Voice interaction method, storage medium, electronic device and system

Abstract

The invention discloses a voice interaction method, a storage medium, an electronic device and a system, and relates to the technical field of smart home, and the voice interaction method comprises the steps: obtaining the attribute information of a plurality of attributes of a voice interaction instruction of a user, classifying the attribute information of each attribute according to a preset standard, obtaining the attribute classification result of each attribute, and storing the attribute classification result of each attribute; obtaining user emotion information of the voice interaction instruction according to an attribute classification result of the at least one attribute; and determining a feedback mode of the voice interaction instruction according to the user emotion information. According to the method, the user emotion information can be obtained through analysis of the attribute information of the multiple attributes of the voice interaction instruction of the user, and the feedback mode of the voice interaction instruction reply content can be flexibly adjusted according to the user emotion requirement. In this way, the feedback mode of the voice interaction instruction is determined according to the user emotion information, the emotion trouble caused by the improper feedback mode of the voice interaction instruction to the user can be avoided, and the interaction experience of the user is improved.

Description

Technical Field

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

[0002] Intelligent voice interaction is a new generation of interaction mode based on voice input, which can get feedback results by speaking. Thanks to the advancement of technology, the expansion of application scenarios, the improvement of user experience and the support of industrial policies, intelligent voice interaction applications have developed rapidly. With the gradual popularization and promotion of various types of intelligent voice interaction devices, users can communicate freely with smart devices through voice interaction operating systems, gradually achieving the goal of smart life.

[0003] At present, in daily voice interaction, smart devices can already achieve non-differentiated reply exchanges, and can provide instant interactive services for any different users, solve user output problems and provide instant feedback to users. However, in actual application scenarios, when providing feedback on user control and interaction, the voice interaction function of smart devices depends on the user's own instructions. When users are in various emotions, smart devices cannot judge the user's emotions and give corresponding emotional care, which often makes the voice interaction experience less intelligent, and even causes emotional distress and harm to users due to inappropriate reply content or reply tone, affecting the user experience.

[0004] Accordingly, a new solution is needed in the art to solve the above problems. Summary of the invention

[0005] The present application aims to solve the above-mentioned technical problem, namely, to solve the problem that the existing voice interaction function cannot judge the user's emotions, resulting in poor user voice interaction experience.

[0006] In a first aspect, the present application provides a voice interaction method, comprising:

[0007] Obtaining attribute information of various attributes of the user's voice interaction command;

[0008] Classify the attribute information of each attribute according to the preset standard to obtain an attribute classification result of each attribute;

[0009] Obtaining user emotion information of the voice interaction instruction according to an attribute classification result of at least one of the attributes;

[0010] Determine a feedback method of the voice interaction instruction according to the user emotion information.

[0011] In a technical solution of the above-mentioned voice interaction method, the attributes of the voice instruction include voice parameters, and the attribute classification result includes a voice parameter label;

[0012] The classifying the attribute information of each attribute according to the preset standard to obtain the attribute classification result of each attribute includes:

[0013] The attribute information of the voice parameters of the voice interaction instruction is classified according to preset standards to obtain voice parameter labels of the voice parameters; the voice parameters include at least one of speech rate, intonation, volume and semantics, and the voice parameter labels include at least one of speech rate labels, intonation labels, volume labels and voice labels.

[0014] In a technical solution of the above-mentioned voice interaction method, if the voice parameter is speech rate, the attribute information of the voice parameter is the actual speech rate value of the voice interaction instruction, and the voice parameter label is the speech rate label;

[0015] The acquiring of the voice parameter tag of the voice parameter comprises:

[0016] Determining a preset standard speaking speed according to the preset standard;

[0017] Determining whether the actual speaking speed value exceeds the preset standard speaking speed;

[0018] If the actual speaking speed value exceeds the preset standard speaking speed, the speaking speed label is determined as an urgent speaking speed label;

[0019] If the actual speaking speed value does not exceed the preset standard speaking speed, the speaking speed label is determined as a slow speaking speed label.

[0020] In a technical solution of the above-mentioned voice interaction method, if the voice parameter is intonation, the attribute information of the voice parameter is the actual intonation value of the voice interaction instruction, and the voice parameter label is the intonation label;

[0021] The acquiring of the voice parameter tag of the voice parameter comprises:

[0022] Determining a standard intonation range according to the preset standard;

[0023] Determining whether the actual intonation value is within the standard intonation range;

[0024] If the actual speaking speed value is within the standard intonation range, determining the intonation label as a normal intonation label;

[0025] If the actual intonation value is greater than the upper limit of the standard intonation range, the intonation label is determined as a high-intonation label.

