Method for generating an avatar of a voice assistant, control device, vehicle and program product

Abstract

The application provides a voice assistant image generation method, control device, vehicle and program product, relates to the technical field of intelligent cockpit, and aims to solve the problems that the voice assistant image of the vehicle is single and lacks personalization in the prior art, resulting in lack of emotional connection and sense of belonging in the human-vehicle interaction experience. The method comprises the following steps: determining a corresponding constellation type of a user; determining a personality characteristic label based on the constellation type, wherein the personality characteristic label comprises a visual characteristic label, an auditory characteristic label and a dialogue characteristic label; generating a visual instruction, an auditory instruction and a dialogue instruction according to the personality characteristic label; generating a visual image of the voice assistant according to the visual instruction, generating a tone of the voice assistant according to the auditory instruction, and generating an interactive dialogue library of the voice assistant according to the dialogue instruction.

Description

Technical Field

[0001] This application relates to the field of smart cockpit technology, and in particular to a method for generating the image of a voice assistant, a control device, a vehicle, and a program product. Background Technology

[0002] With the rapid development of smart cockpit technology, voice assistants have become the core interactive interface of in-vehicle infotainment systems. Users can use voice assistants to perform various functions such as navigation, music playback, phone calls, and vehicle control. However, existing in-vehicle voice assistants typically use preset, generic virtual avatars or offer only a limited number of selectable skins. Regardless of a user's personal characteristics, preferences, or background, they are often faced with the same standardized voice assistant avatar. This one-size-fits-all interactive experience lacks respect for and responsiveness to individual user differences, making it difficult to establish deep emotional connections and reducing user stickiness and a sense of exclusivity. Summary of the Invention

[0003] This application provides a method for generating the image of a voice assistant, a control device, a vehicle, and a program product, aiming to solve the problems in the prior art where the image of an in-vehicle voice assistant is monotonous and lacks personalization, resulting in a lack of emotional connection and a sense of belonging in the human-vehicle interaction experience.

[0004] In a first aspect, this application provides a method for generating the image of a voice assistant, comprising: determining the user's corresponding zodiac sign type; determining personality feature tags based on the zodiac sign type, the personality feature tags including visual feature tags, auditory feature tags, and verbal feature tags; generating visual instructions, auditory instructions, and verbal instructions based on the personality feature tags; generating the visual image of the voice assistant based on the visual instructions, generating the voice tone of the voice assistant based on the auditory instructions, and generating an interactive dialogue library for the voice assistant based on the verbal instructions.

[0005] The voice assistant image generation method provided in this application overcomes the problems of traditional voice assistants having a single image and lacking personalization by determining the user's zodiac sign type and generating personality trait tags based on the zodiac sign type, and then generating multimodal commands. This method can customize the visual image, voice tone, and interactive scripts according to the user's zodiac sign characteristics, enabling the voice assistant to present personalized features that match the user's personality. This enhances the interactive experience and emotional connection between the user and the voice assistant, increasing the user's acceptance and willingness to use the voice assistant.

[0006] In some possible implementations, determining the user's corresponding zodiac sign type includes: in response to the user's authorization instruction, obtaining the user's date of birth information from the vehicle user account system; and determining the user's corresponding zodiac sign type based on the date of birth information according to a preset calculation rule.

[0007] According to the above embodiments, by responding to the user's authorization command, the birth date information is obtained from the vehicle user account system, and the constellation type is determined according to the preset calculation rules. This ensures the authenticity and reliability of the user's constellation information, provides an accurate data foundation for the subsequent personalized generation of the constellation-based voice assistant, avoids personalization deviations caused by incorrect constellation information, and improves the accuracy of matching the voice assistant with the user.

[0008] In some possible implementations, determining the user's corresponding zodiac sign type includes: obtaining the user's input on the zodiac sign type, and determining the user's corresponding zodiac sign type based on the user's input.

[0009] According to the above embodiments, the zodiac sign type is determined by directly obtaining the user's input, avoiding the problem of some users being unable to provide their birth date information. This provides users with a convenient way to directly select their zodiac sign, simplifies the operation process, and increases user participation. It is particularly suitable for users who already know about zodiac signs or prefer preset zodiac sign types, enhancing user autonomy and satisfaction in the voice assistant's generation process.

[0010] In some possible implementations, the voice assistant image generation method further includes: encapsulating the visual image, voice tone, and interactive script library into a multimodal personality package; performing a personality tone matching assessment on the multimodal personality package to ensure that the voice assistant's image, voice tone, and script match; and adjusting the visual image, voice tone, and interactive script library of the multimodal personality package if a conflict is detected.

[0011] According to the above embodiments, by encapsulating visual image, voice tone and interactive script library into a multimodal personality package and conducting personality tone matching assessment, the consistency and coordination of the voice assistant in visual, auditory and linguistic expression are ensured, avoiding the sense of image fragmentation caused by element conflict, improving the professionalism, credibility and naturalness of the voice assistant's overall image, and allowing users to experience a more consistent and coherent interactive experience.

[0012] In some possible implementations, the voice assistant's image generation method further includes: acquiring driving status information, including time context, scene context, and user state context; and adjusting the visual image, voice tone, and interactive dialogue library based on the driving status information according to preset personality elastic mapping rules.

[0013] According to the above embodiments, by acquiring driving status information and adjusting the multimodal performance of the voice assistant according to preset personality elastic mapping rules, the voice assistant can dynamically adjust its image, tone and speech based on factors such as driving time, scenario and user status. This achieves intelligent adaptation of the voice assistant to the driving environment, improves the contextual relevance and practicality of the interaction, and enhances the auxiliary effect of the voice assistant in complex driving environments.

[0014] In some possible implementations, the time context includes time and date type; the scenario context includes navigation destination type, road condition and driving duration; and the user state context includes user mood and user preferences learned through historical interactions.

[0015] According to the above embodiments, by clearly defining the specific content of time context, scene context and user state context, a clear framework is provided for the acquisition and parsing of driving status information, enabling the voice assistant to fully and accurately understand the current driving environment, providing a refined decision-making basis for subsequent personality flexibility adjustments, and improving the voice assistant's perception ability and response accuracy to the driving environment.

[0016] In some possible implementations, visual feature labels include symbols, representative graphics, lucky colors, and related materials; and / or, auditory feature labels include emotional tone, speech rate and rhythm, timbre and intonation patterns; and / or, dialogue feature labels include the degree of initiative in the conversation, preference for commonly used interjections, humor index, empathy index, and recommended dialogue for a preset driving scenario.

[0017] According to the above embodiments, by defining visual elements such as symbols, representative graphics, and lucky colors, auditory elements such as emotional tone and speech rate, and conversational elements such as the degree of initiative in the conversation and the humor index, a comprehensive personality characteristic system is constructed, making the personalized generation of voice assistants richer, more delicate and accurate, and meeting users' diversified and personalized needs for voice assistants.

[0018] In some possible implementations, visual instructions include style cues, color cues, element cues, and composition cues; auditory instructions include timbre characteristics, speech rate range, emotional tone, and intonation patterns; and verbal instructions include the level of initiative in the conversation, preference for commonly used interjections, humor index, empathy index, and recommended verbal phrases for a preset driving scenario.

