Voice interaction method and electronic equipment
By analyzing user intentions and person information in voice content in advance, the problem of poor performance of intelligent assistants after detecting wake-up information is solved, achieving more efficient response and processing efficiency.
Patent Information
- Application Number
- CN202411273418.X
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-09-11
- Publication Date
- 2024-11-19
AI Technical Summary
Existing smart assistants can only perform voice content analysis after detecting wake-up information, resulting in poor performance and inability to analyze user intentions in advance, affecting response efficiency.
By obtaining voice content, determine the person information and user intentions corresponding to the voice, and perform corresponding operations when the wake-up information is detected, analyze and respond to user intentions in advance.
It improves the performance and processing efficiency of the smart assistant, realizes user intention analysis and response operations before detecting wake-up information, and achieves chat effect.
Smart Images

Figure CN118982984A_ABST
Abstract
Description
Technical Field
[0001] The present application relates to the field of artificial intelligence technology, and in particular to a voice interaction method and electronic device. Background Art
[0002] Smart assistants are artificial intelligence products commonly used in people's lives. They can provide users with convenient services based on the voice content input by users, such as opening a device or an application for users based on the voice content input by users, or controlling a device or an application based on the voice content input by users. In some related technologies, smart assistants collect the voice content input by users after detecting the wake-up information, analyze the voice content, and respond based on the analysis results. In this way, since the voice content can only be analyzed after the wake-up information is detected, the performance of smart assistants is poor. Summary of the invention
[0003] The embodiments of the present application provide a voice interaction method and an electronic device to solve the performance problem of intelligent assistants existing in the related art.
[0004] In a first aspect, an embodiment of the present application provides a voice interaction method, comprising:
[0005] Get voice content;
[0006] Determining character information corresponding to the voice content based on the voice content;
[0007] Determining the user's intention for the intelligent assistant based on the character information and the voice content;
[0008] When the wake-up information of the intelligent assistant is detected, a response operation is performed, where the response operation is a response operation of the intelligent assistant in response to the user's intention.
[0009] In a second aspect, an embodiment of the present application provides a voice interaction device, including:
[0010] An acquisition module, used to acquire voice content;
[0011] A first determination module, configured to determine character information corresponding to the voice content based on the voice content;
[0012] A second determination module, configured to determine the user's intention for the intelligent assistant based on the character information and the voice content;
[0013] A control module is used to perform a response operation when the wake-up information of the intelligent assistant is detected, and the response operation is a response operation of the intelligent assistant in response to the user's intention.
[0014] In a third aspect, an embodiment of the present application provides an electronic device, including:
[0015] A processor, used for acquiring voice content;
[0016] The processor is further configured to determine character information corresponding to the voice content based on the voice content;
[0017] The processor is further configured to determine a user intention for the intelligent assistant based on the character information and the voice content;
[0018] The processor is also used to perform a response operation when the wake-up information of the intelligent assistant is detected, and the response operation is a response operation of the intelligent assistant in response to the user's intention.
[0019] In a fourth aspect, an embodiment of the present application further provides an electronic device, comprising: a processor, a memory, and a program stored in the memory and executable on the processor, wherein when the program is executed by the processor, the steps of the voice interaction method provided in the embodiment of the present application are implemented.
[0020] In a fifth aspect, an embodiment of the present application further provides a computer-readable storage medium, on which a computer program is stored, and when the computer program is executed by a processor, the steps of the voice interaction method provided in the embodiment of the present application are implemented.
[0021] In a sixth aspect, an embodiment of the present application also provides a computer program product, comprising computer instructions, which, when executed by a processor, implement the steps of the voice interaction method provided in the embodiment of the present application.
[0022] In an embodiment of the present application, voice content is obtained; character information corresponding to the voice content is determined based on the voice content; user intent for the intelligent assistant is determined based on the character information and the voice content; and a response operation is performed when the wake-up information of the intelligent assistant is detected, wherein the response operation is a response operation of the intelligent assistant in response to the user intent. In this way, the user intent can be analyzed before the wake-up information is detected, and when the wake-up information of the intelligent assistant is detected, the response operation corresponding to the user intent is performed, thereby improving the performance of the intelligent assistant. BRIEF DESCRIPTION OF THE DRAWINGS
[0023] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings required for use in the description of the embodiments of the present application will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative labor.