[0026] In a technical solution of the above-mentioned voice interaction method, if the voice parameter is semantic, the attribute information of the voice parameter is the semantic analysis result of the voice interaction instruction, and the voice parameter label is a semantic label;

[0027] The acquiring of the voice parameter tag of the voice parameter comprises:

[0028] Determine emotion classification keywords according to the preset criteria;

[0029] Classifying the semantic analysis result according to the emotion classification keyword to obtain the semantic label, wherein the semantic analysis result is an analysis result obtained by performing semantic analysis and text sentiment analysis on the voice interaction instruction using natural language processing technology;

[0030] The semantic labels include a no emotion fluctuation label, an emotion fluctuation label and a severe emotion fluctuation label.

[0031] In a technical solution of the above-mentioned voice interaction method, the step of determining the feedback mode of the voice interaction instruction according to the user emotion information includes:

[0032] Obtaining a preset user emotion judgment condition set, and judging whether the user emotion information contains a label in the user emotion judgment condition set according to the label in the user emotion judgment condition set;

[0033] If at least one label in the user emotion judgment condition set exists in the user emotion information, adjusting the feedback mode of the voice interaction instruction according to the label;

[0034] If any label in the user emotion judgment condition set does not exist in the user emotion information, the reply content is determined according to the user's voice interaction instruction, and the reply content is fed back to the user in a default feedback manner.

[0035] In a technical solution of the above-mentioned voice interaction method, the user emotion judgment condition set includes labels of urgent speaking speed, high tone, loud volume and emotional fluctuation;

[0036] The step of adjusting the feedback mode of the voice interaction instruction according to the label includes:

[0037] If the tag includes a tag of urgent speech speed or high tone, extract key information in the reply text corresponding to the voice interaction instruction as the final reply content, determine that the feedback mode of the reply content is a fast feedback mode, and the speech speed value of the fast feedback mode is greater than the speech speed value of the default feedback mode;

[0038] If the labels include labels with a relatively low volume, adjusting the volume value of the feedback mode to the lowest volume;

[0039] If the label includes a label of emotional fluctuations, a preset soothing sentence is added to the reply text corresponding to the voice interaction instruction to obtain the final reply content, and the reply content is fed back to the user in a soothing feedback manner, and the tone value of the soothing feedback manner is less than the tone value of the default feedback manner.

[0040] In a second aspect, an electronic device is provided, which includes at least one processor and at least one memory, 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 voice interaction method described in any one of the technical solutions of the above-mentioned voice interaction method.

[0041] In a third aspect, a computer-readable storage medium is provided, which stores a plurality of program codes, wherein the program codes are suitable for being loaded and run by a processor to execute the voice interaction method described in any one of the technical solutions of the above-mentioned voice interaction method.

[0042] In a fourth aspect, a voice interaction system is provided, the system comprising the electronic device and the interaction module in the second aspect;

[0043] The interaction module is used to receive the user's voice interaction instructions and transmit them to the electronic device, and provide feedback to the user according to the feedback method determined by the electronic device.

[0044] The above one or more technical solutions of the present application have at least one or more of the following beneficial effects:

[0045] In the case of adopting the above technical solution, the present application provides a voice interaction method, including: obtaining attribute information of each attribute of the user's voice interaction instruction, classifying the attribute information of each attribute according to a preset standard, obtaining the attribute classification result of each attribute, and obtaining the user's emotional information of the voice interaction instruction according to the attribute classification result of at least one attribute; determining the feedback method of the voice interaction instruction according to the user's emotional information. Through the above configuration method, the present application can analyze the user's emotional information through the attribute information of the attributes of the user's voice interaction instruction, and predict the user's emotional needs for dialogue and audio according to the user's emotional information, so as to determine the feedback method of the voice interaction instruction. In this way, not only can the accuracy of judging the user's emotional information be improved, but also the feedback method of the voice interaction instruction reply content can be flexibly adjusted according to the user's emotional needs, thereby avoiding the emotional distress caused to the user by the inappropriate tone or intonation of the feedback method of the voice interaction instruction, and improving the user's interactive experience. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] 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.

[0047] 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.

