Voice broadcast control method and device, wearable equipment and storage medium

Abstract

The invention provides a voice broadcast control method and device, wearable equipment and a storage medium, and belongs to the technical field of voice processing, and the method comprises the steps: responding to a received first instruction, and determining initial broadcast content based on first information and second information; the first information comprises voice information of the user, and the second information comprises preference information of the user; in response to the received second instruction, determining an initial broadcast content based on the second information; the content amount of the initial broadcast content is adjusted based on third information, target broadcast content is obtained, and the third information comprises behavior information of the user; and broadcasting the target broadcast content. According to the voice broadcast control method and device, the wearable device and the storage medium provided by the invention, the personalization and intelligence of voice broadcast control can be improved.

Description

Technical Field

[0001] The present disclosure belongs to the field of voice processing technology, and more specifically, relates to a voice broadcast control method and device, a wearable device, and a storage medium. Background Art

[0002] With the continuous development of science and technology, voice broadcasting has been widely used in various fields, such as smart devices, navigation systems, information prompts, etc. However, existing voice broadcasting methods often have some shortcomings, such as the inability to flexibly control according to the specific needs of users, which affects the personalization and intelligence of voice broadcasting.

[0003] Therefore, a personalized and intelligent voice broadcast control method is needed. Summary of the invention

[0004] The purpose of the present disclosure is to provide a voice broadcast control method and device, a wearable device, and a storage medium to improve the personalization and intelligence of voice broadcast control.

[0005] According to a first aspect of an embodiment of the present disclosure, a voice broadcast control method is provided, which is applied to a first device and includes: In response to receiving the first instruction, determining the initial broadcast content based on first information and second information; the first information includes the user's voice information, and the second information includes the user's preference information; In response to receiving the second instruction, determining initial broadcast content based on the second information; Adjusting the content volume of the initial broadcast content based on third information to obtain target broadcast content, wherein the third information includes user behavior information; The target broadcast content is broadcasted.

[0006] According to a second aspect of the embodiments of the present disclosure, a voice broadcast control device is provided, which is applied to a first device and includes: A first broadcast content determination module, configured to determine initial broadcast content based on first information and second information in response to receiving a first instruction; the first information includes user voice information, and the second information includes user preference information; A second broadcast content determination module, in response to receiving the second instruction, determines the initial broadcast content based on the second information; an adjustment module, which adjusts the content volume of the initial broadcast content based on third information to obtain target broadcast content, wherein the third information includes user behavior information; The broadcast module is used to broadcast the target broadcast content.

[0007] According to a third aspect of an embodiment of the present disclosure, a wearable device is provided, comprising a memory, a processor, and a computer program stored in the memory and running on the processor, wherein the processor implements the steps of the above-mentioned voice broadcast control method when executing the computer program.

[0008] According to a fourth aspect of the embodiments of the present disclosure, a computer-readable storage medium is provided, wherein the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the steps of the above-mentioned voice broadcast control method are implemented.

[0009] The beneficial effects of the voice broadcast control method and device, wearable device, and storage medium provided by the embodiments of the present disclosure are: The present disclosure determines the initial broadcast content by integrating user voice information (first information) and preference information (second information), and can accurately locate user needs. Whether it is based on the detailed demand expression of the first instruction or the quick trigger of the second instruction, it can quickly provide content that meets the user's preferences, greatly improving the pertinence and efficiency of information acquisition. Secondly, the user behavior information is introduced to adjust the initial broadcast content to obtain the target broadcast content, so that the broadcast can be dynamically optimized according to the user's real-time feedback. This dynamic adaptability can effectively avoid information redundancy and mismatch, and continuously enhance the user experience. Furthermore, the overall process has a high degree of automation, from command reception, content determination to broadcasting, which is done in one go, reducing user manual operations, and is particularly suitable for wearable devices such as smart glasses or convenient interactive devices such as smart speakers, allowing users to conveniently and efficiently obtain personalized voice information services in multiple scenarios. Therefore, the present disclosure can improve the personalization and intelligence of voice broadcast control. BRIEF DESCRIPTION OF THE DRAWINGS