[0024] Figure 1 is a flow chart of a voice interaction method provided in an embodiment of the present application;
[0025] Figure 2 is an example schematic diagram of a voice interaction method provided in an embodiment of the present application;
[0026] Figure 3 is a structural diagram of a voice interaction device provided in an embodiment of the present application;
[0027] Figure 4 It is a structural diagram of an electronic device provided in an embodiment of the present application. DETAILED DESCRIPTION
[0028] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are 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 are within the scope of protection of this application.
[0029] See also Figure 1 , Figure 1 is a flow chart of a voice interaction method provided in an embodiment of the present application, such as Figure 1 As shown, the following steps are included:
[0030] Step 101: Acquire voice content.
[0031] The above steps may be to collect or monitor the voice content of the user, which may be the voice content of one or more users, such as a conversation between multiple users. For example, the voice content of the user may be monitored by an artificial intelligence (AI) monitoring module, such as periodically or continuously monitoring the voice content of the user.
[0032] Step 102: Determine character information corresponding to the voice content based on the voice content.
[0033] The above-mentioned character information may be the character corresponding to the above-mentioned voice content, such as the user of the electronic device, the owner of the house where the electronic device is located, or the above-mentioned character information may be the character relationship between multiple users corresponding to the above-mentioned voice content, such as the relationship between father and son, the relationship between the host and the neighbor, the relationship between the host and the guest, etc.
[0034] In some implementations, the person corresponding to the voice content may be determined based on the voice features in the voice content, such as identifying the person corresponding to the voice content based on the timbre or voiceprint in the voice content.
[0035] Step 103: Determine the user intention for the intelligent assistant based on the character information and the voice content.
[0036] In the embodiments of the present application, the intelligent assistant may be an independent electronic device, such as an intelligent speaker, an intelligent home control device, an intelligent robot or other intelligent device, or the intelligent assistant may be software installed on the electronic device. In addition, the intelligent assistant may be a voice assistant or a conversational robot.
[0037] The above-mentioned user intention for the smart assistant can be understood as the intention to control or command the above-mentioned smart assistant, such as controlling or commanding the above-mentioned smart assistant to turn on the TV, play XX song, adjust the air-conditioning temperature, etc.
[0038] The above-mentioned determination of the user intention for the intelligent assistant based on the character information and the voice content can be that the voice of a specific character is filtered out from the above-mentioned voice content based on the character information, and then the user intention for the intelligent assistant is determined based on the voice of the specific character; or it can be that the voice of a specific character relationship is filtered out from the above-mentioned voice content based on the above-mentioned character information, and then the user intention for the intelligent assistant is determined based on the voice.
[0039] Step 104: When the wake-up information of the intelligent assistant is detected, a response operation is performed, where the response operation is a response operation of the intelligent assistant in response to the user's intention.
[0040] The above wake-up information may be a wake-up word.
[0041] The above-mentioned detection of the wake-up information of the intelligent assistant may be detection of the wake-up information input by the user, such as the user's voice including the above-mentioned wake-up information.
[0042] The above-mentioned execution of the response operation may be that the electronic device controls the intelligent assistant to execute the above-mentioned response operation, or when the above-mentioned intelligent assistant is an electronic device, the intelligent assistant may execute the above-mentioned response operation.
[0043] In the embodiment of the present application, the above steps can be used to analyze the user's intention before detecting the wake-up information, and when the wake-up information of the intelligent assistant is detected, the intelligent assistant is controlled to perform the response operation corresponding to the user's intention, thereby improving the performance of the intelligent assistant. And it can also improve the processing efficiency and achieve the effect of chatting. For example: User A says: User B, please turn on the music of the intelligent assistant; User B says: OK, and calls the intelligent assistant, so that after user B calls the intelligent assistant, he can directly control the intelligent assistant to turn on the music, without having to tell the intelligent assistant to turn on the music after user B calls the intelligent assistant.
[0044] In the embodiment of the present application, the voice interaction method can be applied to an electronic device, which can be an intelligent assistant, or the electronic device is installed with an intelligent assistant. Specifically, the steps of the voice interaction method are performed by the electronic device.