[0048] Figure 1 This is a flowchart of the main steps of a voice interaction method according to an embodiment of the present application;

[0049] Figure 2 It is a flowchart of the main steps of a voice interaction method according to one implementation method in the embodiments of the present application;

[0050] Figure 3 It is a schematic diagram of an implementation scenario of using a cloud server to perform voice interaction instruction analysis according to an embodiment of the present application;

[0051] Figure 4 It is a hardware environment diagram of an interaction method of a smart device according to an embodiment of the present application. DETAILED DESCRIPTION

[0052] 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.

[0053] 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.

[0054] According to one aspect of the embodiments of the present application, a voice interaction method is provided. Figure 1 , Figure 1 FIG. 1 is a flow chart of the main steps of a voice interaction method according to an embodiment of the present application. Figure 1 As shown, the voice method includes steps S101 to S104:

[0055] Step S101: Acquire attribute information of various attributes of the user's voice interaction instruction.

[0056] In this embodiment, the user's voice interaction instruction is a control instruction, which is mainly used to allow the user to control the smart device or application to perform specific operations through voice commands without manual operation or touching the screen. It is widely used in smart phones, cars, smart homes and other fields.

[0057] In some implementations, a user can speak a voice command to the intelligent voice interaction device, and the audio information spoken by the user is collected by an audio collection device on the intelligent voice interaction device, and the audio information is used as the user's voice interaction instruction.

[0058] In this embodiment, the speech speed, intonation, volume, semantics and other attributes of the voice interaction instruction can be analyzed to determine the user's emotions. The attributes of the voice interaction instruction include but are not limited to one or more of the speech speed, intonation, volume and semantics.

[0059] In one implementation, a voice interaction instruction may be received by a terminal device, and then sent to a cloud server, which extracts attribute information of multiple attributes of the voice interaction instruction based on preset analysis indicators.

[0060] Among them, the preset analysis indicators are pre-set indicators for comprehensively analyzing user emotions in voice interaction instructions, and may include multiple attributes of voice interaction instructions, such as pitch, speaking speed, volume, semantics, etc.

[0061] Step S102: classify the attribute information of each attribute according to a preset standard to obtain an attribute classification result of each attribute.

[0062] In this embodiment, the attribute is a physical attribute of the voice interaction instruction, and the attribute information may include a specific value of the attribute. For example, the attribute may be volume, and the corresponding attribute information is 20. The attribute classification result is a user emotion label obtained according to the specific value of the attribute, such as an emotion label such as a slow speaking speed, a gentle tone, or a normal inquiry in semantics.

[0063] In this embodiment, the preset standard can be pre-set according to actual use requirements, so as to classify the attribute information of each attribute to obtain the attribute classification result. For example, in the actual application scenario, a specific algorithm can be used to calculate and determine the specific value of the intonation of the voice interaction instruction, which can usually be represented by the five-degree marking method, that is, the pitch is divided into five levels, the lowest is 1 and the highest is 5, which is used to describe the change of the tone. At this time, the preset standard can be set to: if the specific value of the intonation is between the range of 1-3, it is classified as normal intonation, and when the specific value of the intonation is 4 or 5, it is classified as high intonation, and then the label of normal intonation or high intonation is used as the attribute classification result of the intonation attribute.

[0064] In this embodiment, the attributes may include voice parameters of the voice interaction instructions, such as at least one of speaking rate, intonation, volume, and semantics; the attribute classification results may include voice parameter labels, such as at least one of speaking rate labels, intonation labels, volume labels, and semantic labels.

[0065] In one implementation, the attributes of the voice instruction may include voice parameters, and the attribute classification result includes a voice parameter label, and step S102 may further include:

[0066] Step S1021: classify the attribute information of the voice parameters of the voice interaction instruction according to a preset standard, and obtain the voice parameter labels of the voice parameters.

[0067] The speech parameters include at least one of speech rate, intonation, volume and semantics, and the speech parameter tags include at least one of speech rate tags, intonation tags, volume tags and speech tags.

[0068] In one implementation, if the voice parameter is speech rate, the attribute information of the voice parameter is the actual speech rate value of the voice interaction instruction, and the voice parameter label is the speech rate label, obtaining the voice parameter label of the voice parameter may further include steps S10211-S10214:

[0069] Step S10211: Determine a preset standard speaking speed according to a preset standard.

[0070] Step S10212: Determine whether the actual speaking speed value exceeds the preset standard speaking speed.

[0071] Step S10213: If the actual speaking speed value exceeds the preset standard speaking speed, the speaking speed label is determined as an urgent speaking speed label.

[0072] Step S10214: If the actual speaking speed value does not exceed the preset standard speaking speed, the speaking speed label is determined as a slow speaking speed label.

