should be noted that the terms "first", "second", etc. in the specification and claims of the present application and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchangeable where appropriate, so that the embodiments of the present application described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any of their variations are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device comprising a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0025] With the development of artificial intelligence technology and the increasing popularity of smart devices, voice interaction has become one of the important ways of human-computer interaction. However, the current voice recognition systems on the market mostly focus on single-person use scenarios. When faced with complex environments where multiple people speak at the same time or their intentions are intertwined, their recognition accuracy drops significantly, which in turn affects the overall quality of the user interaction experience.

[0026] To this end, the present application provides a multi-voice intelligent voice interaction method, including: obtaining voice information; determining whether the voice information contains the voices of multiple users; in the case where the voice information contains the voices of multiple users, determining at least one intention information based on the voice information; determining task execution information based on preset configuration rules and the intention information, the preset configuration rules including at least one of intention logic judgment, priority evaluation and system resource allocation; executing corresponding tasks based on the task execution information. The present application can perform intent recognition on complex voice scenarios where multiple people speak at the same time and have different intentions, determine the intention of each user, and then configure each intention according to the preset configuration rules, obtain the task execution information corresponding to all intentions, and make appropriate responses based on the task execution information, effectively improving the user's interactive experience and the recognition accuracy of multiple people and multiple intentions.

[0027] See attached Figure 1 , Figure 1 1 is a flow chart of the main steps of the multi-voice intelligent voice interaction method according to an embodiment of the present application. Figure 1 As shown, the multi-voice intelligent voice interaction method in the embodiment of the present application mainly includes the following steps S101 to S105.

[0028] Step S101: Acquire voice information.

[0029] Step S102: Determine whether the voice information contains voices of multiple users.

[0030] Step S103: when the voice information contains voices of multiple users, determine at least one intention information based on the voice information.

[0031] Step S104: Determine task execution information based on preset configuration rules and the intention information, wherein the preset configuration rules include at least one of intention logic judgment, priority evaluation, and system resource allocation.

[0032] Step S105: executing the corresponding task based on the task execution information.

[0033] Based on the methods described in steps S101 to S105 above, the present application can perform intent recognition in complex voice scenarios where multiple people speak at the same time and have different intentions, determine the intention of each user, and then configure each intention according to preset configuration rules, obtain task execution information corresponding to all intentions, and make appropriate responses based on the task execution information, effectively improving the user's interactive experience and the recognition accuracy of multiple people and multiple intentions.

[0034] The above steps S101 to S105 are further explained below.

[0035] With respect to step S101, voice information is acquired.

[0036] In this embodiment, multiple intelligent voice devices are set in the interactive environment, and the voice information collected by the audio collection modules of the multiple intelligent voice devices at the same time is accurately obtained.

[0037] With respect to step S102, it is determined whether the voice information contains voices of multiple users.

[0038] Specifically, multiple users refer to two or more users. A method for determining whether the voice information contains the voices of multiple users may be to first divide the voice information into multiple audio segments, identify the sound features in each audio segment, and compare the sound features of all the identified audio segments to determine whether they are the same. If they are the same, it is determined that the voice information does not contain the voices of multiple users. If they are not the same, it is determined that the voice information contains the voices of multiple users.

[0039] The above is the description of step S101 and step S102. The following is a further description of step S103 to step S105.

[0040] For step S103, in one embodiment, when the voice information contains voices of multiple users, determining at least one intention information based on the voice information includes: processing the voice information based on sound features to obtain voice information corresponding to each user; inputting the voice information corresponding to each user into a preset voice recognition model to obtain text information of the voice information corresponding to each user; performing intent recognition on each of the text information respectively to determine the intention information corresponding to each of the text information.

[0041] Specifically, the voice information of each user in the voice information is determined based on the sound features, and the voice information of each user is input into the deep learning-based speech recognition model for recognition to determine the text information corresponding to the voice information of each user. The speech recognition model can convert the voice information into text information with user labels and timestamps by learning a large amount of voice data, and analyze the text information corresponding to the voice information of each user to determine the intention of each user.

