Semantic analysis-based device control method and apparatus, controlled device, and server

Abstract

The present invention relates to the technical fields such as semantic analysis, intelligent control, artificial intelligence, and human-machine interaction, and provides a semantic analysis-based device control method and apparatus, a controlled device, and a server. The method comprises: performing semantic analysis on semantic text to obtain affective feature semantics affecting an affective state of a device user of a controlled device; and automatically controlling a working mode of the controlled device on the basis of the affective feature semantics, so that the controlled device acts on the device user in different working modes on the basis of different affective feature semantics, changing the use experience of the device user when using the controlled device, thereby establishing association among the semantics, the controlled device, and the use experience of the controlled device, improving the degree of intelligence of device control, and improving the user experience.

Description

Semantic analysis-based device control methods, apparatus, controlled devices, and servers Technical Field

[0001] This invention relates to the fields of semantic analysis, intelligent control, artificial intelligence, and human-computer interaction, and in particular to a device control method, apparatus, controlled device, and server based on semantic analysis. Background Technology

[0002] With the rapid development of information technology, many service-oriented smart devices can now interact with users' bodies or living environments through various control methods, providing a variety of personalized user experiences. Existing smart devices typically use traditional methods such as button control, touchscreen control, remote control, and mobile app control to adjust the device's working status and functions. For example, the vibration frequency of a massager can be adjusted via buttons, touchscreen, or remote control, and users can also remotely control its functions through a mobile app. Although these traditional control methods provide users with basic device operation functions, they still have a high degree of user dependence, requiring users to manually interact multiple times to achieve the desired device behavior. Moreover, existing device control methods cannot adjust in real time according to the user's emotions or psychological state, failing to achieve the effect of emotional interaction, resulting in a relatively monotonous and mechanical user experience. Therefore, existing smart devices have significant room for improvement in terms of interactivity, personalization, and emotional responsiveness. Traditional service-oriented smart devices, such as massagers, home appliances, and smart home devices, while possessing certain intelligent control functions, still lack sufficient ability to perceive and respond to the user's emotional state. Most devices still rely on direct user operation to trigger preset functions and cannot proactively adjust according to semantics to provide users with an immersive experience.

[0003] In summary, existing technologies suffer from technical problems such as reliance on user operation, low level of intelligence, inability to proactively adjust based on semantics, and inability to provide users with an immersive experience. Summary of the Invention

[0004] To address the shortcomings of the existing technologies, this invention provides a device control method, apparatus, controlled device, and server based on semantic analysis, thereby improving the intelligence level of intelligent device control and enhancing the user experience of intelligent devices.

[0005] In a first aspect, the present invention provides a device control method based on semantic analysis, comprising:

[0006] Semantic analysis is performed on semantic text to obtain the semantic features of sensory characteristics that affect the sensory state of device users in the controlled device.

[0007] The controlled device's operating mode is automatically controlled based on the sensory feature semantics, so that the controlled device operates in different modes according to the different sensory feature semantics, thereby changing the user experience when using the controlled device.

[0008] Secondly, the present invention provides a device control apparatus based on semantic analysis, comprising:

[0009] The semantic analysis module is used to perform semantic analysis on semantic text in order to obtain the semantic features of the sensory state of the device user that affect the controlled device.

[0010] The working mode control module is used to automatically control the working mode of the controlled device according to the sensory feature semantics, so that the controlled device acts on the device user in different working modes according to the different sensory feature semantics, thereby changing the user experience when the device user uses the controlled device.

[0011] Thirdly, the present invention provides a controlled device, which is controlled using the above-described semantic analysis-based device control method.

[0012] Fourthly, the present invention provides a server that runs the above-described semantic analysis-based device control method to control a controlled device.

