Intelligent broadcasting method and device, electronic equipment and storage medium

Abstract

The application discloses an intelligent broadcasting method and device, electronic equipment and storage medium, and relates to computer technology. The method comprises the following steps: when it is determined that a current user is a common user based on a received face image, acquiring current scene data and generating at least two scene features, each scene feature comprising a corresponding feature priority; wherein the feature priority corresponding to each scene feature is determined by adjusting historical scene feedback records; when it is determined that a proactive broadcasting condition is met according to the feature priority corresponding to each scene feature, acquiring a content preference database associated with the current user; wherein the interest priority corresponding to each content preference identifier is determined by adjusting historical content feedback records; and broadcasting content information corresponding to each content preference identifier according to the interest priority. The scheme can automatically identify the personalized preferences of a user based on the identity of the user, and proactively provide a broadcasting service, thereby improving the intelligent interaction experience of the user.

Description

Technical Field

[0001] This application relates to the field of computer technology, and in particular to an intelligent broadcasting method, device, electronic device, and storage medium. Background Technology

[0002] With the development of artificial intelligence and Internet of Things (IoT) technologies, intelligent voice interaction systems have been widely integrated into various terminal devices, such as smartphones, smart speakers, wearable devices, and in-vehicle systems. These systems typically rely on large-scale models to provide Automatic Speech Recognition (ASR), Natural Language Understanding (NLU), and Text-to-Speech (TTS) capabilities, performing tasks such as information retrieval, content playback, and device control based on user voice commands. Some high-end systems also support multi-turn dialogue, enabling them to respond to detailed follow-up questions from users.

[0003] However, existing intelligent broadcasting systems generally adopt a passive response mode, requiring users to actively wake up the device and issue explicit commands before the system can retrieve content and provide broadcasting services. This mode fails to meet users' needs for an intelligent and seamless interactive experience. Summary of the Invention

[0004] This application provides an intelligent broadcasting method, device, electronic device, and storage medium, which can enhance the user's intelligent interactive experience by proactively providing broadcasting services to the user.

[0005] Firstly, this application provides an intelligent broadcasting method, including: When the current user is determined to be a frequent user based on the received facial image, the current scene data is obtained and at least two scene features are generated, each scene feature including a corresponding feature priority; wherein, the feature priority corresponding to each scene feature is determined by adjusting historical scene feedback records; When the active broadcasting condition is met based on the feature priority corresponding to each of the scene features, a content preference database associated with the current user is obtained. The content preference database contains at least two content preference identifiers, and each content preference identifier includes a corresponding interest priority. The interest priority corresponding to each content preference identifier is determined by adjusting historical content feedback records. The content information corresponding to each content preference identifier is broadcast in order of interest priority.

[0006] Secondly, this application provides an intelligent broadcasting device, the device comprising: The first acquisition module is used to acquire current scene data and generate at least two scene features when determining that the current user is a frequent user based on the received facial image. Each scene feature includes a corresponding feature priority. The feature priority corresponding to each scene feature is determined by adjusting historical scene feedback records. The second acquisition module is used to acquire a content preference database associated with the current user when the active broadcasting conditions are met based on the feature priority corresponding to each scene feature. The content preference database contains at least two content preference identifiers, and each content preference identifier includes a corresponding interest priority. The interest priority corresponding to each content preference identifier is determined by adjusting historical content feedback records. The content broadcasting module is used to broadcast the content information corresponding to each content preference identifier in order of interest priority.

[0007] Thirdly, this application also provides an electronic device, the electronic device comprising: At least one processor; and a memory communicatively connected to said at least one processor; The memory stores a computer program that can be executed by the at least one processor, and the computer program is executed by the at least one processor to enable the at least one processor to execute the intelligent broadcasting method described in any embodiment of this application.

[0008] Fourthly, this application also provides a computer-readable storage medium storing computer instructions that, when executed by a processor, implement the intelligent broadcasting method described in any embodiment of this application.

[0009] Fifthly, this application also provides a computer program product, including a computer program that, when executed by a processor, implements the intelligent broadcasting method described in any embodiment of this application.

[0010] The intelligent broadcasting scheme provided in this application embodiment firstly, when determining that the current user is a frequent user based on the received facial image, it can generate multiple scene features through the current scene data, and determine whether the conditions for proactive broadcasting are met by combining the feature priorities corresponding to each scene feature; it can intelligently decide whether to broadcast based on the specific scene in which the user is located, avoiding blind interruption or missing broadcasting opportunities, and enhancing environmental awareness and adaptive capabilities. Secondly, after determining that the conditions for proactive broadcasting are met, it can broadcast content information corresponding to content preference identifiers according to interest priority order; it realizes personalized content recommendation and proactive broadcasting, making the broadcast content more in line with the user's recent interests, without the need for passive response. Finally, the feature priorities corresponding to scene features and the interest priorities corresponding to content preference identifiers in this scheme are determined by adjusting historical scene feedback records and historical content feedback records, respectively. This scheme can dynamically optimize the timing and preferred content of subsequent proactive broadcasts based on user feedback on historical broadcasts. The scheme provided in this embodiment can automatically identify the user's personalized preferences based on user identity and proactively provide broadcasting services to the user, achieving the beneficial effect of improving the user's intelligent interaction experience.

[0011] It should be noted that the aforementioned computer instructions may be stored, in whole or in part, on a computer-readable storage medium. This computer-readable storage medium may be packaged together with the processor of the intelligent broadcasting device, or it may be packaged separately from the processor of the intelligent broadcasting device; this application does not impose any limitations on this.

[0012] The descriptions of the second, third, fourth, and fifth aspects in this application can be referenced to the detailed description of the first aspect; and the beneficial effects described in the second, fourth, and fifth aspects can be referenced to the analysis of the beneficial effects of the first aspect, which will not be repeated here.

[0013] It should be understood that the description in this section is not intended to identify key or essential features of the embodiments of this application, nor is it intended to limit the scope of this application. Other features of this application will become readily apparent from the following description.