[0045] In one embodiment, the voice content includes voices of multiple users, and determining character information corresponding to the voice content based on the voice content includes:
[0046] Identify voice features of the multiple users based on the voice content, and identify character relationship information of the multiple users according to the voice features of the multiple users;
[0047] The character information includes the character relationship information.
[0048] The voice features may be voice features such as timbre or voiceprint that can distinguish the voices of different users.
[0049] The above-mentioned character relationship information may include but is not limited to at least one of the following:
[0050] the character relationships between the host and his neighbors;
[0051] The relationship between the father and son characters;
[0052] The relationship between siblings.
[0053] The above-mentioned identifying the voice features of the multiple users based on the voice content and identifying the character relationship information of the multiple users based on the voice features of the multiple users may be to identify the user's body information based on the voice features of the multiple users, and then determine to identify the character relationship information of the multiple users based on the identity information. For example: the voice features of the owner and the owner's children may be pre-entered or recorded, and the identities of the users may be identified based on the voice features; or the voice features of the owner and the owner's children may be identified based on historical voice records.
[0054] In this implementation, since the character information includes the character relationship information, it is possible to determine the user intention for the intelligent assistant based on the character relationship information, thereby improving the accuracy of the user intention. For example, the owner's voice can be screened out from the above voice content based on the character relationship information, and the user intention can be identified more accurately based on the voice.
[0055] It should be noted that in the embodiments of the present application, the voice content is not limited to include the voices of multiple users. For example, in some implementations, it may also be the voice of a single user.
[0056] In one embodiment, the identifying voice features of the multiple users based on the voice content, and identifying character relationships among the multiple users according to the voice features of the multiple users, includes:
[0057] recognizing speech features of the multiple users based on the speech content, and determining a weight of each speech feature, determining at least one character relationship corresponding to the multiple users according to the speech features of the multiple users, and calculating a character relationship weight value of each character relationship according to the weights of the speech features of the users involved in the at least one character relationship;
[0058] The character relationship information includes the at least one character relationship and the character relationship weight value of each character relationship.
[0059] The weight of each voice feature may be preset, or may be set according to the user identity information of the voice feature. For example, as shown in Table 1 below:
[0060] Table 1:
[0061] Serial number Character code Relationships timbre Whether to analyze Weight 1 M Owner F(M) yes Q(M)=1 2 D1 Child F(D1) yes Q(D1)=0.5 3 D2 Neighbor F(D2) no Q(D2)=0
[0062] As shown in Table 1 above, the weight of the owner's voice feature is 1, the weight of the child's voice feature is 0.5, and the weight of the neighbor's voice feature is 0.
[0063] It should be noted that the above Table 1 is only an example of weights. The specific weights are not limited in this embodiment and can be set according to actual needs.
[0064] Among them, the higher the weight, the smart assistant takes the voice as the standard. As shown in the table above, when the timbre of M and D1 appears at the same time, the timbre of M shall prevail. In some embodiments, considering that when there are many characters in the scene, useless timbre is filtered out to avoid using unnecessary storage and computing resources, and at the same time, the intention of the main user is better and clearer.
[0065] The "Whether to Analyze" field in Table 1 above can indicate whether to process the timbre when the smart assistant receives it. If not analyzed, the smart assistant will automatically discard the timbre voice to save storage and computing resources.
[0066] The above-mentioned calculation of the character relationship weight value of each character relationship based on the weight of the voice features of the users involved in the at least one character relationship can be to multiply the weights of the voice features of the users involved in the character association, and the result is the character relationship weight value, such as the character relationship weight value of father and son is 1*0.5=0.5, and the character relationship weight value of the owner and the neighbor is 1*0=0.
[0067] Taking Table 1 above as an example, the character relationship weight value can be obtained based on the following matrix:
[0068]
[0069] The above matrix is explained in Table 2 as follows:
[0070] Table 2:
[0071]
[0072]
[0073] W i represents a character node, W ij For example, the weight value of the relationship between two characters is expressed as M, and the weight value of the relationship between father and son is expressed as W. 12 =0.5.
[0074] In the above implementation, since the character relationship information includes at least one character relationship and the character relationship weight value of each character relationship, the voice recognition user intention corresponding to the character relationship with a high character relationship weight value can be selected, thereby increasing the reliability of the user intention.