[0013] Compared with the prior art, the beneficial effects of this invention are as follows:

[0014] This invention provides a device control method, apparatus, controlled device, and server based on semantic analysis. By performing semantic analysis on semantic text, it obtains the sensory feature semantics that affect the sensory state of the device user. Based on the sensory feature semantics, it automatically controls the working mode of the controlled device, so that the controlled device acts on the device user in different working modes according to the different sensory feature semantics, thereby changing the user experience when using the controlled device. This establishes the correlation between semantics, the controlled device, and the user experience of the controlled device, improving the intelligence of device control and enhancing the user experience. Attached Figure Description

[0015] The accompanying drawings, which are included to provide a further understanding of the invention and form part of this invention, illustrate exemplary embodiments of the invention and are used to explain the invention, but do not constitute an undue limitation of the invention. Some specific embodiments of the invention will be described in detail below with reference to the accompanying drawings in an exemplary and non-limiting manner. The same reference numerals in the drawings designate the same or similar parts or components. It should be understood by those skilled in the art that these drawings are not necessarily drawn to scale. In the drawings:

[0016] Figure 1 is a flowchart illustrating a device control method based on semantic analysis according to an embodiment of the present invention.

[0017] Figure 2 is a schematic diagram of the architecture of a device control method based on semantic analysis according to an embodiment of the present invention;

[0018] Figure 3 is a schematic diagram of the server architecture according to an embodiment of the present invention;

[0019] Figure 4 is a schematic diagram of an architecture for communication between the controlled device, the user terminal, and the server according to an embodiment of the present invention. Detailed Implementation

[0020] To enable those skilled in the art to better understand the present invention, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments. Obviously, the described embodiments are merely some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of the present invention.

[0021] Example 1

[0022] Referring to Figures 1-4, this embodiment provides a device control method based on semantic analysis, including steps S101 and S102. Steps S101 and S102 can be run on a server or a user terminal. By performing semantic analysis on semantic text, sensory feature semantics affecting the user's perception of the controlled device are obtained. Based on these sensory feature semantics, the operating mode of the controlled device is automatically controlled, so that the controlled device operates in different modes according to the different sensory feature semantics, changing the user's experience when using the controlled device. This establishes a correlation between semantics, the controlled device, and the user experience, improving the intelligence of device control and enhancing the user experience.

[0023] It should be noted that the server includes memory, processor, and network interface interconnected via a system bus. Those skilled in the art will understand that the server described here is a device capable of automatically performing numerical calculations and / or information processing according to pre-set or stored instructions. Its hardware includes, but is not limited to, microprocessors, application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), digital signal processors (DSPs), embedded devices, etc. The server can interact with users via keyboard, mouse, remote control, touchpad, or voice control devices. The memory includes at least one type of readable storage medium, including flash memory, hard disk, multimedia card, card-type memory (e.g., SD or DX memory), random access memory (RAM), static random access memory (SRAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), programmable read-only memory (PROM), magnetic memory, magnetic disk, optical disk, etc. In some embodiments, the memory can be an internal storage unit of the server, such as the server's hard disk or RAM. In other embodiments, the memory can also be an external storage device of the server, such as a plug-in hard drive, Smart Media Card (SMC), Secure Digital (SD) card, Flash Card, etc., provided on the server. Of course, the memory can also include both the server's internal storage units and its external storage devices.

[0024] Referring to Figures 1-4, the device control method based on semantic analysis provided in this embodiment may include the following steps:

[0025] Step S101: Perform semantic analysis on the semantic text to obtain the semantic features of the sensory state of the device user that affect the controlled device.

[0026] Step S102: Automatically control the working mode of the controlled device according to the sensory feature semantics, so that the controlled device acts on the device user in different working modes according to the different sensory feature semantics, thereby changing the user experience when the device user uses the controlled device.