[0014] It is understood that before using the technical solutions disclosed in the various embodiments of this application, users should be informed of the types, scope of use, and usage scenarios of the personal information involved in this application in an appropriate manner in accordance with relevant laws and regulations, and user authorization should be obtained. Attached Figure Description

[0015] To more clearly illustrate the technical solutions of the embodiments of this application, the accompanying drawings used in the embodiments will be briefly introduced below. It should be understood that the following drawings only show some embodiments of this application and should not be regarded as a limitation of the scope. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort.

[0016] Figure 1 This is a flowchart illustrating the intelligent broadcasting method provided in an embodiment of this application; Figure 2 This is another flowchart illustrating the intelligent broadcasting method provided in this application embodiment; Figure 3 This is a schematic diagram of the intelligent broadcasting device provided in an embodiment of this application; Figure 4 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this application. Detailed Implementation

[0017] To enable those skilled in the art to better understand the present application, the technical solutions of the present application will be clearly and completely described below with reference to the accompanying drawings of the embodiments. Obviously, the described embodiments are only some embodiments of the present application, and not all embodiments. Based on the embodiments of the present application, all other embodiments obtained by those of ordinary skill in the art without creative effort should fall within the scope of protection of the present application.

[0018] It should be noted that the terms "target," "original," etc., used in the specification, claims, and accompanying drawings of this application are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of this application described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0019] The present application will now be described in further detail with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the application and not intended to limit it. Furthermore, it should be noted that, for ease of description, the accompanying drawings show only the parts relevant to the present application, not the entire structure.

[0020] Figure 1This is a flowchart illustrating an intelligent broadcasting method provided in an embodiment of this application. This embodiment is applicable to situations where proactive broadcasting services are provided to users in specific scenarios. The method can be executed by an intelligent broadcasting device, which can be implemented in hardware and / or software and integrated into the electronic device executing the method. Preferably, the electronic device in this embodiment can be a server, or a computer device, etc.

[0021] refer to Figure 1 The intelligent broadcasting method in this embodiment includes, but is not limited to, the following steps: S110. When the current user is determined to be a frequent user based on the received facial image, the current scene data is obtained and at least two scene features are generated.

[0022] It should be noted beforehand that the solution provided in this embodiment can be applied to various scenarios such as home smart terminals, smartphones, smart speakers, wearable devices, smart robots, and in-vehicle systems, etc., and the specific application scenarios are not limited here. To clearly illustrate the specific implementation of this solution, in-vehicle systems will be used as an example in the subsequent illustrative embodiments.

[0023] In the vehicle infotainment system, facial images are obtained through the Driver Monitoring System (DMS) or in-vehicle cameras. Furthermore, biometric information extracted from the facial images, such as facial key points and feature vectors, can be used to verify whether the current user is a frequent user.

[0024] Frequent users are those who have been identified and trained through this scheme. In this embodiment, users whose facial features have been identified and recorded more than a preset number of times (e.g., 5 times) are typically defined on a calendar month basis.

[0025] One implementation method, in this embodiment, is to determine the current user as a frequently used user based on the received facial image as follows: after receiving the vehicle start signal, control the DMS camera to capture the facial image of the driver; extract the facial feature vector on the vehicle terminal and compare it with the locally stored "feature library of frequently used users"; if the match is successful, the current driver is directly identified as a frequently used user; if the match fails, the solution provided in this embodiment is not executed, and the user can be provided with services according to the traditional passive response mode.

[0026] Scene data represents environmental information collected by vehicles or terminal devices at the current moment, including but not limited to: time (time period, precise time point), date type (weekday / holiday), vehicle status data (vehicle speed, driving duration, acceleration, driving mode), location and commuting data (geographical location, driving route, commuting pattern, mileage), environmental data (road condition information), and network environment data (network bandwidth, network latency). Scene features represent quantitative indicators obtained after structuring the scene data. Each feature corresponds to a specific dimension of the environmental state, such as "time period = morning," "location type = home," and "route = frequently used route."

[0027] In this embodiment, each scene feature includes a corresponding feature priority; wherein, the feature priority represents the weight value (such as an integer from 1 to 5) assigned to each scene feature, and the weight value is used to reflect the importance of the feature in determining whether to trigger an active broadcast. The higher the priority, the greater its contribution to the active triggering decision.

[0028] Furthermore, in this embodiment, the feature priority corresponding to each scene feature is determined by adjusting historical scene feedback records. Specifically, whenever an active broadcast is triggered due to a certain scene, the user's choice is recorded: accept the broadcast or refuse the broadcast. For each feature in that scene, during initialization, the feature priority of each scene feature is set to a preset default value; further, its priority is updated according to user feedback: if the user accepts the broadcast, the feature priority of each scene feature in the current scene increases by a first preset step; if the user refuses the broadcast, the feature priority of each scene feature in the current scene decreases by a second preset step. The first and second preset step sizes can be selected as 1, that is, if the user accepts the broadcast, the priority of each feature increases by 1 (upper limit 5); if the user refuses the broadcast, the priority of each feature decreases by 1 (lower limit 1). The specific selection of the first and second preset step sizes is not limited here.

[0029] S120. Based on the feature priority corresponding to each scene feature, when the conditions for active broadcasting are met, obtain the content preference database associated with the current user.

[0030] The conditions for proactive broadcasting are the preset minimum requirements for triggering proactive services. These are typically determined by comparing a weighted average of the current scene's features with a preset threshold. Only when these conditions are met will the subsequent content acquisition and broadcasting process begin.

[0031] The content preference database is a data structure stored on the vehicle's infotainment system or local terminal. It is associated with a specific user identifier (UserIdentifier, or ID) and records the user's interest preferences and their quantified priorities for different content categories (such as finance, technology, and sports). This database is dynamically updated based on user feedback.