[0019] According to the above embodiments, through visual instructions such as style prompts and color prompts, auditory instructions such as timbre features and speech rate ranges, and dialogue instructions such as dialogue initiative and humor index, a detailed operation guide is provided for the generation of voice assistants. This ensures the standardization and controllability of the generation process, improves the consistency and repeatability of voice assistant generation, and facilitates large-scale deployment and maintenance.

[0020] Secondly, this application provides a control device, which includes: a first determining module for determining the user's corresponding zodiac sign type; a second determining module for determining personality trait tags based on the zodiac sign type, the personality trait tags including visual trait tags, auditory trait tags, and verbal trait tags; an instruction generation module for generating visual instructions, auditory instructions, and verbal instructions based on the personality trait tags; and an image generation module for generating a visual image of a voice assistant based on the visual instructions, generating the voice timbre of a voice assistant based on the auditory instructions, and generating an interactive dialogue script library for a voice assistant based on the verbal instructions.

[0021] Thirdly, this application provides a control device, including: a memory and a processor; the memory and the processor are coupled; the memory is used to store instructions executable by the processor; when the processor executes the instructions, it implements the image generation method of the voice assistant of any of the above embodiments.

[0022] Fourthly, this application provides a computer-readable storage medium storing at least one piece of program code, which is loaded and executed by a processor to implement the image generation method of the voice assistant in any of the above embodiments.

[0023] Fifthly, this application provides a computer program product, including a computer program; when the computer program is executed, it can implement the image generation method of the voice assistant in any of the above embodiments.

[0024] Sixthly, this application provides a vehicle, including: an in-vehicle infotainment system for loading a voice assistant; and a control device of any of the above embodiments; or a computer-readable storage medium of any of the above embodiments; or a computer program product of any of the above embodiments. Attached Figure Description

[0025] To more clearly illustrate the technical solutions in this application, the accompanying drawings used in some embodiments of this application will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings.

[0026] Figure 1 The flowchart of a voice assistant image generation method provided for some embodiments of this application Figure 1 ; Figure 2 The flowchart of a voice assistant image generation method provided for some embodiments of this application Figure 2 ; Figure 3 The flowchart of a voice assistant image generation method provided for some embodiments of this application Figure 3 ; Figure 4 The flowchart of a voice assistant image generation method provided for some embodiments of this application Figure 4 ; Figure 5 Schematic diagram of the structure of the control device provided in some embodiments of this application Figure 1 ; Figure 6 Schematic diagram of the structure of the control device provided in some embodiments of this application Figure 2 . Detailed Implementation

[0027] The technical solutions of this application will now be clearly and completely described with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0028] It should be noted that, in this application, the terms "exemplary" or "for example" are used to indicate examples, illustrations, or descriptions. Any embodiment or design described as "exemplary" or "for example" in this application should not be construed as being more preferred or advantageous than other embodiments or design solutions. Specifically, the use of terms such as "exemplary" or "for example" is intended to present the relevant concepts in a concrete manner.

[0029] Hereinafter, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of that feature.

[0030] In the description of this application, unless otherwise stated, " / " means "or". For example, A / B can mean A or B. "And / or" in this document is merely a description of the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A alone, A and B simultaneously, and B alone. Furthermore, "at least one" means one or more, and "multiple" means two or more.

[0031] In smart cockpits, voice assistants have become the core interaction interface. However, current mainstream in-vehicle voice assistants (such as those built into the system or third-party assistants integrated by various brands) are highly homogenized in appearance and interaction. Regardless of who the user is, they are usually faced with the same set of preset, universal virtual avatars (or only a few selectable skins) and standardized voice and scripts. This "one-size-fits-all" interactive experience lacks respect for and responsiveness to the individual user's identity and preferences, making it difficult to establish deep emotional connections and user stickiness, and reducing the fun and personalization of voice interaction.

[0032] As a popular cultural symbol, zodiac signs are often used by users for self-expression, social interaction, and psychological comfort. Many users have a strong sense of identification with the personality traits, lucky elements, and color preferences associated with their own zodiac sign. Integrating zodiac characteristics into in-vehicle voice assistants essentially transforms a widely recognized user identity label into a perceptible interactive experience. For example, a mysterious, profound, and magnetically toned assistant could be generated for a "Scorpio" user, or a faster-talking, lively, and intriguing assistant could be matched for a "Gemini" user. This identity-based customization can instantly bridge the gap between users and machines, giving cold technology a personalized warmth and character.

[0033] Currently, AIGC (Artificial Intelligence Generated Content) technology can achieve high-quality text generation, image generation, and speech synthesis. Large Language Models (LLMs) can generate dialogue scripts that conform to a pre-defined persona. Text-generated images or text-generated 3D models can create unique visual characters. Speech synthesis and voice cloning technologies can customize unique voices. However, how to organically integrate these technologies and automatically and intelligently generate a coordinated and unified multimodal interactive personality based on an abstract "constellation" label remains a challenge.

[0034] Therefore, this application provides a method for generating a voice assistant image. This method is mainly applied to intelligent terminal products that require voice interaction, such as in-vehicle intelligent systems, smart home devices, smart wearable devices, and personal electronic devices. In-vehicle intelligent systems are a classic application scenario for this method. As an important interface for human-vehicle interaction, the intelligent cockpit relies heavily on voice assistants, which have become one of the main ways for users to interact with and control the vehicle. Traditional in-vehicle voice assistants typically use fixed images and uniform interaction patterns, making it difficult to meet the personalized needs of different users. This application generates a unique voice assistant image for each user by mapping the association between constellation type and personality traits, including visual image, vocal characteristics, and interactive speech style, thereby improving user experience and interaction satisfaction.

[0035] This application uses an in-vehicle intelligent system as an example for illustration, which should not be considered a specific limitation on the application scenario. The method is also applicable to devices such as smart speakers, smart TVs, and smart air conditioners in the smart home field; smartphones, tablets, and smartwatches in the personal electronic device field; and AR glasses and VR headsets in the smart wearable device field. When applied to different products, the specific implementation method can be adjusted according to the product characteristics, but the core technical approach remains consistent.

[0036] The following is a detailed description of the image generation method for the voice assistant provided in the embodiments of this application.

[0037] like Figure 1 As shown in the figure, an embodiment of this application provides a method for generating the image of a voice assistant, which includes the following steps: S101. Determine the user's corresponding zodiac sign type.

[0038] In some embodiments, determining a user's zodiac sign type can be achieved by retrieving the user's date of birth information from the vehicle's user account system in response to the user's authorization command. The vehicle's user account system typically stores the user's personal information, including date of birth, contact information, vehicle preferences, and other data. To protect user privacy, the system sends an authorization request to the user before retrieving the date of birth information, and only accesses the relevant data after obtaining explicit authorization from the user. The authorization process employs encrypted communication and permission verification mechanisms to ensure the security of data transmission and access.