[0027] It should be noted that traditional service-oriented smart devices (such as massagers, home appliances, and smart home devices) mainly rely on multiple manual operations by users to achieve the expected functions, making it difficult to make real-time, proactive functional adjustments based on the user's emotions or psychological state. In step S101, semantic analysis is performed on the semantic text to obtain the sensory feature semantics that affect the user's sensory state in the controlled device, thereby enabling proactive mining of emotions, needs, or psychological states within the semantic text. Step S101 identifies and analyzes the semantic information in the semantic text, significantly reducing the frequency of user-initiated operations and achieving automated capture of user emotions and psychological states. In addition, traditional control methods lack emotional interaction, resulting in a monotonous and mechanical user experience. The lack of real-time perception of user emotions makes it difficult to create an immersive experience. In step S102, the operating mode of the controlled device is automatically controlled according to the sensory feature semantics, that is, the sensory feature semantics obtained from semantic analysis are linked with the device's operating mode to achieve automated and personalized adjustments to the device. By mapping the results of emotion semantic recognition to different operating modes (reciprocating motion, rotation, vibration frequency, etc.) of a controlled device (such as a massager), the user experience can be adjusted and enhanced in real time, making the device's interaction more flexible and emotionally responsive, and significantly improving the intelligence and proactive adjustment capabilities of device control. The controlled device includes a massager; the operating modes of the massager include reciprocating motion modes and / or rotational motion modes based on the sensory semantic features, and / or different vibration frequency modes and / or different vibration intensity modes.

[0028] In some preferred embodiments, the semantic text is either text or speech text; the speech text is either dialogue-style audio text or non-dialogue-style audio text; the dialogue-style audio text is either the audio text of the current dialogue or a recorded dialogue. It should be noted that user emotions or needs often exist in multiple forms, such as text or speech text. Users may express emotions or needs through text (e.g., chat logs, novel document content) or audio (e.g., dialogues, recordings). In this embodiment, semantic text can include multiple sources, such as text, dialogue-style audio text, and non-dialogue-style audio text, making the device control method applicable to different input channels and thus meeting more diverse usage scenarios (e.g., real-time interaction or recording file analysis). Speech text can come from dialogue-style audio text (the current dialogue or a recorded dialogue) or non-dialogue-style audio text (e.g., novel recordings). Different speech scenarios contain different contexts and emotional characteristics. In this embodiment, dialogue-style audio text and non-dialogue-style audio text are categorized to ensure comprehensiveness in semantic data collection and analysis. Furthermore, the current dialogue-style audio text indicates that real-time dialogue between the user and the interacting user can be analyzed, allowing for timely capture of current semantic changes. The presence of either dialogue-based or non-dialogue-based audio text indicates that pre-recorded audio can be analyzed offline, satisfying users' semantic analysis needs in non-real-time scenarios (such as playback).

[0029] In some preferred embodiments, when the semantic text is text, the semantic analysis of the semantic text includes: obtaining the text currently selected by the device user or the text automatically recommended by the recommendation algorithm and performing semantic analysis to obtain the sensory feature semantics affecting the sensory state of the device user. It should be noted that in this embodiment, text can be obtained based on user selection or an automatic recommendation mechanism for semantic analysis to meet diverse needs. Specifically, the user can manually select text on the device or the recommendation algorithm can automatically recommend text. When the user already has specific text to analyze, they can select it manually; when the user does not have a specific target text, the algorithm can be used to filter and recommend text that may be related to the user's mood or preferences, thereby better meeting the user's device control needs in text-based scenarios.

[0030] In some preferred embodiments, when the semantic text is non-dialogue audio text, the semantic analysis of the semantic text includes: obtaining the non-dialogue audio text currently selected by the device user or the non-dialogue audio text automatically recommended by the recommendation algorithm, and performing semantic analysis to obtain the sensory feature semantics that affect the sensory state of the device user. It should be noted that in actual use, the audio a user may listen to is not necessarily in dialogue form (such as natural sounds, broadcasts, readings, etc.). By processing non-dialogue audio text separately, semantic analysis can be performed on a wider range of audio sources. This analysis of non-dialogue audio can capture potential emotional elements in the audio, thereby providing a more comprehensive understanding and influence on the device user. Non-dialogue audio can be actively selected by the user or automatically recommended by a recommendation algorithm. When a user is unsure what kind of audio they need, suitable non-dialogue audio can be automatically recommended based on their emotions, preferences, or historical records for semantic analysis, thereby automatically adjusting the operating mode of the controlled device. This allows the device to respond more quickly and accurately to the user's needs, while also reducing the burden of frequent selection and operation for the user and improving the interactive experience.