[0032] Specifically, determining whether the active broadcast condition is met based on the feature priority corresponding to each scene feature can be achieved by summing the feature priority values ​​of each currently activated scene feature to obtain a weighted composite value; when this composite value is greater than or equal to a preset trigger threshold, the active broadcast condition is determined to be met. For example, the sum of the feature priorities of all current scene features can be calculated; if the sum of the feature priorities is greater than or equal to a preset threshold (e.g., 20), active broadcasting is triggered. Another way to determine whether the active broadcast condition is met is to multiply the feature priority of each scene feature by its base weight and then sum the results; when the weighted integral is greater than the trigger threshold of the current scene, active broadcasting is triggered, and the active broadcast condition is determined to be met.

[0033] The aforementioned content preference database is a data structure stored locally on the vehicle's infotainment system or terminal, associated with a specific user ID, and records the user's interest preferences for different content categories (such as finance, technology, and sports) and their quantitative priorities.

[0034] In this embodiment, the content preference database contains at least two content preference identifiers, each of which includes a corresponding interest priority. The content preference identifier is used to uniquely identify a type of content as a keyword or code, such as "Finance-001" or "Technology-002." Each identifier represents a content domain, facilitating quick retrieval and broadcasting of news or information of the corresponding category.

[0035] Interest priority represents the numerical value (e.g., an integer from 1 to 5) assigned to each content preference identifier, reflecting the user's level of interest in that type of content. The higher the priority, the more frequently that type of content appears in the broadcast queue, and the higher its ranking.

[0036] The interest priority corresponding to each content preference identifier is determined by adjusting historical content feedback records. Specifically, in this embodiment, the interest priority value of the content preference identifier can be determined in the following ways: during initialization, the interest priority value of each content preference identifier is set to the initial value; in response to the user's feedback behavior on the broadcasted content, the interest priority value of the corresponding content preference identifier is adjusted: when the user requests detailed expansion of the content corresponding to a certain content preference identifier, the interest priority value of that content preference identifier increases by the first unit. For example, if the user says "explain in detail" to financial news, then the financial priority increases from 3 to 4, with an upper limit of 5; when the user requests to block a certain content preference identifier, the interest priority value of that content preference identifier decreases by the second unit, or is removed from the content preference database. For example, if the user says "don't report sports anymore," then the sports priority decreases from 3 to 2 (or is deleted), with a lower limit of 1; when the user actively requests broadcasting content in a certain field through voice commands, the interest priority value of the content preference identifier corresponding to that field increases by the third unit. For example, if the user says "broadcast technology news for me," then the technology priority increases from 3 to 4; the interest priority value is limited to a preset minimum value (1) and maximum value (5).

[0037] S130. Broadcast the content information corresponding to each content preference identifier according to the priority of interest.

[0038] In this embodiment, the maximum number of entries for each content preference identifier in the current broadcast is determined based on its interest priority value. For example, priority 1 corresponds to a maximum of 1 entry, priority 2 to a maximum of 2 entries, and so on, with priority 5 corresponding to a maximum of 5 entries. The actual broadcast is capped at the number of currently searchable valid content. Next, for each content preference identifier, information sources (such as news APIs) within a specific timeframe (e.g., within 24 hours) are retrieved using that identifier as a keyword to obtain the corresponding content list. After sorting by time (newest first) or relevance, content not exceeding the aforementioned maximum number of entries is extracted and sequentially inserted into the global broadcast queue. Finally, according to the determined order, the specific news content corresponding to each preference identifier in the queue is broadcast sequentially through the speech synthesis module.

[0039] In one application scenario, after driver Zhang San starts the vehicle, verification confirms that Zhang San is a frequent user. Further data on the current scenario is obtained. Based on historical scene characteristics, the current driving scenario indicates that the user is commuting on a weekday morning at 8:00 AM, driving at a speed of 30 km / h, and has a habit of listening to the morning news in this scenario. The current scenario characteristics determine that the conditions for proactive broadcasting are met. The system then retrieves the content preference database associated with the user. Analysis shows that the user recently follows financial news and sports news, with three financial news items and two sports news items found. At this point, a proactive broadcast can be made: "Good morning, Mr. Zhang. We have prepared the latest three financial news items and two sports news items for you, which will be broadcast as follows…." In this driving scenario, the user does not need to actively wake up the vehicle's infotainment system to receive broadcast content that meets their personal preferences, helping the user concentrate on driving and improving driving safety.

[0040] The intelligent broadcasting method provided in this embodiment firstly, when determining that the current user is a frequent user based on the received facial image, it can generate multiple scene features through the current scene data, and determine whether the conditions for proactive broadcasting are met by combining the feature priorities corresponding to each scene feature. It can intelligently decide whether to broadcast based on the specific scene in which the user is located, avoiding blind interruption or missing broadcasting opportunities, and enhancing environmental awareness and adaptability. Secondly, after determining that the conditions for proactive broadcasting are met, it can broadcast content information corresponding to content preference identifiers according to interest priority order; this realizes personalized content recommendation and proactive broadcasting, making the broadcast content more in line with the user's recent interests, without the need for passive response. Finally, the feature priorities corresponding to scene features and the interest priorities corresponding to content preference identifiers in this solution are determined by adjusting historical scene feedback records and historical content feedback records, respectively. This solution can dynamically optimize the timing and preferred content of subsequent proactive broadcasts based on user feedback on historical broadcasts. The solution provided in this embodiment can automatically identify the user's personalized preferences based on user identity and proactively provide broadcasting services to the user, achieving the beneficial effect of improving the user's intelligent interaction experience.

[0041] Figure 2 This is another flowchart illustrating the intelligent broadcasting method provided in this application embodiment. This application embodiment is an optimization based on the above embodiments. Specifically, the optimization is as follows: This embodiment provides a detailed explanation of the implementation process of "determining the conditions for active broadcasting based on the feature priority corresponding to each scene feature", the implementation process of "determining the conditions for active broadcasting based on the feature priority corresponding to each scene feature", and the implementation process of "determining the current user as a frequently used user based on the received facial image".