[0042] In one embodiment, the voice information is processed based on the sound features to obtain voice information corresponding to each user, including: preprocessing the voice information to obtain processed voice information; splitting the processed voice information based on the sound features to obtain multiple audio segments; and synthesizing the audio segments with the same sound features to obtain voice information corresponding to each user.

[0043] Specifically, a speech enhancement algorithm is used to pre-process the acquired speech information. The pre-processing may be to remove background noise, cancel echoes, etc., so as to improve the clarity and recognizability of the speech information.

[0044] The processed voice information is split based on the sound features to obtain multiple audio clips with different sound features. The audio clips with the same sound features are obtained, and the audio clips with the same sound features are synthesized according to the acquisition timestamp to obtain the voice information corresponding to each user.

[0045] In one embodiment, the performing of intent recognition on each of the text messages respectively to determine the intent information corresponding to each of the text messages includes: inputting each of the text messages into a preset intent recognition model, performing lexical analysis, syntactic analysis, semantic understanding and classification processing on each of the text messages to determine the intent information corresponding to each of the text messages.

[0046] Specifically, each text message is input into a preset intent recognition model, and lexical analysis, syntactic analysis, semantic understanding and classification processing are performed on each text message.

[0047] Perform lexical analysis on text information, specifically dividing the text information into basic units such as words and phrases, and marking the part of speech of each unit (such as noun, verb, adjective, etc.).

[0048] For example, the text information is "adjust the air conditioning temperature to 23 degrees". After lexical analysis of the text information, it is determined that the part of speech of "air conditioning" is a noun, the part of speech of "cooling" is a verb, the part of speech of "temperature" is a noun, the part of speech of "adjust" is a verb, the part of speech of "to" is a preposition, and the part of speech of "23 degrees" is a combination of a numeral and a quantifier.

[0049] Performing syntactic analysis on text information, specifically determining the sentence structure and grammatical relationships of the text information, such as subject, predicate, object, etc., helps to understand the relationship between the various parts of the text information.

[0050] Semantic understanding of text information refers to combining context and domain knowledge to conduct in-depth analysis of the meaning of text information and identify that the semantics is to set the cooling temperature of the air conditioner to 23 degrees.

[0051] Classifying the user's intention means using machine learning algorithms, such as classification algorithms (support vector machines, decision trees, etc.) or neural networks in deep learning, to preset multiple intent categories and determine the type of user's intention. For example, if "setting the air conditioner temperature" or a similar intent category is preset, when it is determined that the semantics of the text information is to set the cooling temperature of the air conditioner to 23 degrees, the type of user's intention is determined to be "setting the air conditioner temperature".

[0052] Finally, the intent recognition model outputs the user's intent information based on the results of lexical analysis, syntactic analysis, semantic understanding and classification processing performed on each text information.

[0053] The above is the description of step S103 , and the following is a further description of step S104 .

[0054] For step S104, in one embodiment, determining the task execution information based on the preset configuration rules and the intent information includes: performing intent logic judgment on each of the intent information to determine the intent information that meets the judgment conditions; performing priority evaluation on the intent information that meets the judgment conditions to determine the priority order of all intent information; obtaining system real-time resource information; and allocating system resources to the intent information based on the system real-time resource information and the priority order of all intent information to determine the task execution information.

[0055] Specifically, the preset configuration rules include intent logic judgment, priority evaluation and system resource allocation for each intent information. After analyzing and configuring all intent information using the preset configuration rules, the task execution information corresponding to the intent information is determined.

[0056] In one embodiment, performing intent logical judgment on each of the intent information to determine the intent information that meets the judgment conditions includes: performing intent logical judgment on each of the intent information based on preset judgment dimensions to determine the intent information that meets the judgment conditions, and the preset judgment dimensions include at least one of timestamp judgment, logical relationship judgment, and conflict judgment.

[0057] Specifically, the preset judgment dimensions include timestamp judgment, logical relationship judgment and conflict judgment. The timestamp judgment can be a judgment on the collection timestamp of the voice information corresponding to each intention information.