[0031] In some preferred embodiments, when the semantic text is audio text in the form of a dialogue, the audio text in the form of a dialogue consists of the dialogue content of at least two dialogue subjects in the dialogue scenario; the dialogue content includes the dialogue content of the device user using the controlled device and the dialogue content of the interactive user interacting with the device user. It should be noted that in actual use environments, there are more than one person speaking in an audio dialogue. If only the device user's own words are focused on, it may not be possible to fully and accurately capture their sensory state and needs. By collecting and analyzing the dialogue content of at least two dialogue subjects (including the device user and the interactive user), the speech, tone, and context of both parties can be referenced simultaneously, thereby gaining a richer understanding of the user's psychological and emotional changes and the impact of the interaction process. At least two dialogue subjects in the dialogue scenario can not only distinguish the speeches of different speakers, but also identify the content and emotions expressed by different speakers, thereby more accurately identifying the source of the controlled device user's sensory changes (such as their own emotional fluctuations or being triggered by the words of the interactive user), thus improving the targeting of semantic analysis and device control.

[0032] In some preferred embodiments, semantic analysis of the semantic text includes: acquiring the dialogue content of the dialogue subjects, and performing semantic analysis on the dialogue content to obtain the semantic features of sensory characteristics that affect the sensory state of the device user. It should be noted that in this embodiment, the dialogue content of the dialogue subjects is listed independently, allowing for the identification and analysis of the language content and emotional information of each dialogue subject, ensuring a complete understanding of the context, and thus more accurately inferring the emotional state of the device user and the source of influence. In this embodiment, instead of generically identifying the emotions of all speakers, the focus is on identifying which semantic elements affect the device user. Through such targeted semantic analysis, sensory changes closely related to user experience can be quickly identified in the dialogue scenario, thereby guiding the device to make corresponding automated adjustments or feedback, enhancing the user's personalized and immersive experience.

[0033] In some preferred embodiments, when the audio text in the form of a dialogue is a recorded text in the form of a dialogue, the dialogue content of the dialogue subjects is obtained, and semantic analysis is performed on the dialogue content. This includes: obtaining the recorded text in the form of a dialogue currently selected by the device user or the recorded text in the form of a dialogue automatically recommended by a recommendation algorithm; performing semantic analysis on the obtained recorded text in the form of a dialogue to obtain the semantic features of the sensory state of the device user that affect the controlled device. It should be noted that, compared with real-time dialogue, the recorded text is offline audio and can be played back and semantically analyzed afterward. In this embodiment, the recorded text in the form of a dialogue can be compatible with more types of dialogue scenarios. For example, it supports the analysis of both real-time call content and recordings selected by the user afterward or automatically recommended by a recommendation algorithm. This compatibility can significantly expand the application scope, meet the needs of diverse scenarios, and enable smart devices to achieve emotional responses in more situations. If the user already has specific recorded content that they want to analyze, they can manually select it; if the user does not have specific recorded content, the recommendation algorithm can automatically recommend it based on interests, emotions, historical usage habits, etc.

[0034] In some preferred embodiments, when the audio text in the dialogue format is the audio text of the current dialogue format, the dialogue content of the dialogue subjects is obtained, and semantic analysis is performed on the dialogue content. This includes: obtaining the current dialogue content of the interactive user, and performing semantic analysis based on the current dialogue content of the interactive user to obtain the semantic features of the sensory state of the device user affecting the controlled device. It should be noted that, unlike the post-event analysis of recorded text, the audio text in the current dialogue format emphasizes semantic analysis while the dialogue is in progress. By capturing and analyzing the current dialogue content of the interactive user in real time, the speech, tone, and emotional information of the interactive user can be grasped at the moment the dialogue occurs, and its impact on the sensory state of the device user can be perceived and judged in a timely manner. This real-time analysis can automatically adjust the device's working mode in the first instance, significantly improving the real-time performance and immersive interactive experience of the smart device. Performing semantic analysis based on the current dialogue content of the interactive user can focus on identifying the key information (such as words, emotions, and tone) conveyed by the interactive user, and inferring the sensory impact on the device user accordingly. This process can quickly locate the triggers of the user's feelings, thereby ensuring more accurate and timely identification of the sensory state of the device user, and laying the foundation for the smart device to adjust its mode and optimize its interaction.