[0042] See Figure 2The intelligent broadcasting method provided in this embodiment includes, but is not limited to, the following steps: S210. Obtain the current user's facial image and extract facial features from the facial image.

[0043] In this step, the driver monitoring system (DMS) embedded in the vehicle's infotainment system captures the current driver's facial image using its camera. After preprocessing (such as illumination compensation and pose correction), the captured image is used to extract unique and stable facial feature vectors using algorithms such as Local Binary Pattern (LBP) and Convolutional Neural Network (CNN). These feature vectors can represent the user's identity information. To protect privacy, the original facial image is deleted immediately after feature extraction, retaining only irreversible feature data for subsequent comparison and statistical analysis.

[0044] S211. Obtain the number of times the current facial feature is stored in the feature database.

[0045] In this embodiment, a feature database is maintained to record the facial features of each user and the number of times they have appeared before this identification. After extracting the facial features of the current user in step S210, the feature database is queried using the feature vector as an index. If a record matching the feature already exists in the database, the "Number of Stories" field corresponding to that record is read; if no matching record exists, it is considered the first appearance, and the number of stores is counted as 0. This number essentially reflects the frequency with which the user has been captured and successfully identified within a certain period of time (e.g., a calendar month).

[0046] S212. Determine whether the number of times the current facial feature has been stored in the feature database exceeds the preset number.

[0047] If the current facial feature is stored in the feature database more than the preset number of times, then step S213 is executed.

[0048] If the current facial feature has not been stored in the feature database more than the preset number of times, then proceed to step S260.

[0049] The preset number of times can be 5. This step compares the number of times stored obtained in step S211 with this threshold. If the number of times stored is greater than or equal to the preset number, it is determined to be "exceeded", that is, step S213 is executed; if it is less than the preset number, it is determined to be "not exceeded", that is, no, step S260 is executed.

[0050] S213. Determine the current user as a frequently used user.

[0051] Once a user becomes a frequent user, their facial features and corresponding unique user ID will be permanently stored and will not be periodically deleted. Simultaneously, the system will create or update personalized data for that user, including a content preference database and scene feature priorities, and prioritize these data in subsequent proactive broadcast services. This process achieves seamless, automatic identification of frequent users without requiring manual registration or login.

[0052] S220. Obtain the current scene data and generate at least two scene features.

[0053] Each scene feature includes a corresponding feature priority; the feature priority for each scene feature is determined by adjusting historical scene feedback records.

[0054] The explanation and specific implementation of the current step are the same as the corresponding content in the previous embodiment S110, and will not be repeated here.

[0055] S230. Calculate the weighted composite value based on the feature priority of each scene feature.

[0056] After obtaining multiple features of the current scene and their corresponding feature priorities, this step calculates a quantitative indicator that comprehensively reflects the triggering value of the current scene, namely the weighted composite value. One calculation method is to directly add the feature priorities of all activated scene features. For example, if the current scene has 5 features with priorities of 5, 4, 3, 5, and 2, then the weighted composite value = 5 + 4 + 3 + 5 + 2 = 19.

[0057] S231. When the weighted composite value reaches the preset trigger threshold, it is determined that the active broadcast condition is met.

[0058] A trigger threshold is preset (e.g., 10 or 20). The weighted composite value calculated in step S230 is compared with this threshold: if the weighted composite value is ≥ the preset trigger threshold, it is determined that the active broadcast condition is met, and the process proceeds to the subsequent content preparation stage; if the weighted composite value is < the preset trigger threshold, it is determined that the active broadcast condition is not met, and no active prompts or broadcasts are made in this round, maintaining a silent state to avoid unnecessary disturbance to the user.

[0059] Even when the conditions for proactive announcement are not met, users can still obtain information by actively waking up the assistant and issuing commands, just like with a regular voice assistant. This ensures reduced interaction in environments where users do not wish to be disturbed.

[0060] S240. Obtain the content preference database associated with the current user. The content preference database contains at least two content preference identifiers, and each content preference identifier includes a corresponding interest priority.

[0061] The interest priority corresponding to each content preference identifier is determined by adjusting historical content feedback records.

[0062] The explanation and specific implementation of the current step are the same as the corresponding content in the previous embodiment S120, and will not be repeated here.

[0063] S250. Generate title information based on the content information corresponding to each content preference identifier.

[0064] This step uses the content preference database obtained in S240 to retrieve specific content information for each content preference identifier. For example, for the "Technology" category, it retrieves multiple latest technology news articles published within the last 24 hours; the same applies to the "Finance" category. Then, a brief title (e.g., a summary of no more than 20 words) is generated for each piece of content. The purpose of the title is to allow users to quickly understand the content summary before the broadcast, making it easier for them to decide whether to listen to the details. Each title is associated and stored with its corresponding content preference identifier and complete content details.

[0065] S251. Generate proactive inquiry information according to the priority of interests. The proactive inquiry information includes the title information of the content to be broadcast.

[0066] The title information generated by S250 is further sorted according to interest priority from high to low. For cases with the same interest priority, it can be further sorted according to the publication time of the content corresponding to the title or the time when the user recently viewed it. After sorting, this title information is organized into a proactive query, such as in the form of speech synthesis or screen display: "We found the following news: [Title 1], [Title 2]; ..., would you like to hear them in sequence?" The purpose of the proactive query is to inform the user which content will be read next and to obtain the user's permission.

[0067] S252. After receiving the user's confirmation instruction for the actively requested information, broadcast the content information corresponding to each title information in order of interest priority.

[0068] If a user issues a confirmation command (such as "OK", "Play", "Session Broadcast", etc.), the complete content information corresponding to each title (e.g., expanded news details) will be broadcast one by one according to the pre-arranged interest priority. During the broadcast, the user can still interrupt at any time (to request pause, jump, expand the current content, etc.). If the user issues a rejection command (such as "No need" or "Cancel"), the broadcast will stop, and this rejection will be recorded as feedback for subsequent adjustments to feature priorities or thresholds.