[0073] In this embodiment, the actual speaking speed value can be the average speaking speed of the voice interaction instructions within a unit time, or it can be the maximum speaking speed of the speech interaction instructions within a unit time. It can be set according to user needs or actual application scenario requirements, and is not limited in this embodiment.

[0074] In this embodiment, the speech rate label may include a variety of labels for indicating the speech rate status of voice interaction instructions, including but not limited to urgent speech rate labels and slow speech rate labels, and other labels may also be set without affecting the normal implementation of the embodiments of the present application.

[0075] In other embodiments, the preset standard may also use other classification standards to classify the actual speech speed value. As an example, multiple classification intervals may be set. If the actual speech speed value is in the first interval, the speech speed label is determined to be a slow speech speed label. If the actual speech speed value is in the second interval, the speech speed label is determined to be a slow speech speed label. If the actual speech speed value is in the third interval, the speech speed label is determined to be an urgent speech speed label. For example, the first interval may be 0-280 words / minute, the second interval may be 280-300 words / minute, and the third interval may be greater than 300 words / minute.

[0076] In one implementation, if the voice parameter is intonation, the attribute information of the voice parameter is the actual intonation value of the voice interaction instruction, and the voice parameter label is an intonation label, step S1021 may further include steps S10221-S10224:

[0077] Step S10221: Determine the standard intonation range according to the preset standard.

[0078] Step S10222: Determine whether the actual intonation value is within the standard intonation range.

[0079] Step S10223: If the actual intonation value is within the standard intonation range, the intonation label is determined as a normal intonation label.

[0080] Step S10224: If the actual intonation value is greater than the upper limit of the standard intonation range, the intonation tag is determined as a higher intonation tag.

[0081] In this embodiment, the intonation label may include a variety of labels for indicating the intonation status of voice interaction instructions, which may include a normal intonation label, a lower intonation label, a higher intonation label, etc., and may also include other labels, which do not affect the normal implementation of the embodiments of the present application.

[0082] In this embodiment, there are four basic types of intonation: rising tone, falling tone, rising-falling tone, and falling-rising tone. In actual speech recognition scenarios, the data of actual intonation values ​​usually need to be obtained by comprehensive analysis of the frequency, pitch, amplitude, and tone of the sound signal of the voice interaction instruction, which will not be described in detail in this embodiment.

[0083] In this embodiment, the standard intonation range is a classification standard for classifying the actual intonation value determined by a preset standard. In some other embodiments, multiple intonation classification intervals can also be determined according to the preset standard to classify the actual intonation value, all of which are within the protection scope of the embodiments of the present application.

[0084] In one embodiment, the voice parameter may also be volume, and the attribute information of the volume may include the actual value of the volume of the voice interaction instruction, and the corresponding voice parameter label is a volume label. Step S1021 may further include: determining multiple volume classification intervals according to preset standards to classify the actual volume value. As an example, the volume classification intervals may be set to three intervals of less than 20 decibels, 20-80 decibels, and greater than 80 decibels. When the actual volume value is less than 20 decibels, the volume label is determined to be a low volume label; when the actual volume value is between 20-60 decibels, the volume label is determined to be a normal volume label; when the actual value of the tone is greater than 80 decibels, the volume label is determined to be a high volume label.

[0085] In other implementations, the data of the actual volume value may also be directly used as the volume label, which does not affect the normal implementation of the embodiments of the present application.

[0086] In one implementation, if the voice parameter is semantic, the attribute information of the voice parameter is the semantic analysis result of the voice interaction instruction, and the voice parameter label is a semantic label, step S1021 may further include steps S10241-S10242:

[0087] Step S10241: Determine emotion classification keywords according to preset standards.

[0088] Step S10242: Classify the semantic analysis results according to the emotion classification keywords to obtain semantic labels.

[0089] In this embodiment, the semantic analysis result is the analysis result obtained by performing semantic analysis and text sentiment analysis on the voice interaction instruction using natural language processing technology. Among them, semantic analysis can include word segmentation, part-of-speech tagging, syntactic analysis and semantic tagging of the voice interaction instruction, etc., to extract keywords and key sentences in the voice interaction instruction. Text sentiment analysis can include sentiment classification, sentiment polarity recognition and sentiment intensity analysis to obtain the intensity or degree of user emotion of the voice interaction instruction, determine whether the text expresses positive or negative emotions, and classify the text into different sentiment categories, such as positive, negative or neutral.