[0039] After obtaining the user's birth date information, the system determines the user's corresponding zodiac sign type based on the birth date information according to preset calculation rules. The zodiac sign type refers to the sun sign determined by the user's birth date, including twelve signs: Aries, Taurus, Gemini, Cancer, Leo, Virgo, Libra, Scorpio, Sagittarius, Capricorn, Aquarius, and Pisces. Each zodiac sign has its own unique personality traits, lucky elements, color preferences, and other cultural connotations.

[0040] Each zodiac sign corresponds to a predetermined date range. For example, Aries corresponds to March 21 to April 19, Taurus to April 20 to May 20, Gemini to May 21 to June 21, Cancer to June 22 to July 22, Leo to July 23 to August 22, Virgo to August 23 to September 22, Libra to September 23 to October 23, Scorpio to October 24 to November 22, Sagittarius to November 23 to December 21, Capricorn to December 22 to January 19, Aquarius to January 20 to February 18, and Pisces to February 19 to March 20.

[0041] By analyzing the month and day of the birth date and matching them with the aforementioned date range, the system determines the user's zodiac sign. For example, if a user's birth date is May 15th, the system will determine that this date falls within the range of April 20th to May 20th, thus identifying the user as a Taurus. If the birth date is February 25th, it falls within the range of February 19th to March 20th, identifying the user as a Pisces.

[0042] In other embodiments, determining a user's zodiac sign type can also be achieved by obtaining the user's input regarding their zodiac sign type and determining the user's zodiac sign type based on that input. This method is more direct and convenient, as users can input their zodiac sign information through various means such as voice, touchscreen, and gestures.

[0043] For example, the system can provide a zodiac sign selection interface, displaying icons and names of the twelve zodiac signs. Users can select their zodiac sign by clicking, swiping, or using voice commands. For instance, a user can say "I am a Leo" or touch the Leo icon on the screen to select it. After receiving the user's input, the system parses and validates the input to ensure that the entered zodiac sign type is valid. If the user's entered zodiac sign type is not within the twelve zodiac signs range, the system will prompt the user to re-enter it.

[0044] User-input-based recognition is particularly suitable for users who are unwilling to provide their birth date information or who have a clear understanding of their zodiac sign. This method also supports modification and updating of zodiac sign information; users can change their zodiac sign settings at any time, and the system will regenerate the voice assistant's avatar accordingly.

[0045] In practical applications, these two zodiac sign recognition methods can be used in combination to provide more flexible and accurate results. For example, the system can first attempt to obtain birth date information from the vehicle's user account system for automatic recognition; if this fails or the user refuses authorization, it switches to a recognition method based on user input. Alternatively, the system can use both methods simultaneously, and if the results from the two methods are inconsistent, it confirms the final zodiac sign type with the user.

[0046] S102. Determine personality trait labels based on astrological sign type. Personality trait labels include visual trait labels, auditory trait labels, and verbal trait labels.

[0047] Personality trait tags refer to a structured set of parameters used to describe the personalized characteristics of a voice assistant, including three dimensions: visual feature tags, auditory feature tags, and speech feature tags. Based on a given zodiac sign type, the system extracts the corresponding personality trait tags from a pre-defined zodiac-personality trait mapping map.

[0048] The Zodiac-Personality Trait Map is a structured knowledge base that defines a complete set of multimodal personality parameters for each zodiac sign. This map comprises the following three levels: Visual Feature Layer: Defines the visual elements and style characteristics corresponding to the zodiac sign, including symbols, representative graphics, lucky colors, and related materials. For example, the visual feature tags for Leo include: the symbol is a lion or crown, the representative graphic is radiating rays or a V shape, the lucky colors are gold and orange, and the related materials are metallic and silk textures.

[0049] Auditory Feature Layer: Defines the speech characteristics and timbre attributes corresponding to each zodiac sign, including emotional tone, speech rate and rhythm, timbre quality, and intonation pattern. For example, the auditory feature labels for Leo include: an emotional tone of confidence and high spirits, a speech rate and rhythm of medium to fast, a timbre quality of loud and full, and an intonation pattern of obvious fluctuations and intonation.

[0050] The dialogue feature layer defines the interaction style and dialogue characteristics corresponding to each zodiac sign, including the level of initiative in the conversation, preferred interjections, humor index, empathy index, and recommended dialogue for preset driving scenarios. For example, the dialogue feature tags for Leo include: high level of initiative in the conversation, preferred interjections such as "of course" and "no problem," medium humor index, above-average empathy index, and recommended dialogue for navigation scenarios such as "We have planned the best route for you, please follow with confidence."

[0051] In practical applications, after receiving the zodiac sign type, the system queries the mapping database to obtain the corresponding personality trait tags. To improve the accuracy and personalization of the mapping, the system also supports a fine-tuning mechanism based on users' historical behavioral data. For example, if the system finds that a user's actual interaction style differs from typical Aries characteristics, it can personalize the user's personality trait tags to better reflect the user's true traits.

[0052] S103. Generate visual instructions, auditory instructions, and verbal instructions based on personality trait tags.

[0053] After obtaining personality trait tags, the system needs to convert these abstract trait tags into specific generation instructions to drive the subsequent generation of visual image, voice tone and interactive scripts.

[0054] Visual instructions are a detailed set of prompts used to guide the generation of visual images, including style prompts, color prompts, element prompts, and composition prompts. The system generates structured visual instructions based on visual feature tags. For example, for Leo, the system might generate visual instructions such as: style prompts like "magnificent, regal, confident"; color prompts like "primarily gold, accented with orange and deep red"; element prompts like "lion image, crown element, radiating rays"; and composition prompts like "centrally symmetrical composition, image occupies the center of the frame, simple background," etc.

[0055] Auditory instructions are a set of acoustic parameters used to guide the generation of speech timbre, including timbre characteristics, speech rate range, emotional tone, and intonation pattern. The system generates quantified auditory instructions based on auditory feature labels. For example, for a Leo, the system might generate auditory instructions such as: timbre characteristics of "full mid-low frequencies, bright high frequencies, and overall resonant," speech rate range of "180-200 words per minute," emotional tone of "confident, positive, and uplifting," and intonation pattern of "distinct intonation tendency," etc.

[0056] Dialogue instructions are a set of dialogue rules used to guide the generation of interactive dialogue, including the level of initiative in the dialogue, preference for commonly used interjections, humor index, empathy index, and recommended dialogue for preset driving scenarios. The system generates specific dialogue instructions based on dialogue feature tags. For example, for a Leo, the system-generated dialogue instructions might include: a level of initiative of "actively providing information and suggestions, with a moderate frequency of asking questions," and a preference for affirmative words such as "of course," "no problem," and "okay."

[0057] S104. Generate the visual image of the voice assistant based on visual instructions, generate the voice tone of the voice assistant based on auditory instructions, and generate the interactive dialogue library of the voice assistant based on dialogue instructions.

[0058] After generating various instructions, the system calls the corresponding generation modules to execute the tasks of generating visual images, voice timbre, and interactive dialogue.