[0069] In a practical use case, each title information includes a corresponding serial number identifier; this embodiment can also implement the following scheme: Receive a redirect instruction from the current user, which includes keyword information or a sequence number identifier; broadcast the corresponding target content information based on the keyword information or the sequence number identifier.

[0070] In the actively requested information, each title can be accompanied by a corresponding serial number. For example, "1. [Title 1], 2. [Title 2]; ...". When a user issues a jump command containing keywords of the target title (such as "Play the news about robots") or a serial number (such as "Play the second one"), this embodiment can directly locate and play the corresponding target content without having to play all items sequentially. In this way, the user does not need to listen to all items and can directly jump to the content of interest, saving time, which is especially suitable for scenarios such as driving where information needs to be obtained quickly.

[0071] Another practical use case involves receiving interruption commands from the current user. These commands include content preference update commands and playback status control commands. Content request change commands are used to request the broadcast of new topics unrelated to the current content queue or to insert queries. For example, if the system provides news in the technology and finance categories, and the user requests the latest entertainment news, these commands will change or expand the content to be broadcast and trigger an update to the preference database.

[0072] Playback status control commands are used to control the playback process, such as pausing, resuming, stopping, switching to the previous content, switching to the next content, repeating the current content, and closing the playback. These commands only change the playback status and do not change the content queue itself.

[0073] Specifically, the system retrieves the corresponding updated content based on the content preference update command and broadcasts the updated content.

[0074] When a user issues a content preference update command, such as "Give me the latest AI news," the system first uses automatic speech recognition and natural language understanding to analyze the semantics of the command and extract new content preference keywords, such as "artificial intelligence." Then, using this keyword as a search criterion, it retrieves timely content (such as news from within the last 24 hours) and generates a brief summary or detailed content. The current broadcast queue is immediately interrupted to prioritize broadcasting the updated content.

[0075] Furthermore, in the current scenario, this embodiment can also implement the following scheme: Based on the content preference update instruction, determine the content preference keywords. If the content preference keywords already exist in the content preference database, increase the interest priority of the content preference identifier corresponding to the content preference keywords. If the content preference keywords do not exist in the content preference database, add the content preference keywords to the content preference database and determine the interest priority of the content preference identifier corresponding to the content preference keywords as the initial value.

[0076] The system extracts keywords representing user interests from the parsed command text. For example, the command "broadcast AI news" extracts the keyword "AI"; "tell me about that financial news story" extracts "finance". If the content preference keyword already exists in the current user's content preference database, the interest priority of the corresponding content preference identifier is increased by one unit, for example, from 3 to 4, with a maximum of 5. This indicates that the user's interest is positively reinforced, and this type of content will receive more broadcasts and a higher ranking in subsequent broadcasts. If the content preference keyword does not exist in the current user's content preference database, the keyword is added as a new content preference identifier to the database and assigned an initial interest priority, for example, a default of 3. This represents the first time the system has learned the user's new interest in this area.

[0077] Furthermore, this embodiment can also execute the following scheme: After the updated content has been broadcast, continue broadcasting the content information in the title that was not broadcast.

[0078] The solution provided in this embodiment needs to save the state of the original broadcast queue, namely the list of broadcast content, the current broadcast progress, and the list of unbroadcast content. Once the updated content has finished broadcasting, the original broadcast queue is automatically restored, and broadcasting resumes from the unbroadcast content in the original order. If the user interrupts the broadcast of the updated content again, the above-described save and restore mechanism is repeated.

[0079] With the above solution, new content requested by users during the broadcast is immediately identified as a preference signal, and the priority of the corresponding keywords is increased or new preferences are added in a timely manner, which can quickly respond to changes in user interests; and during the journey, users can insert new information requests at any time, and can temporarily switch and automatically return, which not only meets dynamic needs, but also does not lose the original broadcast progress.

[0080] S260. Determine that the current user is a regular user.

[0081] Regular users can still use the vehicle's voice commands to perform one-time information queries and broadcasts, but will not be automatically prompted to receive personalized recommendation services that adapt to different scenarios.

[0082] For ordinary users, the solution provided in this embodiment offers a preferred implementation: a) Accept the voice control commands issued by the current user and parse the voice control commands to obtain text information. In this solution, the user's voice signal is captured via a microphone array, and the audio is converted into a text string using an automatic speech recognition model. This text information may contain the user's requested content category (e.g., "Give me today's tech news"), specific queries (e.g., "What's the weather like in Hangzhou today?"), or other control commands (e.g., "Pause playback"). The purpose of this step is to convert the user's spoken request into machine-processable text. Further, a Natural Language Understanding (NLU) module is used to semantically analyze the text information obtained in step A, extracting preference keywords representing the user's content interests. For example, "Technology News" extracts the keyword "technology," and "Financial News" extracts "finance." For statements without explicit category instructions (e.g., "What's new?"), a default strategy (e.g., polling from the default domain) or inference from the context can be used.

[0083] b) Extract preference keywords from the text information and determine the broadcast content based on the preference keywords. After extracting preferred keywords, these keywords are used as search criteria to retrieve timely content (such as news from within the last 24 hours) via Web Service calls or local caching. A large model is then used to generate brief summaries or directly broadcast the relevant content. This process is consistent with the content generation logic for frequently used users, but it does not trigger proactive recommendation services.

[0084] c) Map and store the facial features of the current user, the current driving scene data, and preference keywords to obtain the training dataset for the current user.

[0085] The information extracted in the preceding steps is stored in a temporary training dataset based on mapping relationships. Although the user is not currently identified as a frequent user, once this data is accumulated, if the user's frequency of appearance increases in the future and they become a frequent user, their content preference database and scene feature priorities can be quickly initialized without having to start from scratch.

[0086] In another preferred embodiment, if the number of times the current user's facial features are stored does not exceed a preset number within a preset time period, the training dataset is cleared.

[0087] Within the aforementioned preset time period, each time a user with the same facial features is captured, the "number of times stored" counter for that user will be incremented. Simultaneously, the user's training dataset will be saved. The preset time period can be 30 days or 45 days; the specific choice of preset time period is not limited here.