[0035] In further embodiments, when the audio text of the dialogue is the audio text of the current dialogue, the dialogue content of the dialogue subjects is obtained, and semantic analysis is performed on the dialogue content. This includes: obtaining the current dialogue content of the device user and the current dialogue content of the interactive user; performing semantic analysis based on the current dialogue content of the device user and the current dialogue content of the interactive user to obtain the sensory feature semantics affecting the sensory state of the device user. It should be noted that, unlike analyzing only the dialogue content of the interactive user or only the device user, this embodiment simultaneously obtains the current dialogue content of both the device user and the interactive user. This allows for a more comprehensive and richer understanding of the current dialogue context, such as: what did the device user say? What was their tone? What did the interactive user say? What were their emotions? How did the language of both parties influence each other? Simultaneously collecting and combining information from both parties during the dialogue allows for a more accurate understanding of the context and underlying emotional logic of the dialogue, enabling more appropriate semantic judgments.

[0036] In further embodiments, semantic analysis of the dialogue content includes: combining the dialogue content of the device user and the dialogue content of the interactive user into contextual content with a specific context; and analyzing the contextual content to obtain the semantic features of the sensory characteristics that affect the sensory state of the device user. It should be noted that the dialogue between the device user and the interactive user is often a continuous and mutually influential process. The device user's words will elicit a response from the interactive user, and the interactive user's words will in turn affect the device user. In this embodiment, merging the dialogue content of both parties into contextual content with a specific context allows for understanding the complete dialogue flow, thereby more accurately identifying various factors affecting the sensory state of the device user. When the dialogue content is integrated into contextual content, not only can the semantics of single sentences or paragraphs be identified, but the contextual relationships of the dialogue can also be tracked (e.g., whether a sentence responds to a previous topic, or whether there is a progression or opposition in emotion). This contextual relationship can distinguish between short-term emotional fluctuations and persistent emotional attitudes, and determine changes in the device user's feelings (e.g., changes in emotional direction or psychological state). Ultimately, based on more accurate and richer sensory feature semantics, the controlled device can be automatically controlled, thereby significantly enhancing the effect of emotional interaction and personalized response.

[0037] In some further embodiments, analyzing the contextual content includes: analyzing the emotional semantics within the contextual content to obtain the sensory feature semantics that influence the emotional state of the device user in the dialogue scenario. It should be noted that this embodiment emphasizes the identification and extraction of emotional semantics, focusing on the direct or implicit emotional components in the dialogue scenario. This allows for precise identification of which emotions, attitudes, or psychological factors most influence the device user, providing more valuable emotional basis for subsequent automatic adjustments of the device. Simply obtaining general semantic information (such as topics, intentions, etc.) is insufficient for achieving deep emotional interaction. In this embodiment, by deeply mining the emotional semantics in the dialogue and focusing on analyzing these semantics, sensory feature semantics that influence the emotional state of the device user are extracted. These results are then mapped to the automatic control of different operating modes of the device (such as a massager). This chain from emotional semantics to device control can significantly enhance the user's immersive experience and satisfaction, making the response of smart devices more human-centered and emotional.

[0038] In some further embodiments, analyzing the emotional semantics in the contextual content includes: parsing the semantic text embodying the contextual content to obtain emotional keywords, tone keywords, and contextual keywords in the semantic text; and analyzing the emotional semantics in the contextual content based on the emotional keywords, tone keywords, and contextual keywords to obtain the sensory feature semantics affecting the emotional state of the device user in the dialogue scenario. It should be noted that in a dialogue scenario, a user's emotions are often reflected in multiple dimensions: whether the words themselves carry a positive or negative tendency (emotional keywords), whether the tone contains a strong or subtle attitude (tone keywords), and how the dialogue background or context shapes the language (contextual keywords). In this embodiment, clues are extracted from the above multiple dimensions simultaneously. This allows for a more refined layered understanding of textual information, thereby making sentiment analysis more in-depth and accurate technically, moving beyond simple sentiment label recognition. When analyzing text, if only a single keyword (such as "anger" or "happiness") is used to judge the user's emotions, it is often impossible to accurately distinguish the intensity of the tone or the true meaning of the context. By combining emotional keywords (indicating emotional tendency), tone keywords (indicating expression and attitude), and contextual keywords (indicating scene and contextual relationship), we can better grasp the overall meaning and emotional trajectory of the dialogue. This multi-faceted fusion analysis approach can quickly and accurately identify semantic elements that have a key impact on changes in user perception, providing a more precise basis for subsequent automatic device control (such as vibration mode, lighting adjustment, etc.), thereby enhancing the user's immersive experience and emotional identification.