[0059] The generation of visual avatars is based on visual commands and can utilize artificial intelligence image generation technologies (such as text-based image models and 3D model generation technologies) to create the visual appearance of the voice assistant. The generated visual avatars include various forms such as static avatars, dynamic standby animations, and full-screen interactive avatars. Based on the style, color, elements, and composition requirements in the visual commands, the system generates visual avatars that conform to the characteristics of a zodiac sign. For example, for Leo, the system might generate a virtual avatar with a gold tone and a confident posture against a simple and elegant background.

[0060] Voice timbre generation is based on auditory commands, and speech synthesis technology can be used to create a unique timbre for the voice assistant. The system synthesizes a voice timbre that matches the characteristics of a zodiac sign based on the timbre features, speech rate range, emotional tone, and intonation pattern in the auditory commands. For example, for a Leo, the system-generated voice timbre might be loud and full, with a moderate to fast speech rate and a confident and uplifting tone.

[0061] The interactive dialogue library is generated based on dialogue instructions and can utilize Large Language Model (LLM) to generate dialogue data that matches astrological characteristics. The system generates rich interactive dialogues based on the degree of dialogue initiative, preference for frequently used interjections, humor index, empathy index, and recommended dialogue templates in the dialogue instructions. The interactive dialogue library includes opening lines, high-frequency question-and-answer templates, emotional response rules, and dialogue examples for typical scenarios. For example, for Leos, the system-generated interactive dialogue library might include phrases such as "Hello! I'm your personal voice assistant, how can I help you?", "No problem, I'll handle it for you right away," and "Of course, I'll arrange it for you right away," reflecting confidence and positivity.

[0062] like Figure 2 As shown, determining a user's corresponding zodiac sign type includes the following steps: S201. In response to the user's authorization command, obtain the user's date of birth information from the vehicle user account system.

[0063] In this embodiment, the system first needs to obtain the user's birth date information as the basic data for zodiac sign calculation. To ensure user privacy and security, the system employs an explicit authorization mechanism. When a user uses the personalized voice assistant function for the first time, the system will display an authorization request interface, explaining to the user that birth date information is required for zodiac sign calculation and personalized voice assistant generation. The user can choose to click the "Agree" or "Decline" button. The system will only perform the subsequent birth date information acquisition operation after the user explicitly clicks the "Agree" button.

[0064] The in-vehicle user account system refers to the user account management module integrated into the vehicle's infotainment system. It stores and manages user information, preferences, usage history, and other data. A user's date of birth is typically provided during account registration and stored in the account system's database. The system retrieves the user's date of birth in an encrypted manner by calling the account system's application programming interface (API).

[0065] The obtained birth date information includes data in three dimensions: year, month, and day. The system will perform format validation and validity checks on the obtained birth date information to ensure the accuracy and completeness of the data. If an incorrect data format or invalid date is detected, the system will prompt the user to re-enter or update the birth date information.

[0066] S202. Based on the birth date information, determine the user's corresponding zodiac sign type according to the preset calculation rules.

[0067] In this embodiment, after obtaining the user's birth date information, the system determines the user's zodiac sign type according to a preset calculation rule. The preset calculation rule can be based on the correspondence between Gregorian calendar dates and sun signs. For example, if the user's birth date is May 15, 1990, the system extracts the month as 5 and the day as 15. Referring to the zodiac date range table, it finds that May 15 falls within the Taurus date range (April 20 - May 20), therefore the user's zodiac sign type is determined to be Taurus.

[0068] After determining the zodiac sign type, the zodiac sign type identifier will be stored in the user's profile and will serve as a core driving parameter for the subsequent personality generation process. The determined zodiac sign type will also be displayed in the user interface for user confirmation. If the user believes the determined zodiac sign type is incorrect, they can correct it manually by entering it back into the user's profile.

[0069] like Figure 3 As shown in the embodiments of this application, the method for generating the image of a voice assistant further includes the following steps: S301. Encapsulate the visual image, voice, and interactive dialogue library into a multimodal personality package.

[0070] In this embodiment, after generating the visual image, voice tone, and interactive dialogue library, these dispersed multimodal resources are integrated and packaged to form a structured "multimodal personality package." The multimodal personality package refers to a data package containing a complete personality definition for the voice assistant, including visual resource files, voice model files, dialogue logic configuration files, etc.

[0071] For example, a multimodal personality package includes the following components: Visual resource files: including static avatar images, dynamic standby animations, and full-screen interactive images of the voice assistant.

[0072] Speech model files: These include the voice assistant's proprietary timbre model and speech samples. The speech model files contain acoustic feature parameters of the timbre, such as fundamental frequency range, formant frequencies, and spectral envelope, which are used for speech synthesis at runtime.

[0073] The dialogue logic configuration file includes structured data for the interaction script library, dialogue rule definitions, and emotional response logic. It defines the voice assistant's response strategies, commonly used script templates, and rules for using interjections in different scenarios.

[0074] Metadata file: Includes basic information about the personality package, such as zodiac sign type, generation time, version number, and verification status. The metadata file is used for personality package management and version control.

[0075] S302. Conduct a personality tone matching assessment on the multimodal personality package to ensure that the voice assistant's image, tone, and speech are matched.

[0076] In this embodiment, after encapsulating the multimodal personality package, a consistency check is performed on the personality package to ensure that the visual image, voice tone, and interactive script maintain consistency in personality tone. Personality tone matching assessment refers to evaluating the degree of consistency of multimodal content at the personality level through quantitative indicators.

[0077] For example, a multi-dimensional assessment index is used to evaluate personality tone fit, including: Visual-auditory consistency assessment: This evaluates whether the stylistic features of a visual image match the acoustic features of a speech timbre. For example, it assesses the correlation between the color saturation and line complexity of a visual image and the brightness and fullness of a speech timbre. Specifically, a pre-trained visual-auditory consistency assessment model can be used to calculate the similarity score between visual feature vectors and auditory feature vectors.

[0078] Visual-to-verbal consistency assessment: This evaluates whether the visual characteristics of the visual identity match the language style of the interactive language. For example, it assesses the correlation between the formality and approachability of the visual identity and the politeness and humor of the verbal language. Specifically, this can be achieved by extracting the linguistic features of the verbal language using natural language processing techniques and comparing them with the stylistic features of the visual identity.

[0079] Auditory-verbal consistency assessment: This evaluates whether the emotional characteristics of the voice tone match the emotional expression of the spoken message. For example, it assesses the consistency between the emotional tone of the voice (e.g., positive, negative, neutral) and the emotional tendency of the spoken message (e.g., encouraging, comforting, objective). Specifically, this can be achieved by extracting the emotional features of the voice and the spoken message separately using sentiment analysis techniques, and then calculating the emotional matching degree.

[0080] Based on the evaluation results of the above three dimensions, a comprehensive matching score is calculated. When the comprehensive matching score is greater than or equal to the threshold, the multimodal personality package is deemed to have passed the consistency check; when the comprehensive matching score is less than the threshold, the multimodal personality package is deemed to have conflicts and needs to be adjusted.

[0081] S303. If a conflict is detected, adjust the visual image, tone, and interactive script library of the multimodal personality package.