[0075] In one embodiment, determining the user intention for the intelligent assistant based on the character information and the voice content includes:
[0076] Determine the voice corresponding to the target character relationship in the voice content, and determine the user intention for the intelligent assistant based on the voice corresponding to the target character relationship, wherein the target character relationship is a character relationship whose character relationship weight value is higher than a preset threshold.
[0077] The preset threshold may be set according to actual scene requirements, such as 0, 0.1 or 0.4.
[0078] Since the target character relationship is a character relationship whose character relationship weight value is higher than the preset threshold, this indicates that the target character relationship is an important character relationship in the above voice content, and thus the user intention is determined based on the voice corresponding to the important character relationship, which can improve the reliability of the user intention. For example: the number of characters in the above voice content is identified through the above steps, such as identifying that there are 3 people in the space: M (owner), D1 (child), D2 (neighbor). Among them, the weight of D2 is 0, and the voice information of D2 is directly ignored, Q(M)>Q(D1), and the voice input information of M is processed and analyzed first, and since the characters of M and D1 are identified, and the relationship W between M and D1 is 12 =0.5>0, then the speech of M and D1 is analyzed first, that is, the dialogue relationship between M and D1.
[0079] In addition, since the user intention for the intelligent assistant is determined based on the voice corresponding to the target person relationship, a large amount of useless information can be filtered out, which greatly reduces the storage and computing power.
[0080] It should be noted that the embodiments of the present application are not limited to determining the user intention for the intelligent assistant based on the voice corresponding to the target person relationship. For example, in some embodiments, the user intention for the intelligent assistant can also be determined based on the voice corresponding to the person relationship with the highest person relationship weight value.
[0081] In one embodiment, determining the user intention for the intelligent assistant based on the character information and the voice content includes:
[0082] Based on the character information, dialogue indication information is identified from the voice content, and user intention for the intelligent assistant is identified based on the dialogue indication information.
[0083] The above-mentioned dialogue instruction information may be dialogue instruction information such as imperative language, imperative verbs or answering dialogue, or the above-mentioned dialogue instruction information may include two of the imperative language, imperative verbs and answering dialogue. For example, user A says: user B, please turn on the music of the smart assistant; user B says: OK, then the dialogue instruction information indicates to turn on the music of the smart assistant.
[0084] In some implementations, conversation indication information of speech content may be identified based on a deep learning model.
[0085] In this implementation, since the user intention to the intelligent assistant is identified based on the dialogue instruction information, the accuracy of the user intention can be improved. In addition, the dialogue instruction information can be determined based on the above character relationship information, so that the dialogue relationship of the characters can be analyzed through the analysis of the character relationship, so that the intelligent assistant can further learn the instruction relationship of the spatial characters, and truly achieve the effect of multi-person dialogue.
[0086] It should be noted that the embodiments of the present application are not limited to identifying the user intention for the intelligent assistant based on the dialogue indication information. In some implementations, the user intention for the intelligent assistant can also be determined based on keywords in the above-mentioned voice content.
[0087] In one embodiment, when the wake-up information of the intelligent assistant is detected, performing a response operation includes:
[0088] When the wake-up information of the intelligent assistant is detected, an inquiry instruction for inquiring whether to execute a response operation corresponding to the user intention is output;
[0089] When an instruction to execute the response operation is received, the response operation is executed.
[0090] The above-mentioned output for asking whether to execute the inquiry instruction corresponding to the user intention can be a voice playing for asking whether to execute the inquiry instruction corresponding to the user intention, or a display information for asking whether to execute the inquiry instruction corresponding to the user intention.
[0091] For example: output inquiry instructions on whether to turn on the TV, play XX song, or adjust the air conditioner temperature.
[0092] The instruction to execute the response operation may be an instruction voice or an instruction operation to instruct the execution of the response operation.
[0093] In this implementation, the reliability of executing the response operation can be improved by querying the indication.
[0094] The embodiments of the present application are not limited to outputting the above inquiry indication. For example, in some implementations, the above response operation may also be directly executed to improve efficiency.
[0095] In one embodiment, the method further comprises:
[0096] The user intention is stored, and the voice content is deleted.
[0097] Among them, the above-mentioned storage of the user intention can be stored in an overwriting manner, that is, only one user intention is stored, or the user intentions are stored in chronological order, and when performing a response operation, the response operation is performed according to the most recent user intention.