[0088] When the cycle ends (e.g., the last day of the month), the number of times the user's facial features have been stored is checked: if it exceeds a preset number (e.g., 5 times), the user is upgraded to a frequently used user, and the training dataset is officially migrated to the frequently used user's content preference database and scene feature database, and converted to persistent storage; if it does not exceed the preset number, all training dataset records for that user are automatically deleted, including facial features, scene data, and preference keywords. After data clearing is complete, the user's next appearance will be treated as a completely new user, and counting and database construction will start again.

[0089] Through the above embodiments, ordinary users can enjoy the instant voice broadcast function without increasing the system load due to temporary use; the temporary data of ordinary users will not be retained for a long time, preventing the risk of leakage caused by data accumulation.

[0090] The intelligent broadcasting method provided in this embodiment firstly, when determining that the current user is a frequent user based on the received facial image, it can generate multiple scene features through the current scene data, and determine whether the conditions for proactive broadcasting are met by combining the feature priorities corresponding to each scene feature. It can intelligently decide whether to broadcast based on the specific scene in which the user is located, avoiding blind interruption or missing broadcasting opportunities, and enhancing environmental awareness and adaptability. Secondly, after determining that the conditions for proactive broadcasting are met, it can broadcast content information corresponding to content preference identifiers according to interest priority order; this realizes personalized content recommendation and proactive broadcasting, making the broadcast content more in line with the user's recent interests, without the need for passive response. Finally, the feature priorities corresponding to scene features and the interest priorities corresponding to content preference identifiers in this solution are determined by adjusting historical scene feedback records and historical content feedback records, respectively. This solution can dynamically optimize the timing and preferred content of subsequent proactive broadcasts based on user feedback on historical broadcasts. The solution provided in this embodiment can automatically identify the user's personalized preferences based on user identity and proactively provide broadcasting services to the user, achieving the beneficial effect of improving the user's intelligent interaction experience.

[0091] Figure 3 This is a schematic diagram of a smart broadcasting device provided in an embodiment of this application. This device is suitable for executing the smart broadcasting method provided in an embodiment of this application. Figure 3 As shown, the device may specifically include: a first acquisition module 310, a second acquisition module 320, and a content broadcasting module 330, wherein: The first acquisition module 310 is used to acquire current scene data and generate at least two scene features when determining that the current user is a frequent user based on the received facial image. Each scene feature includes a corresponding feature priority. The feature priority corresponding to each scene feature is determined by adjusting historical scene feedback records. The second acquisition module 320 is used to acquire a content preference database associated with the current user when the active broadcasting condition is met based on the feature priority corresponding to each scene feature. The content preference database contains at least two content preference identifiers, and each content preference identifier includes a corresponding interest priority. The interest priority corresponding to each content preference identifier is determined by adjusting historical content feedback records. The content broadcasting module 330 is used to broadcast the content information corresponding to each content preference identifier according to the interest priority order.

[0092] The intelligent broadcasting device provided in this embodiment firstly, when determining that the current user is a frequent user based on the received facial image, it can generate multiple scene features through the current scene data, and determine whether the conditions for proactive broadcasting are met by combining the feature priorities corresponding to each scene feature. It can intelligently decide whether to broadcast based on the specific scene in which the user is located, avoiding blind interruption or missing broadcast opportunities, and enhancing environmental awareness and adaptive capabilities. Secondly, after determining that the conditions for proactive broadcasting are met, it can broadcast content information corresponding to content preference identifiers according to interest priority order; this realizes personalized content recommendation and proactive broadcasting, making the broadcast content more in line with the user's recent interests, without the need for passive response. Finally, the feature priorities corresponding to scene features and the interest priorities corresponding to content preference identifiers in this solution are determined by adjusting historical scene feedback records and historical content feedback records, respectively. This solution can dynamically optimize the timing and preferred content of subsequent proactive broadcasts based on user feedback on historical broadcasts. The solution provided in this embodiment can automatically identify the user's personalized preferences based on user identity and proactively provide broadcasting services to the user, achieving the beneficial effect of improving the user's intelligent interaction experience.

[0093] In one embodiment, the second acquisition module 320 includes a weighted calculation unit and a condition determination unit, wherein: A weighted calculation unit is used to calculate a weighted comprehensive value based on the feature priority of each scene feature; The condition determination unit is used to determine that the active broadcasting condition is met when the weighted composite value reaches a preset trigger threshold.

[0094] In one embodiment, the second acquisition module 320 further includes an information generation unit and an information broadcasting unit, wherein: The information generation unit is used to generate title information based on the content information corresponding to each content preference identifier; The information generation unit is also used to generate proactive inquiry information according to the interest priority order, wherein the proactive inquiry information includes the title information of the content to be broadcast; The information broadcasting unit is used to broadcast the content information corresponding to each title information in the order of interest priority after receiving the user's confirmation instruction for the actively asked information.

[0095] In one embodiment, each of the title information includes a corresponding serial number identifier; The information broadcasting unit is also used to broadcast the corresponding target content according to the target title information when it receives a jump instruction from the user containing target title information or target sequence number identifier during the process of broadcasting the content information corresponding to each title information in the order of interest priority.

[0096] In one embodiment, the information broadcasting unit is further configured to receive an interruption command issued by the current user, the interruption command including a content preference update command; retrieve corresponding updated content according to the content preference update command, and broadcast the updated content; In one embodiment, the second acquisition module 320 further includes a preference update unit, wherein: The preference update unit is used to determine content preference keywords according to the content preference update instruction. If the content preference keywords already exist in the content preference database, the interest priority of the content preference identifier corresponding to the content preference keywords is increased. If the content preference keywords do not exist in the content preference database, the content preference keywords are added to the content preference database, and the interest priority of the content preference identifier corresponding to the content preference keywords is determined to be the initial value. In one embodiment, the information broadcasting unit is further configured to continue broadcasting the content information in the title information that was not broadcast after the updated content has been broadcast.