[0058] The logical relationship judgment can be to judge the logical dependency relationship corresponding to each intention information. For example, the first intention information is "set the air conditioning cooling to 23 degrees", and the second intention information is "turn on the air conditioning". There is an obvious order requirement between the two intention information. It is impossible to set the air conditioning cooling temperature before turning on the air conditioning. Therefore, the result of the logical relationship judgment is that the second intention information should be executed before the first intention information.

[0059] The conflict judgment can be to judge whether there is a conflict between the control events corresponding to each intent information. For example, the second intent information is "turn on the air conditioner", and the third intent information is "turn off the air conditioner". There is an obvious conflict between the control events corresponding to the two intent information. Therefore, the result of the conflict judgment is that there is a conflict between the second intent information and the third intent information.

[0060] After performing intention logic judgment on each intention information based on the preset judgment dimension, the intention information that can be executed will be regarded as the intention information that meets the judgment conditions according to the intention logic judgment result.

[0061] For example, the first intention information is "Set the air conditioning cooling to 23 degrees", the second intention information is "Turn on the air conditioning", and the third intention information is "Turn off the air conditioning". The time converted from the timestamp of the first intention information is 17:55:02, the time converted from the timestamp of the second intention information is 17:55:01, and the time converted from the timestamp of the third intention information is 17:55:00.

[0062] The three intent information are subjected to intention logic judgment based on timestamp, logical relationship judgment and conflict judgment. The result of timestamp judgment is to execute the third intent information first, then the second intent information, and finally the first intent information.

[0063] The result of the logical relationship judgment is that the second intention information should be executed before the first intention information.

[0064] The conflict determination result indicates that the third intention information conflicts with the second intention information and the first intention information.

[0065] After performing intention logic judgment on the first intention information, the second intention information and the third intention information based on the preset judgment dimension, it is determined that the third intention information will not be executed, and the second intention information and the first intention information are regarded as intention information that meets the judgment conditions. The intention information also includes the logical relationship between the two intention information.

[0066] In one embodiment, the priority evaluation of the intent information that meets the judgment conditions is performed to determine the priority order of all intent information, including: based on preset priority evaluation rules, priority evaluation of each intent information is performed to determine the priority order of all intent information, and the preset priority evaluation rules include at least one of urgency, importance and user authority.

[0067] The urgency can be the urgency with which the intent information needs to be processed or responded to. The importance can be the impact of the intent information on the overall system. The user authority can be the user identity or authority level corresponding to the intent information.

[0068] Specifically, after determining the intent information that meets the judgment conditions, score each intent information according to the urgency and importance of the intent information and the user authority corresponding to the intent information, and finally sum the three scores of the intent information through a preset calculation method to obtain the priority score of the intent information. The preset calculation method can be a weighted summation, etc. Finally, sort the intent information based on the priority scores of all intent information to determine the priority order of all intent information.

[0069] In one embodiment, the system real-time resource information includes at least one of real-time computing power information, real-time memory information, and real-time network bandwidth information; the system resources are allocated to the intent information based on the priority order of the system real-time resource information and all intent information, and the task execution information is determined, including: determining the execution unit information and execution thread information corresponding to each of the intent information based on the priority order of the system real-time resource information and all intent information; determining the task execution information based on the execution unit information and execution thread information corresponding to all the intent information.

[0070] Specifically, the real-time monitoring system monitors the real-time resource information of the system, such as real-time computing power information, real-time memory information, and real-time network bandwidth information. The real-time computing power information may include indicators such as CPU usage and remaining computing power. The real-time memory information may include indicators such as memory usage and remaining memory. The real-time network bandwidth information may include indicators such as current network bandwidth usage and remaining bandwidth.

[0071] According to the priority order of intent information and the real-time resource information of the system, appropriate task execution units and execution threads are allocated to each intent information. Task execution units include CPU cores, memory blocks, etc., and execution threads refer to the execution order of tasks.

[0072] The execution unit information and execution thread information corresponding to all intent information are integrated to generate detailed task execution information. The task execution information may include the execution unit of each task, the required resources (such as CPU time, memory size, network bandwidth, etc.) and the execution order.