[0098] In this embodiment, since the user intention is stored, deleting the voice content can save storage resources and significantly reduce the storage capacity.
[0099] It should be noted that, in the embodiment of the present application, the voice obtained by the user may be obtained with the user's prior authorization or prior notification to the user.
[0100] In an embodiment of the present application, voice content is obtained; character information corresponding to the voice content is determined based on the voice content; user intent for the intelligent assistant is determined based on the character information and the voice content; and a response operation is performed when the wake-up information of the intelligent assistant is detected, wherein the response operation is a response operation of the intelligent assistant in response to the user intent. In this way, the user intent can be analyzed before the wake-up information is detected, and when the wake-up information of the intelligent assistant is detected, the response operation corresponding to the user intent is performed, thereby improving the performance of the intelligent assistant.
[0101] Below Figure 2 Take the example of the voice interaction method provided in the embodiment of the present application as an example:
[0102] The user inputs the voice, that is, the AI monitoring template obtains the voice content. If it is not a wake-up word, the voice preprocessing, voice processing / analysis and user intention storage are performed in sequence. The process specifically includes character relationship entry / recognition, user timbre entry / recognition, voice storage, common language recognition / storage, common dialogue time analysis, and spatial character quantity analysis. Among them, common language recognition / storage and common dialogue time analysis are used to identify dialogue indication information, and spatial character quantity analysis is used to identify character relationship information. Historical information input means obtaining the stored user intention. In this way, when the wake-up word is received, it is determined whether the user intention is stored, that is, historical information analysis. If the user intention is stored, it means that the historical information matches, thereby generating reply information, that is, the inquiry indication in the above embodiment, and specifically feedback to the user through machine voice. When determining whether to perform a response operation, if the user intention is not stored, and other analysis modes are adopted, it can be specifically adopted to use the common listening monitoring mode (such as monitoring voice according to a preset period) for monitoring, and voice recognition / analysis of the monitored voice, and feedback the results to the user. In this way, the user intention can be analyzed in advance to improve the performance of the intelligent assistant.
[0103] See also Figure 3 , Figure 3 is a structural diagram of a voice interaction device provided in an embodiment of the present application, such as Figure 3 As shown, the voice interaction device 300 includes:
[0104] An acquisition module 301 is used to acquire voice content;
[0105] A first determination module 302, configured to determine character information corresponding to the voice content based on the voice content;
[0106] A second determination module 303, configured to determine the user's intention for the intelligent assistant based on the character information and the voice content;
[0107] The control module 304 is used to perform a response operation when the wake-up information of the intelligent assistant is detected, and the response operation is a response operation of the intelligent assistant in response to the user's intention.
[0108] Optionally, the voice content includes voices of multiple users, and the first determination module 302 is used to: identify voice features of the multiple users based on the voice content, and identify character relationship information of the multiple users according to the voice features of the multiple users;
[0109] The character information includes the character relationship information.
[0110] Optionally, the first determination module 302 is used to: identify voice features of the multiple users based on the voice content, and determine the weight of each voice feature, determine at least one character relationship corresponding to the multiple users according to the voice features of the multiple users, and calculate the character relationship weight value of each character relationship according to the weight of the voice features of the users involved in the at least one character relationship;
[0111] The character relationship information includes the at least one character relationship and a character relationship weight value of each character relationship.
[0112] Optionally, the second determination module 303 is used to: determine the voice corresponding to the target character relationship in the voice content, and determine the user intention for the intelligent assistant based on the voice corresponding to the target character relationship, wherein the target character relationship is a character relationship whose character relationship weight value is higher than a preset threshold.
[0113] Optionally, the second determination module 303 is used to: identify dialogue indication information from the voice content based on the character information, and identify the user's intention for the intelligent assistant based on the dialogue indication information.
[0114] Optionally, the control module 304 is used to: when detecting the wake-up information of the intelligent assistant, output an inquiry instruction for inquiring whether to execute the response operation corresponding to the user's intention; when receiving an instruction to execute the response operation, execute the response operation.
[0115] Optionally, the device further comprises:
[0116] A processing module is used to store the user intention and delete the voice content.
[0117] The above-mentioned voice interaction device is capable of implementing each process of each embodiment of the above-mentioned voice interaction method. The technical features correspond one to one and can achieve the same technical effect. To avoid repetition, they will not be described here.