[0010] In order to more clearly illustrate the technical solutions in the embodiments of the present disclosure, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present disclosure. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0011] Figure 1 A flowchart of a voice broadcast control method provided by an embodiment of the present disclosure; Figure 2 A structural block diagram of a voice broadcast control device provided in an embodiment of the present disclosure; Figure 3 A schematic block diagram of a wearable device provided in one embodiment of the present disclosure. DETAILED DESCRIPTION

[0012] In the following description, specific details such as specific system structures and technologies are provided for the purpose of illustration rather than limitation, so as to provide a thorough understanding of the embodiments of the present disclosure. However, it should be clear to those skilled in the art that the present disclosure may be implemented in other embodiments without these specific details. In other cases, detailed descriptions of well-known systems, devices, circuits, and methods are omitted to avoid obstructing the description of the present disclosure with unnecessary details.

[0013] In order to make the purpose, technical solutions and advantages of the present disclosure more clear, specific embodiments will be described below in conjunction with the accompanying drawings.

[0014] Please refer to Figure 1 , Figure 1 This is a flow chart of a voice broadcast control method provided in an embodiment of the present disclosure. The method is applied to a first device and may include S101 to S104.

[0015] S101: In response to receiving a first instruction, determining initial broadcast content based on first information and second information; the first information includes user voice information, and the second information includes user preference information.

[0016] In this embodiment, the first device may be a device that can interact with the user in real time and without obstacles. The first device can quickly receive the user's command information and immediately feedback the corresponding content to the user, so as to achieve efficient information exchange. For example, the first device may be a smart wearable device (such as smart glasses), a smart phone, and a smart speaker. The user is the user of the first device.

[0017] The first instruction may be a voice instruction (such as a task instruction or a query request) issued by the user to the first device, which means that the user has a voice input. For example, the first instruction may be "I want to know the latest technology news".

[0018] The first information is the user's voice information, that is, the voice content that the user can directly say to the first device, which is also obtained from the first instruction, covering various demand expressions, question asking, operation commands, etc., such as "check the weather in Beijing tomorrow."

[0019] The second information is the user's preference information, which can be a collection of information about the user's personal preferences, habits, settings, etc., or it can be content pre-set by the user in the first device, such as favorite music style (rock, classical, etc.), news category (sports, entertainment, etc.), voice broadcast speed (fast, normal, slow), timbre (gentle female voice, calm male voice, etc.), or it can be the preference tendency analyzed and summarized through the user's past interaction behavior with the first device.

[0020] Determining initial broadcast content based on the first information and the second information includes: extracting a first feature set of first information; extracting a second feature set of second information; The first feature set and the second feature set are input into the first model to obtain initial broadcast content.

[0021] Among them, the first model is a model trained based on the user's first historical information, second historical information and corresponding historical initial broadcast content.

[0022] For example, in the stage of extracting the first feature set of the first information, natural language processing technology can be used to accurately understand the semantics, intentions and key features of the user's voice information. The text after speech transcription is divided into words through lexical analysis, and part-of-speech tagging is performed to determine the grammatical role of each word. Then, syntactic analysis is used to construct the grammatical structure of the sentence, and then key features such as "soothing music" and "science and technology news" are accurately extracted. At the same time, the text content in the user preference information is also processed in the same way to better obtain the initial broadcast content.

[0023] The first model may be a machine learning algorithm to establish the relationship between the first feature set, the second feature set and the initial broadcast content. For example, a neural network model in deep learning may be used to learn the mapping relationship between different user voice features, preference features and broadcast content through a large amount of training data, so as to accurately output the initial broadcast content when facing a new user request.