[0073] In some embodiments, complex intent information can also be decomposed into multiple subtasks and reasonably allocated to different resources for parallel processing.

[0074] With respect to step S105 , a corresponding task is executed based on the task execution information.

[0075] According to the task execution information, the corresponding tasks are assigned to each execution unit in the system, and the execution unit processes the tasks according to the predetermined process and algorithm. For example, for the task of querying the database, the execution unit connects to the database, performs the query operation, and obtains the result. For the task of controlling the device, the corresponding communication protocol is used to send instructions to the device to realize the control and operation of the device. For the tasks that require calculation, the system's computing resources are used to perform calculations and obtain the results.

[0076] Furthermore, after the execution unit completes the task, the task execution result can be fed back to the user, and the feedback method can be voice broadcast or other interactive forms to meet the needs and usage scenarios of different users.

[0077] See attached Figure 2 , Figure 2 FIG. 1 is a detailed flow chart of a multi-voice intelligent voice interaction method according to another embodiment of the present application. Figure 2 As shown, in this embodiment, the multi-voice intelligent voice interaction method includes the following steps:

[0078] Step S201: Start.

[0079] Step S202: Determine whether there are multiple intelligent voice devices; if yes, execute step S204; if no, execute step S203. Specifically, determine whether the intelligent voice device situation in the interactive environment includes multiple intelligent voice devices.

[0080] Step S203: a single device starts picking up sound.

[0081] Step S204: Multiple devices start picking up sound simultaneously. Specifically, the microphone array of each intelligent voice device continuously monitors the sound in the environment.

[0082] Step S205: Upload the collected voice information to the cloud. Specifically, the intelligent voice device uploads all collected voice information to the cloud, and the cloud processes the voice information.

[0083] Step S206: Determine whether the voice information contains voices of multiple users; if so, execute step S207; if not, execute step S210.

[0084] Step S207: Process the voice information based on the sound features to obtain the voice information of each user.

[0085] Specifically, a speech enhancement algorithm is used to preprocess the acquired voice information. The preprocessing may be to remove background noise, echo elimination, etc., and then the preprocessed voice information is split based on sound features to obtain multiple audio clips with different sound features. The audio clips with the same sound features are synthesized to obtain the voice information corresponding to each user.

[0086] Step S208: Based on the voice information of each user, obtain text information corresponding to the voice information of each user.

[0087] Step S209: Perform intent recognition on each text message respectively to determine the intent information corresponding to each text message.

[0088] Step S210: Process the voice information to obtain text information corresponding to the voice information.

[0089] Specifically, when the voice information does not include the voices of multiple users, the voice information is directly converted into text to obtain corresponding text information.

[0090] Step S211: perform intent recognition on the text information to determine the intent information corresponding to the text information.

[0091] Step S212: Determine whether there are multiple intent information; if so, execute step S213; if not, execute step S218.

[0092] Step S213: Perform intent logic judgment on the intent information based on the preset judgment dimension to determine the intent information that meets the conditions.

[0093] Step S214: Perform a priority evaluation on the intent information that meets the judgment conditions, and determine the priority order of all intent information.

[0094] Step S215: Obtain system real-time resource information.

[0095] Step S216: Determine the task execution information based on the system real-time resource information and the priority order of all intent information.

[0096] Step S217: Execute the corresponding task based on the task execution information. Specifically, the cloud allocates the corresponding task to each execution unit in the system according to the task execution information, and the execution unit processes the task according to the predetermined process and algorithm.

[0097] Step S218: Acquire system real-time resource information. Specifically, when it is determined that there is only single-intent information, directly acquire system real-time resource information.

[0098] Step S219: Allocate tasks to the single-intent information based on the system real-time resource information.

[0099] Step S220: Execute the task of the single-intent information. Specifically, the execution unit executes the task of the single-intent information sent from the cloud according to a predetermined process and algorithm.

[0100] Step S221: Feedback the task execution result.

[0101] Step S222: End.