[0118] See also Figure 4 , Figure 4 is a structural diagram of an electronic device provided in an embodiment of the present application, such as Figure 4 As shown, the electronic device 400 includes: a bus 401 , a transceiver 402 , an antenna 403 , a bus interface 404 , a processor 405 and a memory 406 .
[0119] exist Figure 4In the embodiment, a bus architecture (represented by bus 401) is shown, bus 401 may include any number of interconnected buses and bridges, bus 401 links various circuits including one or more processors represented by processor 405 and memory represented by memory 406. Bus 401 may also link various other circuits such as peripherals, voltage regulators, and power management circuits, which are well known in the art and are therefore not further described herein. Bus interface 404 provides an interface between bus 401 and transceiver 402. Transceiver 402 may be one element or multiple elements, such as multiple receivers and transmitters, providing a unit for communicating with various other devices on a transmission medium. Data processed by processor 405 is transmitted on a wireless medium via antenna 403, and further, antenna 403 also receives data and transmits the data to processor 405.
[0120] Processor 405 is responsible for managing bus 401 and general processing, and may also provide various functions, including timing, peripheral interfaces, voltage regulation, power management, and other control functions. Memory 406 may be used to store data used by processor 405 when performing operations.
[0121] Optionally, the processor 405 may be a CPU, an ASIC, an FPGA or a CPLD.
[0122] The processor 405 is used to obtain voice content;
[0123] The processor 405 is further configured to determine character information corresponding to the voice content based on the voice content;
[0124] The processor 405 is further configured to determine the user intention for the intelligent assistant based on the character information and the voice content;
[0125] The processor 405 is also used to perform a response operation when the wake-up information of the intelligent assistant is detected, and the response operation is a response operation of the intelligent assistant in response to the user's intention.
[0126] Optionally, the voice content includes voices of multiple users, and determining character information corresponding to the voice content based on the voice content includes:
[0127] Identify voice features of the multiple users based on the voice content, and identify character relationship information of the multiple users according to the voice features of the multiple users;
[0128] The character information includes the character relationship information.
[0129] Optionally, the identifying voice features of the multiple users based on the voice content, and identifying character relationships among the multiple users according to the voice features of the multiple users, includes:
[0130] recognizing speech features of the multiple users based on the speech content, and determining a weight of each speech feature, determining at least one character relationship corresponding to the multiple users according to the speech features of the multiple users, and calculating a character relationship weight value of each character relationship according to the weights of the speech features of the users involved in the at least one character relationship;
[0131] The character relationship information includes the at least one character relationship and a character relationship weight value of each character relationship.
[0132] Optionally, determining the user intention for the intelligent assistant based on the character information and the voice content includes:
[0133] Determine the voice corresponding to the target character relationship in the voice content, and determine the user intention for the intelligent assistant based on the voice corresponding to the target character relationship, wherein the target character relationship is a character relationship whose character relationship weight value is higher than a preset threshold.
[0134] Optionally, determining the user intention for the intelligent assistant based on the character information and the voice content includes:
[0135] Based on the character information, dialogue indication information is identified from the voice content, and user intention for the intelligent assistant is identified based on the dialogue indication information.
[0136] Optionally, when the wake-up information of the intelligent assistant is detected, performing a response operation includes:
[0137] When the wake-up information of the intelligent assistant is detected, an inquiry instruction for inquiring whether to execute a response operation corresponding to the user intention is output;
[0138] When an instruction to execute the response operation is received, the response operation is executed.
[0139] Optionally, the processor 405 is further configured to:
[0140] The user intention is stored, and the voice content is deleted.
[0141] The above-mentioned electronic device is capable of implementing each process of each embodiment of the above-mentioned voice interaction method, the technical features correspond one to one, and can achieve the same technical effect. To avoid repetition, they will not be repeated here.
[0142] The present application also provides a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, the above Figure 1 The various processes of the voice interaction method embodiment shown in the figure can achieve the same technical effect, and will not be repeated here to avoid repetition. Among them, the computer-readable storage medium, such as read-only memory (ROM), random access memory (RAM), disk or optical disk, etc.
[0143] The embodiment of the present application also provides an electronic device, comprising: a processor, a memory, and a program stored in the memory and executable on the processor, wherein the program is executed by the processor to implement the above Figure 1 The various processes of the voice interaction method embodiment shown can achieve the same technical effect, and will not be described again here to avoid repetition.