[0090] In this embodiment, the semantic tags may include a no emotion fluctuation tag, an emotion fluctuation tag, and an intense emotion fluctuation tag. In other embodiments, the semantic tags may also include other tags, which do not affect the normal implementation of the embodiment of the present application.

[0091] In this embodiment, the emotion classification keywords may be a plurality of pre-set keyword sets for representing different user emotions, wherein each keyword set is a set of commonly used keywords of a user emotion type.

[0092] In one implementation, emotion classification keywords can be pre-set based on an emotion dictionary. The emotion dictionary is an emotion information library based on vocabulary, which contains vocabulary and its corresponding emotion value. The emotion value is usually a numerical value used to represent the emotional tendency of the vocabulary. The emotion dictionary can be customized according to different emotion analysis tasks. For example, the emotion dictionary for a specific field (such as home, medical, etc.) saves the keywords and the corresponding emotion values ​​into a keyword set representing different user emotions according to the size of the emotion value.

[0093] In other implementations, a convolutional neural network (CNN) model can also be used to perform convolution, pooling and other operations on the preprocessed speech signal to extract emotion-related features, and then use a classifier to perform emotion classification to obtain semantic labels, which does not affect the normal implementation of the embodiments of the present application.

[0094] Step S103: obtaining user emotion information of the voice interaction instruction according to the attribute classification result of at least one attribute.

[0095] In this embodiment, the user emotion information is the result of integrating the attribute classification results of multiple attributes of the voice interaction instruction.

[0096] In one implementation, the attribute types of the voice interaction instruction may include speech speed, intonation, volume, and semantics, and the attribute classification results of each attribute, namely, speech speed label, intonation label, volume label, and semantic label, may be directly integrated to obtain user emotion information. For example, the user emotion information may be a label set including labels such as "normal speech speed", "gentle intonation", "normal volume", and "normal semantics".

[0097] Step S104: Determine the feedback method of the voice interaction instruction according to the user emotion information.

[0098] In this embodiment, the feedback method is a method of providing feedback to the user through sound, which can intuitively show the user the operation result or system status of the voice interaction instruction. This feedback method relies on speech synthesis technology, which can convert computer-generated language into natural and fluent speech output, so that users can experience a more humanized interactive experience. For example, in a smart home scenario, when a user controls lights or music through a voice assistant, the assistant can reply to the user through speech synthesis technology to confirm the operation result or provide relevant information, thereby realizing an immersive voice interaction experience.

[0099] In one implementation, the feedback method may specifically include information such as the text content, intonation, volume, and speech speed of the output voice.

[0100] In one implementation, step S104 may further include steps S1041 to S1043:

[0101] Step S1041: obtaining a preset user emotion judgment condition set, and judging whether a tag in the user emotion judgment condition set exists in the user emotion information according to a plurality of tags in the user emotion judgment condition set.

[0102] Step S1042: If the user emotion information contains at least one label in the user emotion judgment condition set, adjust the feedback method of the voice interaction instruction according to the label.

[0103] Step S1043: If any label in the user emotion judgment condition set does not exist in the user emotion information, the reply content is determined according to the user's voice interaction instruction, and the reply content is fed back to the user in a default feedback manner.

[0104] In this embodiment, the labels and default feedback methods in the user emotion judgment set can be pre-set according to user needs or actual application scenarios. For example, the user emotion judgment set may include "urgent speech speed", "high tone", "loud volume", "emotional fluctuations" and other labels, then step S1042 may further include steps S10421-step S10423:

[0105] Step S10421: If the tag includes a tag of urgent speech speed or high tone, extract the key information in the reply text corresponding to the voice interaction instruction as the final reply content, and determine the feedback mode of the reply content as a fast feedback mode.

[0106] Step S10422: If the tags include tags with a lower volume, the volume value of the feedback mode is adjusted to the lowest volume.

[0107] Step S10423: If the tag includes a tag of emotional fluctuation, a preset soothing sentence is added to the reply text corresponding to the voice interaction instruction to obtain the final reply content, and the reply content is fed back to the user in a soothing feedback manner.

[0108] Among them, the speech speed value of the quick feedback mode is greater than the speech speed value of the default feedback mode, and the tone value of the soothing feedback mode is less than the tone value of the default feedback mode.

[0109] In one embodiment, please refer to the attached Figure 2 , Figure 2 FIG. 1 is a flow chart of the main steps of a voice interaction method according to one implementation of the present application. Figure 2 As shown, the voice interaction method may further include steps S201 to S206:

[0110] Step S201: The terminal device obtains a voice interaction instruction.