[0039] In further embodiments, the emotional semantics in the contextual content are analyzed using an AI sentiment analysis model. This model analyzes the emotional semantics in the contextual content based on the emotional keywords, tone keywords, and contextual keywords to obtain the sensory feature semantics affecting the emotional state of the device user in the dialogue scenario. The AI ​​sentiment analysis model can include any one of the following: convolutional neural networks, recurrent neural networks, long short-term memory networks, deep neural networks, and Transformer networks. It should be noted that, compared to manually defined rule systems, AI models possess self-learning and generalization capabilities, continuously optimizing recognition accuracy through accumulated dialogue data and feedback. Through comprehensive learning of emotional keywords, tone keywords, and contextual keywords, the AI ​​model can analyze user emotional fluctuations in real-time or near real-time and produce accurate sensory feature semantics affecting the emotional state of the device user. This allows intelligent devices to quickly and flexibly adjust their operating modes according to the user's emotional state, thereby significantly improving the user's immersive experience and intelligent interaction effects.

[0040] In some preferred embodiments, when the device user selects the text, the dialogue-style audio text, or the non-dialogue-style audio text, the selection is made through a user interface or voice command.

[0041] Example 2

[0042] Referring to Figures 2 and 4, this embodiment provides a device control apparatus based on semantic analysis, including:

[0043] The semantic analysis module is used to perform semantic analysis on semantic text in order to obtain the semantic features of the sensory state of the device user that affect the controlled device.

[0044] The working mode control module is used to automatically control the working mode of the controlled device according to the sensory feature semantics, so that the controlled device acts on the device user in different working modes according to the different sensory feature semantics, thereby changing the user experience when the device user uses the controlled device.

[0045] In this embodiment, semantic analysis is performed on semantic text to obtain the sensory feature semantics that affect the sensory state of the device user. The working mode of the controlled device is automatically controlled according to the sensory feature semantics, so that the controlled device acts on the device user in different working modes according to the different sensory feature semantics, thereby changing the user experience when using the controlled device. This establishes the correlation between semantics, the controlled device, and the user experience of the controlled device, improves the intelligence of device control, and enhances the user experience.

[0046] Example 3

[0047] Referring to Figures 1, 3, and 4, this embodiment provides a controlled device that uses any of the semantic analysis-based device control methods described in the above embodiments. The controlled device can connect and communicate with a user terminal and a server. By performing semantic analysis on semantic text, it obtains the sensory feature semantics that affect the user's sensory state. Based on these sensory feature semantics, it automatically controls the operating mode of the controlled device, allowing it to operate in different modes according to the different sensory feature semantics. This alters the user's experience when using the controlled device, thereby establishing a correlation between semantics, the controlled device, and the user experience, enhancing the intelligence of device control and improving the user experience.

[0048] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some or all of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention.

Claims

1. A device control method based on semantic analysis, characterized in that, include: Semantic analysis is performed on semantic text to obtain the semantic features of sensory characteristics that affect the sensory state of device users in the controlled device. The controlled device's operating mode is automatically controlled based on the sensory feature semantics, so that the controlled device operates in different modes according to the different sensory feature semantics, thereby changing the user experience when using the controlled device.

2. The device control method based on semantic analysis as described in claim 1, characterized in that, The semantic text can be text or audio text; the audio text can be audio text in dialogue form or audio text in non-dialogue form; the audio text in dialogue form can be audio text in the current dialogue form or audio recording text in dialogue form.