[0102] It should be pointed out that although the various steps in the above embodiments are described in a specific order, those skilled in the art can understand that in order to achieve the effect of the present application, different steps do not have to be executed in such an order. They can be executed simultaneously (in parallel) or in other orders. These adjusted schemes are equivalent to the technical schemes described in this application, and therefore will also fall within the scope of protection of this application.

[0103] It is understood by those skilled in the art that all or part of the processes in the method for implementing the above-mentioned embodiment of the present application can also be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a computer-readable storage medium, and the computer program can implement the steps of the above-mentioned various method embodiments when executed by the processor. Among them, the computer program includes computer program code, and the computer program code can be in source code form, object code form, executable file or some intermediate form. The computer-readable storage medium may include: any entity or device, medium, U disk, mobile hard disk, disk, optical disk, computer memory, read-only memory, random access memory, electrical carrier signal, telecommunication signal and software distribution medium, etc. that can carry the computer program code.

[0104] Another aspect of the present application provides an electronic device.

[0105] See attached Figure 3 , Figure 3 is a schematic diagram of the main structure of an electronic device according to an embodiment of the present application, Figure 3 FIG. 4 exemplarily shows that the processor 11 and the memory 12 are connected to each other through a bus.

[0106] In an embodiment of an electronic device according to the present application, the electronic device may include at least one processor; and a memory connected to the at least one processor in communication; wherein a computer program is stored in the memory, and when the computer program is executed by at least one processor, the method described in any of the above embodiments is implemented. The electronic device described in the present application may be, but is not limited to, a mobile phone, a tablet computer, a desktop, a laptop, a handheld computer, a notebook computer, a vehicle-mounted device, an ultra-mobile personal computer (UMPC), a netbook, a personal digital assistant (PDA), an augmented reality (AR) / virtual reality (VR) device, etc., and the embodiments of the present application are not limited to this.

[0107] Another aspect of the present application also provides a computer-readable storage medium.

[0108] In an embodiment of a computer-readable storage medium according to the present application, the computer-readable storage medium may be configured to store a program for executing the multi-voice intelligent voice interaction method of the above-mentioned method embodiment, and the program may be loaded and run by the processor to implement the above-mentioned multi-voice intelligent voice interaction method. For ease of explanation, only the parts related to the embodiment of the present application are shown. For specific technical details not disclosed, please refer to the method section of the embodiment of the present application. The computer-readable storage medium may be a storage device including various electronic devices. Optionally, the computer-readable storage medium in the embodiment of the present application is a non-transitory computer-readable storage medium.

[0109] Further, it should be understood that the method of the present application is widely used in smart home (Smart Home), smart home, smart home device ecology, smart residential (Intelligence House) ecology and other whole-house intelligent digital control application scenarios. The above-mentioned network may include but is not limited to at least one of the following: wired network, wireless network. The above-mentioned wired network may include but is not limited to at least one of the following: wide area network, metropolitan area network, local area network, and the above-mentioned wireless network may include but is not limited to at least one of the following: WI FI (Wireless Fidelity, Wireless Fidelity), Bluetooth. Terminal devices may not be limited to PCs, mobile phones, tablet computers, smart air conditioners, smart range hoods, smart refrigerators, smart ovens, smart stoves, smart washing machines, smart water heaters, smart washing equipment, smart dishwashers, smart projection equipment, smart TVs, smart clothes drying racks, smart curtains, smart audio and video, smart sockets, smart speakers, smart speakers, smart fresh air equipment, smart kitchen and bathroom equipment, smart bathroom equipment, smart sweeping robots, smart window cleaning robots, smart mopping robots, smart air purification equipment, smart steamers, smart microwave ovens, smart kitchen treasures, smart purifiers, smart water dispensers, smart door locks, etc.

[0110] So far, the technical solution of the present application has been described in conjunction with an embodiment shown in the accompanying drawings, but it is easy for those skilled in the art to understand that the protection scope of the present application is obviously not limited to these specific embodiments. Without departing from the principles of the present application, those skilled in the art can make equivalent changes or substitutions to the relevant technical features, and the technical solutions after these changes or substitutions will fall within the protection scope of the present application.