[0111] In actual application scenarios, please refer to the attached Figure 3 , Figure 3 Schematic diagram of the implementation scenario of using a cloud server to analyze voice interaction instructions in an embodiment of the present application. Figure 3 As shown, voice interaction can be performed with the user through the terminal device. After the terminal device collects the voice interaction instructions, it is uploaded to the cloud server. The cloud server analyzes the user emotions in the voice interaction instructions and adjusts the feedback method to send to the terminal device.

[0112] In this embodiment, the terminal device can obtain voice interaction instructions through an audio collection device such as a microphone.

[0113] Step S202: The cloud server receives the voice interaction instruction.

[0114] In this embodiment, the cloud server has computing resources and tools related to natural language processing (NLP), so the cloud server has the function of parsing language interaction instructions, such as text analysis, sentiment analysis, and machine translation.

[0115] Step S203: The cloud server confirms and analyzes the user's emotions in the voice interaction command, and proceeds to steps S204 and S205.

[0116] In this embodiment, after receiving the interaction instruction, the cloud server may further include determining analysis indicators to comprehensively analyze the emotional points of the user's voice interaction instruction. In some embodiments, the analysis indicator may be a tag set including multiple attribute type tags. For example, the analysis indicator may be a tag set including a speech rate tag, a tone tag, a volume tag, a semantic tag, and other type tags.

[0117] Step S204: Analyze the speech speed, intonation and volume of the voice interaction instruction.

[0118] In this embodiment, step S204 may further include: extracting the speaking speed, intonation and volume of the voice interaction instruction; classifying the specific data corresponding to the speaking speed, intonation and volume according to preset standards to obtain the classification results corresponding to the speaking speed, intonation and volume.

[0119] Step S205: Analyze the semantics of the voice interaction instruction.

[0120] In this embodiment, step S205 may further include: analyzing the voice interaction instruction; judging the analysis result according to a preset standard to obtain a semantically corresponding classification result.

[0121] Step S206: Comprehensively judge and obtain the user emotion information of the voice interaction instruction.

[0122] In one embodiment, step S206 may further include: integrating the classification results of speech rate, intonation, volume and semantics to obtain user emotion information. For example, if the classification result of speech rate is a label of slow speech rate, the classification result of intonation is a label of mild intonation, the classification result of volume is a label of 20 decibels, and the classification result of semantics is a normal inquiry, then the user emotion information may include emotion labels such as "slow speech rate", "mild intonation", current volume of 20 (volume range: 10-100), and "semantics of normal inquiry".

[0123] Step S207: According to the judgment result, corresponding emotional words and audio responses are given as feedback methods of the voice interaction instructions.

[0124] In this embodiment, emotional speech refers to guiding, alleviating or enhancing the emotional response of others through specific language strategies. Among them, emotional speech can include concise speech, soothing speech and normal speech. Normal speech refers to the default reply content corresponding to the voice interaction instruction set in the terminal device; concise speech refers to simplifying the default reply content; soothing speech refers to adding soothing sentences based on the application scenario on the basis of the default reply content.

[0125] In this embodiment, the feedback method can be adjusted according to each emotional label in the judgment result. For example, if the speech rate label is an urgent speech rate label, the emotional words and audio content of the reply are automatically adjusted; if the tone label is a high tone label, the emotional words and audio content of the reply are automatically adjusted; if the volume label is a high volume label, the emotional words and audio content of the reply are automatically adjusted; if the semantic label is a semantic angry inquiry label, the emotional words and audio content of the reply are automatically adjusted.

[0126] As an example, the feedback method may also be adjusted according to the judgment basis in Table 1 and each emotion label.

[0127] Table 1 Feedback method adjustment judgment basis

[0128]

[0129] In this embodiment, soothing words can be added to the reply content, such as "I'm sorry, we will accept your suggestion with an open mind", "We understand your dissatisfaction and anxiety very well" or "I understand your mood very well" and soothing words.

[0130] Through the above steps S101 to S104, the present application can analyze the user's emotional information through the attribute information of multiple attributes of the user's voice interaction command, and can flexibly adjust the feedback method of the voice interaction command reply content according to the user's emotional needs. In this way, the user's emotional needs for dialogue and audio are predicted based on the user's emotional information, so as to determine the feedback method of the voice interaction command, which can avoid the inappropriate tone or intonation of the feedback method of the voice interaction command causing emotional distress to the user, and enhance the user's interactive experience.