3. The device control method based on semantic analysis as described in claim 2, characterized in that, When the semantic text is text, the semantic analysis of the semantic text includes: obtaining the text selected by the current device user or the text automatically recommended by the recommendation algorithm and performing semantic analysis to obtain the sensory feature semantics that affect the sensory state of the device user of the controlled device.

4. The device control method based on semantic analysis as described in claim 2, characterized in that, When the semantic text is non-dialogue audio text, the semantic analysis of the semantic text includes: obtaining the non-dialogue audio text currently selected by the device user or the non-dialogue audio text automatically recommended by the recommendation algorithm and performing semantic analysis to obtain the sensory feature semantics that affect the sensory state of the device user of the controlled device.

5. The device control method based on semantic analysis as described in claim 2, characterized in that, When the semantic text is audio text in the form of a dialogue, the audio text in the form of a dialogue consists of the dialogue content of at least two dialogue subjects in the dialogue scene; the dialogue content includes the dialogue content of the device user using the controlled device and the dialogue content of the interactive user interacting with the device user.

6. The device control method based on semantic analysis as described in claim 5, characterized in that, The semantic analysis of the semantic text includes: obtaining the dialogue content of the dialogue subjects, performing semantic analysis on the dialogue content, and obtaining the semantic features of the sensory state of the device user that affect the controlled device.

7. The device control method based on semantic analysis as described in claim 6, characterized in that, When the audio text in the form of a dialogue is a recorded text in the form of a dialogue, the dialogue content of the dialogue subjects is obtained, and semantic analysis is performed on the dialogue content, including: obtaining the recorded text of the dialogue format currently selected by the device user or the recorded text of the dialogue format automatically recommended by the recommendation algorithm, and performing semantic analysis on the obtained recorded text of the dialogue format to obtain the sensory feature semantics that affect the sensory state of the device user of the controlled device.

8. The device control method based on semantic analysis as described in claim 6, characterized in that, When the audio text in the form of the dialogue is the audio text in the current dialogue form, the dialogue content of the dialogue subject is obtained, and semantic analysis is performed on the dialogue content, including: obtaining the current dialogue content of the device user and the current dialogue content of the interactive user, and performing semantic analysis based on the current dialogue content of the device user and the current dialogue content of the interactive user to obtain the sensory feature semantics that affect the sensory state of the device user of the controlled device.

9. The device control method based on semantic analysis as described in claim 6, characterized in that, Semantic analysis of the dialogue content includes: combining the dialogue content of the device user and the dialogue content of the interactive user into contextual content; and analyzing the contextual content to obtain the semantic features of the sensory characteristics that affect the sensory state of the device user in the controlled device.

10. The device control method based on semantic analysis as described in claim 9, characterized in that, Analyzing the contextual content includes: analyzing the emotional semantics in the contextual content to obtain the sensory feature semantics that affect the emotional state of the device user in the dialogue scenario.

11. The device control method based on semantic analysis as described in claim 10, characterized in that, Analyzing the emotional semantics in the contextual content includes: parsing the semantic text that embodies the contextual content to obtain the emotional keywords, tone keywords, and contextual keywords in the semantic text; and analyzing the emotional semantics in the contextual content based on the emotional keywords, tone keywords, and contextual keywords to obtain the sensory feature semantics that affect the emotional state of the device user in the dialogue scenario.

12. A device control apparatus based on semantic analysis, characterized in that, include: The semantic analysis module is used to perform semantic analysis on semantic text in order to obtain the semantic features of the sensory state of the device user that affect the controlled device. The working mode control module is used to automatically control the working mode of the controlled device according to the sensory feature semantics, so that the controlled device acts on the device user in different working modes according to the different sensory feature semantics, thereby changing the user experience when the device user uses the controlled device.

13. A controlled device, characterized in that, The controlled device is controlled using the semantic analysis-based device control method as described in any one of claims 1-11.

14. A server, characterized in that, include: The device control method based on semantic analysis as described in any one of claims 1-11 is executed to control the controlled device.

Images