[0097] In one embodiment, the first acquisition module 310 includes a feature extraction unit and a user determination unit, wherein: The feature extraction unit is used to acquire the current user's facial image and extract facial features from the facial image; The user determination unit is used to obtain the number of times the current facial feature is stored in the feature database; if the number of times it is stored exceeds a preset number, the current user is determined to be a frequently used user.

[0098] In one embodiment, the first acquisition module 310 further includes an instruction parsing unit, a content determination unit, a mapping storage unit, and a data clearing unit, wherein: The instruction parsing unit is used to accept the voice control instructions issued by the current user and parse the voice control instructions to obtain text information; The content determination unit is used to extract preference keywords from the text information and determine the broadcast content based on the preference keywords; The content determination unit is used to map and store the facial features corresponding to the current user, the current driving scene data, and the preference keywords to obtain the training dataset for the current user.

[0099] The data clearing unit is used to clear the training dataset if the number of times the current user's facial features are stored does not exceed the preset number within a preset time period.

[0100] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the above-described division of functional modules is merely an example. In practical applications, the above functions can be assigned to different functional modules as needed, that is, the internal structure of the device can be divided into different functional modules to complete all or part of the functions described above. The specific working process of the functional modules described above can be referred to the corresponding process in the foregoing method embodiments, and will not be repeated here.

[0101] This application also provides an electronic device, which includes: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores a computer program executable by the at least one processor, and the computer program is executed by the at least one processor to enable the at least one processor to execute the intelligent broadcasting method described in any embodiment of this application.

[0102] This application also provides a computer-readable medium storing computer instructions that, when executed by a processor, implement the intelligent broadcasting method described in any embodiment of this application.

[0103] The following is for reference. Figure 4 , Figure 4 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this application. It illustrates a schematic diagram of the structure of a computer system 500 suitable for implementing the electronic device in the embodiment of this application. Figure 4 The electronic device shown is merely an example and should not impose any limitation on the functionality and scope of use of the embodiments of this application.

[0104] like Figure 4 As shown, the computer system 500 includes a central processing unit (CPU) 501, which can perform various appropriate actions and processes based on programs stored in read-only memory (ROM) 502 or programs loaded from storage section 508 into random access memory (RAM) 503. The RAM 503 also stores various programs and data required for the operation of the system 500. The CPU 501, ROM 502, and RAM 503 are interconnected via a bus 504. An input / output (I / O) interface 505 is also connected to the bus 504.

[0105] The following components are connected to I / O interface 505: an input section 506 including a keyboard, mouse, etc.; an output section 507 including a cathode ray tube (CRT), liquid crystal display (LCD), etc., and speakers, etc.; a storage section 508 including a hard disk, etc.; and a communication section 509 including a network interface card such as a LAN card, modem, etc. The communication section 509 performs communication processing via a network such as the Internet. A drive 510 is also connected to I / O interface 505 as needed. A removable medium 511, such as a disk, optical disk, magneto-optical disk, semiconductor memory, etc., is installed on drive 510 as needed so that computer programs read from it can be installed into storage section 508 as needed.

[0106] Specifically, according to the embodiments disclosed in this application, the processes described above with reference to the flowcharts can be implemented as computer software programs. For example, embodiments disclosed in this application include a computer program product comprising a computer program carried on a computer-readable medium, the computer program containing program code for performing the methods shown in the flowcharts. In such embodiments, the computer program can be downloaded and installed from a network via communication section 509, and / or installed from removable medium 511. When the computer program is executed by central processing unit (CPU) 501, it performs the functions defined above in the system of this application.

[0107] It should be noted that the computer-readable medium shown in this application can be a computer-readable signal medium or a computer-readable storage medium, or any combination of the two. A computer-readable storage medium can be, for example,—but not limited to—an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination thereof. More specific examples of a computer-readable storage medium may include, but are not limited to: an electrical connection having one or more wires, a portable computer disk, a hard disk, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage device, magnetic storage device, or any suitable combination thereof. In this application, a computer-readable storage medium can be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system, apparatus, or device. In this application, a computer-readable signal medium can include a data signal propagated in baseband or as part of a carrier wave, carrying computer-readable program code. Such propagated data signals can take various forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination thereof. Computer-readable signal media can also be any computer-readable medium other than computer-readable storage media, which can send, propagate, or transmit a program for use by or in connection with an instruction execution system, apparatus, or device. The program code contained on the computer-readable medium can be transmitted using any suitable medium, including but not limited to: wireless, wireline, and optical fiber, or any suitable combination thereof.

[0108] The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to various embodiments of this application. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of code containing one or more executable instructions for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions indicated in the blocks may occur in a different order than those indicated in the drawings. For example, two consecutively indicated blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in a block diagram or flowchart, and combinations of blocks in a block diagram or flowchart, may be implemented using a dedicated hardware-based system that performs the specified function or operation, or using a combination of dedicated hardware and computer instructions.

[0109] The modules and / or units described in the embodiments of this application can be implemented in software or hardware. The described modules and / or units can also be housed in a processor; for example, a processor can be described as including a first acquisition module, a second acquisition module, and a content playback module. The names of these modules do not necessarily limit the functionality of the module itself.

[0110] In another aspect, this application also provides a computer-readable medium, which may be included in the device described in the above embodiments; or it may exist independently and not assembled into the device. The computer-readable medium carries one or more programs, which, when executed by the device, cause the device to: when determining that the current user is a frequent user based on a received facial image, acquire current scene data and generate at least two scene features, each scene feature including a corresponding feature priority; wherein the feature priority corresponding to each scene feature is determined by adjusting historical scene feedback records; when determining that an active broadcast condition is met according to the feature priority corresponding to each scene feature, acquire a content preference database associated with the current user, the content preference database containing at least two content preference identifiers, each content preference identifier including a corresponding interest priority; wherein the interest priority corresponding to each content preference identifier is determined by adjusting historical content feedback records; and broadcast content information corresponding to each content preference identifier in order of interest priority.