[0082] In this embodiment, when a conflict is detected in the multimodal personality package, the system needs to initiate a conflict correction process to adjust the content of the conflicting modalities.

[0083] In some embodiments, the conflict resolution process includes four steps: conflict location, adjustment strategy formulation, adjustment execution, and re-verification.

[0084] Conflict localization: The system analyzes the evaluation results across various dimensions to determine the specific location and type of conflict. For example, low scores for visual and auditory consistency indicate a mismatch between the visual image and the voice tone. Low scores for auditory and verbal consistency indicate a mismatch between the voice tone and the interactive script. By comparing the feature vectors of each modality, the specific features leading to the conflict are identified.

[0085] Adjustment Strategy Formulation: Based on the conflict localization results, the system formulates corresponding adjustment strategies. These strategies include prioritizing modal selection, determining the magnitude of adjustment, and determining the direction of adjustment. The system determines the modalities to be adjusted based on the severity of the conflict and the adjustability of each modality. For example, if the cost of generating the visual image is high and the adjustment space is limited, the system may prioritize adjusting the voice tone and interactive script.

[0086] Adjustment Execution: Based on the adjustment strategy, the corresponding modal content is adjusted. Adjustment methods include parameter fine-tuning, regeneration, and feature replacement. For example, to adjust the timbre of speech, the system can achieve this by adjusting the acoustic parameters of the model (such as fundamental frequency and formants). To adjust interactive dialogue, the system can achieve this by modifying the dialogue template and adjusting the frequency of use of interjections.

[0087] Re-verification: After the adjustment is completed, the system re-executes the personality tone matching assessment and calculates the adjusted overall matching score. If the adjusted score meets the threshold requirement, the conflict correction is considered successful. If the threshold requirement is not met, the adjustment process continues until the verification is passed or the maximum number of adjustments is reached.

[0088] like Figure 4 As shown in the embodiments of this application, the method for generating the image of a voice assistant further includes the following steps: S401. Obtain driving status information, which includes time context, scene context, and user status context.

[0089] In this embodiment, during the operation of the voice assistant, the system needs to acquire driving status information in real time in order to dynamically adjust the interactive behavior of the voice assistant according to the current context. Driving status information refers to a multi-dimensional data set reflecting the current driving environment and user status, including three dimensions: time context, scene context, and user status context.

[0090] Time context refers to time-related environmental information, including time and date type. Time information includes the current specific time (e.g., 8:30, 14:45, 23:10), which can be obtained through the vehicle system's clock module. Date type information includes whether the current date is a weekday or a holiday (e.g., Monday to Friday are weekdays, Saturday, Sunday, and public holidays are holidays), which can be obtained through the calendar module.

[0091] Scene context refers to environmental information related to the driving scenario, including navigation destination type, road condition status, and driving duration. Navigation destination type information includes the category of the current navigation destination (e.g., company, shopping mall, scenic spot, residence, hospital, etc.), which can be obtained through the navigation module. Road condition status information includes the current road congestion level (e.g., smooth traffic, slow traffic, congested) and road type (e.g., highway, urban road, rural road), which can be obtained through road condition services. Driving duration information includes the duration of this drive (e.g., 30 minutes, 2 hours), which can be obtained through the timing module.

[0092] User state context refers to information related to a user's current state, including user emotions and user preferences learned through historical interactions. User emotional information can be obtained through voice emotion recognition technology, which analyzes the acoustic features of the user's voice (such as speech rate, pitch, and volume) to determine the user's emotional state (such as calm, tired, excited, anxious, etc.). User preference information can be learned from historical interaction data through machine learning algorithms, including user preferences for voice assistant response methods (such as concise or detailed), level of interest in preset topics, and receptiveness to humor.

[0093] In practice, the above driving status information can be obtained in real time through the vehicle system's sensors, network services, and data analysis modules, and updated to the context information cache.

[0094] S402. Based on driving status information, adjust the visual image, tone of voice, and interactive dialogue library according to the preset personality elastic mapping rules.

[0095] In these embodiments, after obtaining driving status information, the interactive behavior of the voice assistant needs to be dynamically adjusted according to the current context. Personality elastic mapping rules refer to a set of rules defining the magnitude and direction of adjustment of the basic personality in different contexts, ensuring scenario adaptability while maintaining the stability of the personality core.

[0096] In some embodiments, the personality flexibility mapping rule may include three sub-rule sets: time adjustment rule, scenario adjustment rule, and user state adjustment rule.

[0097] The time adjustment rules define adjustment strategies based on time context. For example, in the early morning (6:00-9:00), for all zodiac signs' voice assistants, the speaking speed and volume should be appropriately reduced, and a gentler greeting should be used to avoid information overload in the early morning. For example, "Good morning, the weather is nice today" could be adjusted to "Good morning, did you sleep well last night? The weather is great today." In the late night (22:00-24:00), for lively zodiac signs (such as Gemini and Sagittarius), the speaking speed and activity level should be appropriately reduced, and a quieter interaction method should be used. For calm and composed zodiac signs (such as Capricorn and Scorpio), the original concise style should be maintained, but more caring expressions should be added.

[0098] On weekdays, for career-oriented zodiac signs (such as Capricorn and Virgo), add information prompts related to work efficiency. For leisure-oriented zodiac signs (such as Leo and Sagittarius), add suggestions related to leisure and entertainment. On holidays, for all zodiac signs, appropriately increase lighthearted and cheerful expression elements and reduce serious work-related topics.

[0099] The scenario adjustment rules define adjustment strategies based on the scenario context. For example, in congested traffic, for impatient zodiac signs (such as Aries and Leo), soothing expressions and music recommendations are added. For patient zodiac signs (such as Taurus and Capricorn), traffic analysis and alternative route suggestions are provided.

[0100] For long-distance driving, fatigue alerts and rest suggestions will be added to the voice assistant for all zodiac signs, but the wording will be adjusted according to the characteristics of each sign. At shopping mall destinations, shopping-oriented signs (such as Libra and Taurus) will receive mall promotions and shopping suggestions. Practical signs (such as Virgo and Capricorn) will be provided with mall layout and parking information.

[0101] User state adjustment rules define adjustment strategies based on the user's state context. For example, when a user is tired, for lively signs, reduce activity and humor, and increase caring expressions. For calm signs, maintain a simple style, but increase the frequency of reminders. When a user is excited, for introverted signs (such as Cancer and Scorpio), appropriately increase relatable expressions. For extroverted signs (such as Leo and Sagittarius), increase interactivity and sharing suggestions.

[0102] When users prefer concise responses, for talkative zodiac signs (such as Gemini and Aquarius), shorten the response length and highlight the core information. For concise zodiac signs (such as Capricorn and Virgo), maintain the original concise style.

[0103] In practice, based on the current driving status information, corresponding personality elasticity mapping rules can be matched to calculate the adjustment parameters for each modality. These adjustment parameters can include speech rate adjustment coefficients, volume adjustment coefficients, and speech style adjustment parameters (such as the adjustment range for humor and empathy).