[0144] The present application also provides a computer program product, including computer instructions, which are executed by a processor to implement the following Figure 1 The various processes of the voice interaction method embodiment shown can achieve the same technical effect, and will not be described again here to avoid repetition.
[0145] It should be noted that, in this article, the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, an element defined by the sentence "comprises a ..." does not exclude the existence of other identical elements in the process, method, article or device including the element.
[0146] Through the description of the above implementation methods, those skilled in the art can clearly understand that the above-mentioned embodiment methods can be implemented by means of software plus a necessary general hardware platform, and of course by hardware, but in many cases the former is a better implementation method. Based on such an understanding, the technical solution of the present application, or the part that contributes to the prior art, can be embodied in the form of a software product, which is stored in a storage medium (such as ROM / RAM, a disk, or an optical disk), and includes a number of instructions for a terminal (which can be a mobile phone, a computer, a server, an air conditioner, or a network device, etc.) to execute the methods of each embodiment of the present application.
[0147] The embodiments of the present application are described above in conjunction with the accompanying drawings, but the present application is not limited to the above-mentioned specific implementation methods. The above-mentioned specific implementation methods are merely illustrative and not restrictive. Under the guidance of the present application, ordinary technicians in this field can also make many forms without departing from the purpose of the present application and the scope of protection of the claims, all of which are within the protection of the present application.
Claims
1. A voice interaction method, characterized in that: include: Get voice content; Determining character information corresponding to the voice content based on the voice content; Determining the user's intention for the intelligent assistant based on the character information and the voice content; When the wake-up information of the intelligent assistant is detected, a response operation is performed, where the response operation is a response operation of the intelligent assistant in response to the user's intention.
2. The method according to claim 1, characterized in that The voice content includes voices of multiple users, and determining character information corresponding to the voice content based on the voice content includes: Identify voice features of the multiple users based on the voice content, and identify character relationship information of the multiple users according to the voice features of the multiple users; The character information includes the character relationship information.
3. The method according to claim 2, characterized in that The identifying the voice features of the multiple users based on the voice content, and identifying the character relationships of the multiple users according to the voice features of the multiple users, includes: recognizing speech features of the multiple users based on the speech content, and determining a weight of each speech feature, determining at least one character relationship corresponding to the multiple users according to the speech features of the multiple users, and calculating a character relationship weight value of each character relationship according to the weights of the speech features of the users involved in the at least one character relationship; The character relationship information includes the at least one character relationship and a character relationship weight value of each character relationship.
4. The method according to claim 3, characterized in that The determining the user intention for the intelligent assistant based on the character information and the voice content includes: Determine the voice corresponding to the target character relationship in the voice content, and determine the user intention for the intelligent assistant based on the voice corresponding to the target character relationship, wherein the target character relationship is a character relationship whose character relationship weight value is higher than a preset threshold.
5. The method according to any one of claims 1 to 4, characterized in that The determining the user intention for the intelligent assistant based on the character information and the voice content includes: Based on the character information, dialogue indication information is identified from the voice content, and user intention for the intelligent assistant is identified based on the dialogue indication information.
6. The method according to any one of claims 1 to 4, characterized in that When the wake-up information of the intelligent assistant is detected, performing a response operation includes: When the wake-up information of the intelligent assistant is detected, outputting an inquiry instruction for inquiring whether to execute a response operation corresponding to the user intention; When an instruction to execute the response operation is received, the response operation is executed.
7. A voice interaction device, characterized in that: include: An acquisition module, used to acquire voice content; A first determination module, configured to determine character information corresponding to the voice content based on the voice content; A second determination module, configured to determine the user's intention for the intelligent assistant based on the character information and the voice content; A control module is used to perform a response operation when the wake-up information of the intelligent assistant is detected, and the response operation is a response operation of the intelligent assistant in response to the user's intention.
8. An electronic device, characterized in that: include: A processor, a memory, and a program stored in the memory and executable on the processor, wherein the program implements the steps of the method according to any one of claims 1 to 6 when executed by the processor.
9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 6 are implemented.
10. A computer program product, characterized in that The method comprises computer instructions, which implement the steps of the method according to any one of claims 1 to 6 when executed by a processor.