[0111] According to the technical solution of this embodiment, firstly, when the current user is determined to be a frequent user based on the received facial image, multiple scene features can be generated through the current scene data. The feature priority corresponding to each scene feature is then combined to determine whether the conditions for proactive broadcasting are met. This method can intelligently decide whether to broadcast based on the specific scene in which the user is located, avoiding blind interruption or missing broadcast opportunities, and enhancing environmental awareness and adaptability. Secondly, after determining that the conditions for proactive broadcasting are met, content information corresponding to the content preference identifier can be broadcast according to the interest priority order. This method realizes personalized content recommendation and proactive broadcasting, making the broadcast content more in line with the user's recent interests, without the need for passive response. Finally, the feature priority corresponding to the scene features and the interest priority corresponding to the content preference identifier in this solution are determined by adjusting historical scene feedback records and historical content feedback records, respectively. This indicates that this solution has continuous learning capabilities and can dynamically optimize the timing and preferred content of subsequent proactive broadcasts based on user feedback on historical broadcasts. The solution provided in this embodiment can automatically identify the user's personalized preferences based on user identity and proactively provide broadcast services to the user, achieving the beneficial effect of improving the user's intelligent interaction experience.

[0112] The specific embodiments described above do not constitute a limitation on the scope of protection of this application. Those skilled in the art should understand that various modifications, combinations, sub-combinations, and substitutions can occur depending on design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of this application should be included within the scope of protection of this application.

Claims

1. An intelligent broadcasting method, characterized in that, include: When the current user is determined to be a frequent user based on the received facial image, the current scene data is obtained and at least two scene features are generated, each scene feature including a corresponding feature priority; wherein, the feature priority corresponding to each scene feature is determined by adjusting historical scene feedback records; When the active broadcasting condition is met based on the feature priority corresponding to each of the scene features, a content preference database associated with the current user is obtained. The content preference database contains at least two content preference identifiers, and each content preference identifier includes a corresponding interest priority. The interest priority corresponding to each content preference identifier is determined by adjusting historical content feedback records. The content information corresponding to each content preference identifier is broadcast in order of interest priority.

2. The intelligent broadcasting method according to claim 1, characterized in that, The step of determining whether the active broadcasting conditions are met based on the feature priority corresponding to each scene feature includes: Calculate a weighted composite value based on the feature priority of each scene feature; When the weighted composite value reaches the preset trigger threshold, it is determined that the active broadcast condition is met.

3. The intelligent broadcasting method according to claim 1, characterized in that, The step of broadcasting the content information corresponding to each content preference identifier according to the interest priority order includes: Title information is generated based on the content information corresponding to each content preference identifier; Active inquiry information is generated according to the interest priority order, and the active inquiry information includes the title information of the content to be broadcast; Upon receiving confirmation from the user regarding the actively requested information, the content information corresponding to each title is broadcast in order of interest priority.

4. The intelligent broadcasting method according to claim 3, characterized in that, Each of the title information includes a corresponding serial number identifier; In the process of broadcasting the content information corresponding to each title information in the order of interest priority, the method further includes: Receive a redirection command issued by the current user, the redirection command including keyword information or sequence number identifier; The corresponding target content information is broadcast based on the keyword information or the serial number identifier.

5. The intelligent broadcasting method according to claim 3, characterized in that, In the process of broadcasting the content information corresponding to each title information in the order of interest priority, the method further includes: Receive interruption instructions from the current user, including content preference update instructions; Retrieve the corresponding updated content according to the content preference update instruction, and broadcast the updated content; The method further includes: The content preference keywords are determined according to the content preference update instruction. If the content preference keywords already exist in the content preference database, the interest priority of the content preference identifier corresponding to the content preference keywords is increased. If the content preference keyword does not exist in the content preference database, the content preference keyword is added to the content preference database, and the interest priority of the content preference identifier corresponding to the content preference keyword is determined as the initial value. Accordingly, the method further includes: After the updated content has been broadcast, the content information in the title information that was not broadcast will continue to be broadcast.

6. The intelligent broadcasting method according to claim 1, characterized in that, Determining the current user as a frequent user based on the received facial image includes: Obtain the current user's facial image and extract facial features from the facial image; Get the number of times the current facial feature is stored in the feature database; If the number of times a user is stored exceeds a preset number, the current user is determined to be a frequently used user.

7. The intelligent broadcasting method according to claim 6, characterized in that, If the number of times the data is stored does not exceed the preset number, the method further includes: Accept the voice control command issued by the current user, and parse the voice control command to obtain text information; Extract preference keywords from the text information, and determine the broadcast content based on the preference keywords; The facial features of the current user, the current driving scene data, and the preference keywords are mapped and stored to obtain the training dataset for the current user; Accordingly, the method further includes: If the number of times the current user's facial features are stored does not exceed the preset number within a preset time period, the training dataset is cleared.

8. An intelligent broadcasting device, characterized in that, include: The first acquisition module is used to acquire current scene data and generate at least two scene features when determining that the current user is a frequent user based on the received facial image. Each scene feature includes a corresponding feature priority. The feature priority corresponding to each scene feature is determined by adjusting historical scene feedback records. The second acquisition module is used to acquire a content preference database associated with the current user when the active broadcasting conditions are met based on the feature priority corresponding to each scene feature. The content preference database contains at least two content preference identifiers, and each content preference identifier includes a corresponding interest priority. The interest priority corresponding to each content preference identifier is determined by adjusting historical content feedback records. The content broadcasting module is used to broadcast the content information corresponding to each content preference identifier in order of interest priority.

9. An electronic device, characterized in that, The electronic device includes: At least one processor; and a memory communicatively connected to said at least one processor; The memory stores a computer program that can be executed by the at least one processor, which is then executed by the at least one processor to enable the at least one processor to perform the intelligent broadcasting method according to any one of claims 1-7.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the program is executed by the processor, it implements the intelligent broadcasting method as described in any one of claims 1-7.

Images