[0104] The foregoing mainly describes the solutions of the embodiments of this application from a methodological perspective. It is understood that, in order to achieve the above functions, the control device includes at least one of the hardware structures and software modules corresponding to the execution of each function. Those skilled in the art should readily recognize that, in conjunction with the units and algorithm steps of the various examples described in the embodiments applied herein, the embodiments of this application can be implemented in hardware or a combination of hardware and computer software. Whether a function is executed by hardware or by computer software driving hardware depends on the preset application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each preset application, but such implementation should not be considered beyond the scope of the embodiments of this application.

[0105] This application embodiment can divide the control device into functional modules according to the above method embodiment. For example, each function can be divided into a separate functional module, or two or more functions can be integrated into one functional module. The integrated module can be implemented in hardware or software. It should be noted that the module division in this application embodiment is illustrative and only represents one logical functional division. In actual implementation, there may be other division methods. The following description uses the example of dividing each functional module according to each function.

[0106] This application also provides a voice assistant image generation system, including: a user identity and constellation analysis module, a constellation feature multidimensional knowledge base, a multimodal personality generation AI Agent engine, an image and voice multimodal synthesis module, a personality package assembly and verification module, and a runtime interaction adaptation module.

[0107] The user identity and zodiac sign analysis module is used to obtain birth date information from the vehicle system user account, determine the user's zodiac sign through a zodiac calculation algorithm, and trigger the personality generation process.

[0108] The Zodiac Sign Feature Multidimensional Knowledge Base is used to store the structured mapping relationship between zodiac signs and multimodal personality traits, specifically including three dimensions: visual feature layer, auditory feature layer, and personality language layer.

[0109] The multimodal personality generation AI Agent engine, as the core coordination unit of the system, is built on a large language model. It receives target constellation information, parses multidimensional feature tags from the knowledge base, plans and coordinates the generation tasks of visual, auditory, and verbal communication, and performs cross-consistency checks to ensure that the multimodal output is consistent in personality tone.

[0110] The image and voice multimodal synthesis module includes a visual generation subunit and a voice generation subunit, which are used to call the underlying AIGC model to generate the visual image and unique voice of the voice assistant according to the detailed generation instructions issued by the AI ​​Agent engine.

[0111] The personality package assembly and verification module is used to encapsulate the generated image resources, voice models, and dialogue logic configuration files to form a structured "multimodal personality package" and perform a final consistency verification.

[0112] The runtime interaction adaptation module, deployed on the vehicle's infotainment system, is used to load the corresponding personality package based on the current user account, connect to the underlying voice interaction system, and dynamically fine-tune the interaction behavior at runtime in conjunction with the real-time context.

[0113] The knowledge base consists of a zodiac sign and personality trait mapping map (hereinafter referred to as the "trait map"). This map not only includes typical visual symbols of each zodiac sign (such as the bull and earth tones for Taurus; the fish and ocean blue-purple for Pisces), but more importantly, it defines their corresponding computable "interactive personality parameter set," which specifically includes: Voice parameter dimensions: emotional tone (composure index, liveliness index), speech rate and rhythm (fast-slow scale), timbre quality (warmth, brightness, depth), and intonation pattern (fluctuation, inflection tendency).

[0114] Conversation style dimensions: level of initiative in the conversation (frequency of proactive care, tendency to ask questions), preference for commonly used interjections, humor index, empathy index, and recommended conversation templates for pre-set driving scenarios (traffic jams, long distances, fatigue reminders).

[0115] Behavioral logic dimensions: Information detail preference (concise / detailed), decision support style (guiding / delegating), and emotional feedback pattern (rational analysis / emotional support).

[0116] The AI ​​Agent engine uses the target constellation as an index to extract the aforementioned parameter set from the feature map, forming a "multimodal personality generation blueprint," and then uses this blueprint to parameterize and drive the generation of content for each modality, ensuring that all outputs originate from the same set of personality definitions.

[0117] The AI ​​Agent engine acts as the "director," employing a three-stage workflow of "planning-generation-verification" to ensure a high degree of consistency between visual, auditory, and verbal communication at the personality level.

[0118] Planning phase: Based on the parameter set extracted from the feature map, generate three-way collaborative generation instructions: 1. Visual instructions: Send detailed prompts to the compositing module, including style, color, elements, composition, etc. (e.g., "Leo: regal demeanor, gold tone, lion image or emblem elements, gorgeous texture, confident posture").

[0119] 2. Auditory instructions: Send acoustic parameters, including timbre characteristics, speech rate range, and emotional tone, to the synthesis module (e.g., "Leo: loud timbre, full mid-low frequency, medium to fast speech rate, confident and uplifting tone").

[0120] 3. Conversation Instructions: The engine's built-in LLM automatically generates a seed dialogue corpus based on personality tags (such as "leadership", "generosity", "confidence"), including opening remarks, high-frequency question and answer templates, emotional response rules, and typical scenario dialogue examples.

[0121] 4. Generation stage: The synthesis module performs visual and auditory generation tasks in parallel, while the engine synchronously generates the LLM of the speech library.

[0122] 5. Verification Phase: The engine performs a "cross-consistency check," evaluating the personality tone matching of the generated image preview, voice sample, and speech sample. If a conflict is detected (e.g., an image that is dignified but the voice is light and cheerful), a correction process is triggered, adjusting the relevant generation parameters until the verification is passed. For example, when generating a visual image with a regal demeanor for a "Leo," the system ensures that the synchronously generated voice has a confident and resonant quality, while the speech appropriately incorporates expressions that reflect "leadership" or "generosity," thus forming a highly unified "regal personality" from the outside in.

[0123] In some embodiments, the generated base personality package is not static. The runtime interaction adaptation module has a built-in "scene perception and dynamic fine-tuning engine" that obtains multi-dimensional real-time contextual information through the vehicle system, including: 1. Time context: time (early morning / late night), date type (weekday / holiday).

[0124] 2. Context: Navigation destination type (company / shopping mall / scenic spot), road condition (congested / smooth), driving time.

[0125] 3. User state context: Initially determine the user's emotions (calm / fatigued / excited) through voice emotion recognition technology, and learn user preferences through historical interactions.

[0126] Based on the aforementioned contextual information, the fine-tuning engine dynamically adjusts the current interactive behavior according to preset "personality flexibility mapping rules." The adjustment range is controlled within the flexibility allowed by the "basic personality," ensuring the stability of the personality core while achieving scene adaptability. For example: The calm and reserved Capricorn assistant maintains a concise and efficient speaking style during weekday commutes. However, in weekend travel scenarios, the tone of their navigation announcements can be subtly adjusted to convey a sense of ease and anticipation. For example, instead of "Turn right in 300 meters ahead," it could be changed to "Turn right in 300 meters ahead, and you're one step closer to the seaside cafe you're looking for! Are you excited?"

[0127] The enthusiastic and lively "Sagittarius" assistant automatically reduces its speaking speed and volume when it detects that the user is speaking slowly in the morning and may not be fully awake. It then announces the weather and schedule in a gentler way to avoid information overload and sensory stimulation in the morning.