Claims

1. A multi-voice intelligent speech interaction method, characterized in that: The method comprises: Get voice information; Determining whether the voice information contains voices of multiple users; In a case where the voice information includes voices of multiple users, determining at least one intention information based on the voice information; Determine task execution information based on preset configuration rules and the intention information, wherein the preset configuration rules include at least one of intention logic judgment, priority evaluation, and system resource allocation; The corresponding task is executed based on the task execution information.

2. The multi-voice intelligent speech interaction method according to claim 1, characterized in that: When the voice information includes voices of multiple users, determining at least one intention information based on the voice information includes: Processing the voice information based on the sound features to obtain voice information corresponding to each user; Inputting the voice information corresponding to each user into a preset voice recognition model to obtain text information of the voice information corresponding to each user; Intent recognition is performed on each of the text messages respectively to determine the intent information corresponding to each of the text messages.

3. The multi-voice intelligent speech interaction method according to claim 2 is characterized in that: The processing of the voice information based on the sound feature to obtain the voice information corresponding to each user includes: Preprocessing the voice information to obtain processed voice information; Splitting the processed voice information based on sound features to obtain multiple audio segments; Audio clips with the same sound features are synthesized to obtain the voice information corresponding to each user.

4. The multi-voice intelligent speech interaction method according to claim 2, characterized in that: The performing intent recognition on each of the text messages respectively to determine the intent information corresponding to each of the text messages includes: Each of the text information is input into a preset intention recognition model, and lexical analysis, syntactic analysis, semantic understanding and classification processing are performed on each of the text information to determine the intention information corresponding to each of the text information.

5. The multi-voice intelligent speech interaction method according to claim 1, characterized in that: The determining of the task execution information based on the preset configuration rule and the intention information includes: Performing intention logic judgment on each of the intention information to determine the intention information that meets the judgment conditions; Performing a priority evaluation on the intent information that meets the judgment condition, and determining the priority order of all intent information; Get real-time resource information of the system; Based on the system real-time resource information and the priority order of all intent information, system resources are allocated to the intent information to determine task execution information.

6. The multi-voice intelligent speech interaction method according to claim 5, characterized in that: The performing intention logic judgment on each of the intention information to determine the intention information that meets the judgment condition includes: Based on preset judgment dimensions, intent logic judgment is performed on each of the intent information to determine the intent information that meets the judgment conditions. The preset judgment dimensions include at least one of timestamp judgment, logical relationship judgment, and conflict judgment.

7. The multi-voice intelligent speech interaction method according to claim 5, characterized in that: The performing priority evaluation on the intention information meeting the judgment condition to determine the priority order of all the intention information includes: Based on preset priority evaluation rules, priority evaluation is performed on each of the intent information to determine the priority order of all intent information, and the preset priority evaluation rules include at least one of urgency, importance and user authority.

8. The multi-voice intelligent speech interaction method according to claim 5, characterized in that: The system real-time resource information includes at least one of real-time computing capability information, real-time memory information, and real-time network bandwidth information; The method of allocating system resources to the intention information based on the system real-time resource information and the priority order of all the intention information, and determining the task execution information, includes: Based on the system real-time resource information and the priority order of all intent information, determining the execution unit information and execution thread information corresponding to each of the intent information; Task execution information is determined based on the execution unit information and execution thread information corresponding to all the intention information.

9. An electronic device comprising at least one memory and at least one processor, characterized in that: The memory is suitable for storing multiple program codes, and is characterized in that the program codes are suitable for being loaded and run by the processor to execute the multi-voice intelligent speech interaction method described in any one of claims 1 to 8.

10. A computer-readable storage medium storing a plurality of program codes, characterized in that: The program code is suitable for being loaded and run by a processor to execute the multi-voice intelligent speech interaction method described in any one of claims 1 to 8.

Citation Information

Cited By

  • Voice call management method, device, equipment, storage medium and product

    CN119788121A

  • Multi-player voice game performance optimization method and device based on dynamic resource management

    CN120045337A

  • Household equipment control method and device, equipment and medium

    CN120353346A

  • User intention scheduling method and device, storage medium and electronic device

    CN120415934A