[0131] According to one aspect of an embodiment of the present application, a computer-readable storage medium is provided. In a computer-readable storage medium embodiment according to the present application, the computer-readable storage medium can be configured to store a program for executing the voice interaction method of the above-mentioned method embodiment, and the program can be loaded and run by the processor to implement the above-mentioned 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 part of the embodiment of the present application. The computer-readable storage medium can be a memory device formed by various electronic devices. Optionally, the computer-readable storage medium in the embodiment of the present application is a non-temporary computer-readable storage medium.

[0132] According to one aspect of an embodiment of the present application, an electronic device is provided. In an electronic device embodiment according to the present application, the electronic device includes a processor and a memory. The memory can be configured to store a program for executing the method for adjusting the fixed thread pool of the above method embodiment. The processor can be configured to execute the program in the memory, which includes but is not limited to the program for executing the method for adjusting the fixed thread pool of the above method embodiment. 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 part of the embodiment of the present application. The electronic device can be an electronic device device formed by various electronic devices.

[0133] In the embodiment of the present application, the electronic device may be an electronic device device formed by various electronic devices. In some possible implementations, the electronic device may include multiple memories and multiple processors. The program for executing the method for adjusting the fixed thread pool of the above method embodiment can be divided into multiple subprograms, and each subprogram can be loaded and run by the processor to execute the different steps of the method for adjusting the fixed thread pool of the above method embodiment. Specifically, each subprogram can be stored in different memories, and each processor can be configured to execute the program in one or more memories to jointly implement the method for adjusting the fixed thread pool of the above method embodiment, that is, each processor executes the different steps of the method for adjusting the fixed thread pool of the above method embodiment, to jointly implement the method for adjusting the fixed thread pool of the above method embodiment.

[0134] The above-mentioned multiple processors may be processors deployed on the same device. For example, the above-mentioned electronic device may be a high-performance device composed of multiple processors, and the above-mentioned multiple processors may be processors configured on the high-performance device. In addition, the above-mentioned multiple processors may also be processors deployed on different devices. For example, the above-mentioned electronic device may be a server cluster, and the above-mentioned multiple processors may be processors on different servers in the server cluster.

[0135] According to one aspect of an embodiment of the present application, a voice interaction system is provided. In an embodiment of a voice interaction system according to the present application, the system includes an interaction module and the electronic device in the above embodiment;

[0136] The interaction module is used to receive the user's voice interaction command and transmit it to the electronic device, and provide feedback to the user according to the feedback method determined by the electronic device.

[0137] According to one aspect of the embodiment of the present application, a method for interacting with a smart home device is also provided. The method for interacting with the smart home device 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. Optionally, in this embodiment, the method for interacting with the smart home device can be applied to Figure 4 In the hardware environment composed of the terminal device 102 and the server 104 shown in FIG. Figure 4 As shown, the server 104 is connected to the terminal device 102 via a network, and can be used to provide services (such as application services, etc.) for the terminal or a client installed on the terminal. A database can be set on the server or independently of the server to provide data storage services for the server 104. Cloud computing and / or edge computing services can be configured on the server or independently of the server to provide data computing services for the server 104.

[0138] The network may include but is not limited to at least one of the following: wired network, wireless network. The 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 wireless network may include but is not limited to at least one of the following: WIFI (Wireless Fidelity), Bluetooth. The terminal device 102 may be but is not limited to a PC, a mobile phone, a tablet computer, a smart air conditioner, a smart range hood, a smart refrigerator, a smart oven, a smart stove, a smart washing machine, a smart water heater, a smart washing device, a smart dishwasher, a smart projection device, a smart TV, a smart clothes drying rack, a smart curtain, a smart audio and video, a smart socket, a smart speaker, a smart fresh air device, a smart kitchen and bathroom device, a smart bathroom device, a smart sweeping robot, a smart window cleaning robot, a smart mopping robot, a smart air purification device, a smart steamer, a smart microwave oven, a smart kitchen treasure, a smart purifier, a smart water dispenser, a smart door lock, etc.

[0139] The above is only a preferred implementation of the present application. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principles of the present application. These improvements and modifications should also be regarded as the scope of protection of the present application.

Claims

1. A voice interaction method, It is characterized in that The method comprises: Obtaining attribute information of various attributes of the user's voice interaction command; Classify the attribute information of each attribute according to the preset standard to obtain an attribute classification result of each attribute; Obtaining user emotion information of the voice interaction instruction according to an attribute classification result of at least one of the attributes; Determine a feedback method of the voice interaction instruction according to the user emotion information.