[0128] This application upgrades the in-vehicle voice assistant from a functional tool to an emotionally engaging partner with a distinct "personality." By deeply integrating the user's astrological sign as an identity tag and utilizing AI technology to automatically generate a complete set of matching interactive personas, a deep level of personalization is achieved. This not only enhances the user's sense of novelty, surprise, and emotional identification, strengthening their sense of brand exclusivity and loyalty, but also explores a new path for multimodal intelligent adaptation based on cultural identity at the interaction design level, improving the differentiated competitiveness and humanistic warmth of the smart cockpit.

[0129] In a specific embodiment, the development environment can be set up as follows: 1. Hardware Environment: The in-vehicle infotainment system must be network-enabled and possess certain graphics and audio processing capabilities. Some generation tasks can be completed on a more powerful cloud server.

[0130] 2. Software Environment: The vehicle-mounted system needs to integrate an account management module and an AI inference framework. The server-side needs to deploy an AI Agent development framework, AIGC model services (such as image generation and speech synthesis), knowledge base management services, and a dialogue model fine-tuning platform.

[0131] System integration and testing are as follows: 1. Knowledge Base Construction and Model Preparation: Collect and structure constellation culture data to build a feature knowledge base. Pre-train or select visual and speech synthesis models for image generation.

[0132] 2. In-vehicle integration: Add a "Personalized Personality Management" module to the in-vehicle voice assistant framework to interface with the above services and enable the download, switching, and runtime management of personality packages.

[0133] 3. Testing and Optimization: Recruit users from different zodiac signs to conduct experience testing. Evaluate the acceptability, consistency, fun, and positive emotional feedback of the generated personality. Optimize the accuracy of the zodiac feature map and the generation strategy of the AI ​​Agent based on the test results.

[0134] Figure 5 This is a schematic diagram of a control device 500 provided in an embodiment of this application. The control device 500 is applied to a vehicle and can execute the voice assistant image generation method provided in the above-described method embodiment. For example... Figure 5 As shown, the control device 500 includes: a first determining module 501, a second determining module 502, an instruction generating module 503, and an image generating module 504.

[0135] The first determining module 501 is used to determine the user's corresponding zodiac sign type; The second determining module 502 is used to determine personality trait labels based on constellation type. The personality trait labels include visual trait labels, auditory trait labels, and verbal trait labels. The instruction generation module 503 is used to generate visual instructions, auditory instructions, and verbal instructions based on personality trait tags; The image generation module 504 is used to generate the visual image of the voice assistant based on visual instructions, generate the voice tone of the voice assistant based on auditory instructions, and generate the interactive dialogue library of the voice assistant based on dialogue instructions.

[0136] In some embodiments, the control device 500 further includes an adjustment module, which is used to encapsulate the visual image, voice tone, and interactive script library into a multimodal personality package; to perform a personality tone matching assessment on the multimodal personality package to ensure that the image, voice tone, and script of the voice assistant match; and to adjust the visual image, voice tone, and interactive script library of the multimodal personality package if a conflict is detected.

[0137] In some embodiments, the adjustment module is further configured to obtain driving status information, including time context, scene context and user status context; and adjust the visual image, voice and interactive dialogue library according to the preset personality elastic mapping rules based on the driving status information.

[0138] The control device 500 described above can be implemented in hardware or as a software functional module. When implemented as a software functional module, the control device 500 may include a processor and a memory. The processor is coupled to the memory, and the memory is used to store instructions executable by the processor. When the processor executes instructions, it can execute the image generation method for the voice assistant in Embodiment 1 described above.

[0139] In implementing the functions of the integrated modules described above using hardware, this application provides a possible structure for the control device 600 involved in the above embodiments. For example... Figure 6 As shown, the control device 600 includes a processor 602 and a bus 604. Optionally, the control device 600 may also include a memory 601; alternatively, the control device 600 may also include a communication interface 603.

[0140] Processor 602 may implement or execute various exemplary logic blocks, modules, and circuits described in conjunction with embodiments of this application. Processor 602 may be a central processing unit, a general-purpose processor, a digital signal processor, an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA), or other programmable logic devices, transistor logic devices, hardware components, or any combination thereof. It may implement or execute various exemplary logic blocks, modules, and circuits described in conjunction with embodiments of this application. Processor 602 may also be a combination that implements computing functions, such as including one or more microprocessor combinations, a combination of a DSP and a microprocessor, etc.

[0141] Communication interface 603 is used to connect to other devices via a communication network. This communication network can be Ethernet, wireless access network, wireless local area network (WLAN), etc.

[0142] The memory 601 may be a read-only memory (ROM) or other type of static storage device capable of storing static information and instructions, random access memory (RAM) or other type of dynamic storage device capable of storing information and instructions, or electrically erasable programmable read-only memory (EEPROM), disk storage medium or other magnetic storage device, or any other medium capable of carrying or storing desired program code in the form of instructions or data structures and accessible by a computer, but is not limited thereto.

[0143] In one possible implementation, the memory 601 can exist independently of the processor 602. The memory 601 can be connected to the processor 602 via a bus 604 and is used to store instructions or program code. When the processor 602 calls and executes the instructions or program code stored in the memory 601, it can implement the image generation method for the voice assistant provided in this embodiment. In another possible implementation, the memory 601 can also be integrated with the processor 602.

[0144] Bus 604 can be an extended industry standard architecture (EISA) bus, etc. Bus 604 can be divided into address bus, data bus, control bus, etc. For ease of representation, Figure 6 The 604 bus is represented by a single thick line, but this does not mean that there is only one bus 604 or only one type of bus 604.

[0145] Some embodiments of this application provide a computer-readable storage medium (e.g., a non-transitory computer-readable storage medium) storing at least one piece of program code that, when loaded and executed by a processor, causes a computer to perform a voice assistant image generation method as described in any of the above embodiments.

[0146] For example, the aforementioned computer-readable storage media may include, but are not limited to: magnetic storage devices (e.g., hard disks, floppy disks, or magnetic tapes), optical discs (e.g., compact disks (CDs), digital versatile disks (DVDs), etc.), smart cards, and flash memory devices (e.g., erasable programmable read-only memory (EPROMs), cards, sticks, or key drives, etc.). The various computer-readable storage media described in this application may represent one or more devices and / or other machine-readable storage media for storing information. The term "machine-readable storage medium" may include, but is not limited to, wireless channels and various other media capable of storing, containing, and / or carrying instructions and / or data.

[0147] This application provides a computer program product containing instructions that, when run on a computer, cause the computer to execute the voice assistant image generation method of any of the above embodiments.

[0148] This application provides a vehicle, including: an in-vehicle infotainment system for loading a voice assistant; and a control device as described in any of the preceding embodiments; or a computer-readable storage medium as described in any of the preceding embodiments; or a computer program product as described in any of the preceding embodiments.

[0149] The vehicle provided in this application may be a passenger vehicle or a freight vehicle, and may also be an electric vehicle or a hybrid vehicle. This application does not limit the specific purpose or power type of the vehicle, and the choice can be made according to actual needs.

[0150] A vehicle consists of a body and wheels. The body is used for passengers and for carrying goods, while the wheels are mounted underneath the body to support it and allow it to roll on the road, thus enabling the vehicle to move.

[0151] In some possible examples, the vehicle is equipped with a control system, which typically adopts a layered distributed architecture. From the bottom layer to the top layer, it can be roughly divided into a perception layer, a control layer, a coordination layer, and an interaction layer. The layers communicate with each other through an in-vehicle network.

[0152] The perception layer mainly consists of various sensors distributed inside and outside the vehicle, including but not limited to external environment cameras, millimeter-wave radar, lidar, ultrasonic sensors, in-vehicle driver monitoring cameras, microphone arrays, and various vehicle status sensors (such as wheel speed sensors, inertial measurement units, temperature sensors, etc.). The perception layer is responsible for collecting multi-dimensional data such as the vehicle's own operating status, driver behavior, and external driving environment in real time.

[0153] The control layer consists of dozens to hundreds of electronic control units (ECUs), distributed across multiple functional domains including powertrain, chassis, body, intelligent driving, and infotainment. Each ECU embeds real-time control software that performs closed-loop control of the vehicle's actuators based on preset control strategies or upper-level commands, and generates corresponding alarm signals when abnormal conditions are detected. Typical ECUs include the engine control unit, transmission control unit, brake control unit, steering control unit, vehicle stability control unit, airbag control unit, intelligent driving domain controller, and in-vehicle infotainment unit.

[0154] The coordination layer typically exists in the form of a domain controller or a central computing platform, responsible for cross-domain data fusion, global state management, and collaborative decision-making. The coordination layer centrally processes and schedules the sensing data and control commands that were originally scattered across various functional domains, connecting downwards to various electronic control units and supporting human-machine interaction functions upwards.

[0155] The interaction layer mainly includes in-cabin display devices (such as instrument panel, central control screen, head-up display), voice interaction system, haptic feedback device, etc., which are responsible for presenting vehicle status, warning information and driving suggestions to the driver in the form of visual, auditory or tactile, while receiving the driver's touch, voice and other input commands.

[0156] Data transmission and interaction between different layers are achieved through the vehicle bus network. Common vehicle bus protocols include CAN (Controller Area Network), CAN FD (CAN with Flexible Data-Rate), LIN (Local Interconnect Network), FlexRay, and vehicle Ethernet, which supports high-bandwidth data transmission. Among these, CAN and CAN FD buses are widely used for communication in real-time control domains such as powertrain and chassis, while vehicle Ethernet is gradually being applied to high-bandwidth sensor data transmission in the intelligent driving domain and multimedia interaction scenarios in the cockpit domain.

[0157] In some possible examples, the in-vehicle infotainment unit, as the core of the in-vehicle computing platform, is equipped with a processor and memory to execute the image generation method of the voice assistant provided in the embodiments of this application.

[0158] For example, the control device of this application is arranged in the vehicle infotainment unit.

[0159] For example, the in-vehicle infotainment unit is equipped with multiple functional modules for performing the various method steps.

[0160] The above are merely specific embodiments of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions within the technical scope disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

Claims

1. A method for generating the image of a voice assistant, characterized in that, include: Determine the user's corresponding zodiac sign type; Personality trait labels are determined based on the aforementioned zodiac sign type, and these personality trait labels include visual trait labels, auditory trait labels, and verbal trait labels. Visual instructions, auditory instructions, and verbal instructions are generated based on the personality trait tags. The visual image of the voice assistant is generated based on the visual instructions, the voice timbre of the voice assistant is generated based on the auditory instructions, and the interactive dialogue script library of the voice assistant is generated based on the dialogue script instructions.

2. The method according to claim 1, characterized in that, Determining the user's corresponding zodiac sign type includes: In response to the user's authorization command, the user's date of birth information is obtained from the vehicle user account system; Based on the birth date information, the user's corresponding zodiac sign type is determined according to a preset calculation rule.

3. The method according to claim 1, characterized in that, Determining the user's corresponding zodiac sign type includes: Obtain the user's input regarding the zodiac sign type, and determine the zodiac sign type corresponding to the user based on the user's input.

4. The method according to claim 1, characterized in that, Also includes: The visual image, the voice, and the interactive dialogue library are encapsulated into a multimodal personality package; The personality tone matching degree of the multimodal personality package is evaluated to ensure that the image, tone and speech of the voice assistant are matched. If a conflict is detected, the visual image, voice, and interactive dialogue library of the multimodal personality package are adjusted.

5. The method according to claim 1, characterized in that, Also includes: Obtain driving status information, which includes time context, scene context, and user status context; Based on the driving status information, the visual image, the voice tone, and the interactive dialogue library are adjusted according to the preset personality elastic mapping rules.

6. The method according to claim 5, characterized in that, The time context includes time and date types; The scenario context includes navigation destination type, road condition status, and driving duration; The user state context includes user emotions and user preferences learned through historical interactions.

7. The method according to any one of claims 1 to 6, characterized in that, The visual feature labels include symbols, representative graphics, lucky colors, and related materials; And / or, the auditory feature tags include emotional tone, speech rate and rhythm, timbre and intonation patterns; And / or, the dialogue feature tags include the degree of initiative in the conversation, preference for commonly used interjections, humor index, empathy index, and recommended dialogue when facing preset driving scenarios.

8. The method according to any one of claims 1 to 6, characterized in that, The visual instructions include style prompts, color prompts, element prompts, and composition prompts; The auditory instructions include timbre characteristics, speech rate range, emotional tone, and intonation pattern; The dialogue instructions include the level of initiative in the conversation, preference for commonly used interjections, humor index, empathy index, and recommended dialogue for facing preset driving scenarios.

9. A control device, characterized in that, include: The first determining module is used to determine the user's corresponding zodiac sign type; The second determining module is used to determine personality trait labels based on the constellation type, wherein the personality trait labels include visual trait labels, auditory trait labels, and verbal trait labels. The instruction generation module is used to generate visual instructions, auditory instructions, and verbal instructions based on the personality trait tags. The image generation module is used to generate a visual image of the voice assistant based on the visual instructions, generate the voice timbre of the voice assistant based on the auditory instructions, and generate an interactive dialogue library for the voice assistant based on the dialogue instructions.

10. A control device, characterized in that, include: Memory and processor; Memory and processor are coupled; The memory is used to store instructions that can be executed by the processor; When the processor executes the instructions, it performs the image generation method for the voice assistant as described in any one of claims 1 to 8.

11. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores at least one piece of program code, which is loaded and executed by a processor to implement the image generation method for a voice assistant as described in any one of claims 1 to 8.

12. A computer program product, characterized in that, It includes a computer program; when the computer program is executed, it is capable of implementing the image generation method for the voice assistant as described in any one of claims 1 to 8.

13. A vehicle, characterized in that, include: In-vehicle infotainment system, used to load voice assistants; And, the control device as described in claim 10; Alternatively, the computer-readable storage medium as claimed in claim 11; or the computer program product as claimed in claim 12.

Images