2. The voice interaction method according to claim 1, It is characterized in that The attributes of the voice instruction include voice parameters, and the attribute classification result includes a voice parameter label; The classifying the attribute information of each attribute according to the preset standard to obtain the attribute classification result of each attribute includes: The attribute information of the voice parameters of the voice interaction instruction is classified according to preset standards to obtain voice parameter labels of the voice parameters; the voice parameters include at least one of speech rate, intonation, volume and semantics, and the voice parameter labels include at least one of speech rate labels, intonation labels, volume labels and voice labels.

3. The voice interaction method according to claim 2, It is characterized in that If the voice parameter is speech rate, the attribute information of the voice parameter is the actual speech rate value of the voice interaction instruction, and the voice parameter label is the speech rate label; The acquiring of the voice parameter tag of the voice parameter comprises: Determining a preset standard speaking speed according to the preset standard; Determining whether the actual speaking speed value exceeds the preset standard speaking speed; If the actual speaking speed value exceeds the preset standard speaking speed, the speaking speed label is determined as an urgent speaking speed label; If the actual speaking speed value does not exceed the preset standard speaking speed, the speaking speed label is determined as a slow speaking speed label.

4. The voice interaction method according to claim 2, It is characterized in that If the voice parameter is intonation, the attribute information of the voice parameter is the actual intonation value of the voice interaction instruction, and the voice parameter label is the intonation label; The acquiring of the voice parameter tag of the voice parameter comprises: Determining a standard intonation range according to the preset standard; Determining whether the actual intonation value is within the standard intonation range; If the actual speaking speed value is within the standard intonation range, determining the intonation label as a normal intonation label; If the actual intonation value is greater than the upper limit of the standard intonation range, the intonation label is determined as a high-intonation label.

5. The voice interaction method according to claim 2, It is characterized in that If the voice parameter is semantic, the attribute information of the voice parameter is a semantic analysis result of the voice interaction instruction, and the voice parameter label is a semantic label; The acquiring of the voice parameter tag of the voice parameter comprises: Determine emotion classification keywords according to the preset criteria; Classifying the semantic analysis result according to the emotion classification keyword to obtain the semantic label, wherein the semantic analysis result is an analysis result obtained by performing semantic analysis and text sentiment analysis on the voice interaction instruction using natural language processing technology; The semantic labels include a no emotion fluctuation label, an emotion fluctuation label and a severe emotion fluctuation label.

6. The voice interaction method according to claim 2, It is characterized in that The step of determining the feedback mode of the voice interaction instruction according to the user emotion information includes: Obtaining a preset user emotion judgment condition set, and judging whether the user emotion information contains a label in the user emotion judgment condition set according to the label in the user emotion judgment condition set; If at least one label in the user emotion judgment condition set exists in the user emotion information, adjusting the feedback mode of the voice interaction instruction according to the label; If any label in the user emotion judgment condition set does not exist in the user emotion information, the reply content is determined according to the user's voice interaction instruction, and the reply content is fed back to the user in a default feedback manner.

7. The voice interaction method according to claim 6, It is characterized in that The user emotion judgment condition set includes labels such as urgent speaking speed, high tone, loud volume and emotional fluctuation; The step of adjusting the feedback mode of the voice interaction instruction according to the label includes: If the tag includes a tag of urgent speech speed or high tone, extract key information in the reply text corresponding to the voice interaction instruction as the final reply content, determine that the feedback mode of the reply content is a fast feedback mode, and the speech speed value of the fast feedback mode is greater than the speech speed value of the default feedback mode; If the labels include labels with a relatively low volume, adjusting the volume value of the feedback mode to the lowest volume; If the label includes a label of emotional fluctuations, a preset soothing sentence is added to the reply text corresponding to the voice interaction instruction to obtain the final reply content, and the reply content is fed back to the user in a soothing feedback manner, and the tone value of the soothing feedback manner is less than the tone value of the default feedback manner.

8. A computer-readable storage medium, It is characterized in that The computer-readable storage medium includes a stored program, wherein the program executes the voice interaction method according to any one of claims 1 to 7 when running.

9. An electronic device comprising a memory and a processor, It is characterized in that A computer program is stored in the memory, and the processor is configured to execute the voice interaction method according to any one of claims 1 to 7 through the computer program.

10. A voice interaction system, It is characterized in that The system comprises the electronic device and the interaction module of claim 9; The interaction module is used to receive the user's voice interaction instructions and transmit them to the electronic device, and provide feedback to the user according to the feedback method determined by the electronic device.

Images