[0024] The initial broadcast content is the preliminarily determined content that needs to be broadcast, and it will need to be adjusted later to determine the final broadcast content for broadcast.

[0025] When the first device receives the first command issued by the user, it will immediately parse the user's voice information (first information), extract the key required content, and combine the user preference information (second information) stored or analyzed by the first device to preliminarily determine the content to be voice broadcast (initial broadcast content).

[0026] For example, when a user says "I want to listen to some relaxing music" (first instruction) to the smart glasses (first device), the user's voice information (first information) can be obtained from the instruction. The first device will analyze the user's voice information and determine that the user wants to listen to music in a relaxing style. Then, the first device will check the user's preference information (second information). If it is found that the user often listens to songs by singers A, B, etc., and prefers the lyrical category in pop music, then the first device will determine that the initial broadcast content can be a list of some relaxing and lyrical songs by singers A or B, and prepare to perform voice broadcast.

[0027] For another example, when a user says "I want to listen to the commentary of the Four Great Classics" (first instruction) to a smartphone (first device), the user's voice information (first information) can be obtained from the instruction. The first device analyzes the first information and learns that the user's demand is to play the commentary of the Four Great Classics. Then, the user's preference information (second information) is checked to see that it is a normal speaking speed and a steady male voice. The initial broadcast content can be content related to "the commentary of the Four Great Classics with a normal speaking speed and a steady male voice".

[0028] S102: In response to receiving the second instruction, determine initial broadcast content based on second information.

[0029] In this embodiment, the second instruction is different from the first instruction. The second instruction can be a broadcast instruction issued by the user to the first device (such as a quick trigger instruction based on the user's pre-set or recognized by the first device). Therefore, when determining the initial broadcast content, only the second information (user's preference information) needs to be considered. At this time, the main focus is on the various preferences, habitual settings, etc. shown by the user during the use of the first device.

[0030] When the first device (such as smart glasses) receives the second instruction, it directly determines the initial content to be broadcast based on the user preference information it has obtained.

[0031] For example, for the smart glasses (first device), if the user sets "My Music" as the second instruction. When the smart glasses receive "My Music" (second instruction), the first device will check the user's preference information. If it is found that the user's favorite music type is classical music, the initial broadcast content determined by the smart glasses may be "Start your classical music journey", and then the voice broadcast can be prepared and the corresponding music playback operation can be executed.

[0032] For another example, if the second instruction is "Today's Briefing", the smart glasses (first device) will provide the user with a briefing of information that meets their preferences based on the user's preference information for news areas. If the user prefers financial and technological news, the initial broadcast content may be "Here are today's financial news: [Specific financial news title 1], [Specific financial news title 2]; Technological news: [Specific technological news title 1], [Specific technological news title 2]", thereby quickly providing the user with a summary of information that meets their preferences.

[0033] S103: adjusting the content volume of the initial broadcast content based on the third information to obtain the target broadcast content, wherein the third information includes the user's behavior information.

[0034] In this embodiment, the third information (user behavior information) refers to various behavior actions and operation data displayed by the user during the interaction with the first device. These behaviors can be operations directly targeting the voice broadcast of the first device, such as pausing, fast-forwarding, and skipping certain content during the voice broadcast; they can also be behaviors during the use of other functions of the first device, such as frequent searches for certain keywords, length of time spent on a specific page, clicks on or ignores certain recommended content, etc. For example, when using the music playback function, the user often skips a certain singer's songs, and this behavior will be recorded as behavior information; or when browsing news information, the user always reads a certain type of news article for a long time, which also constitutes part of the behavior information.

[0035] The first device will optimize, modify or supplement the previously determined initial broadcast content based on the collected user behavior information (third information), so as to generate the final target broadcast content to better meet the user's actual needs and preference changes.

[0036] The initial broadcast content is adjusted based on the third information to obtain the target broadcast content, including: Establishing screening rules based on third party information; If the first line of information in the third information is greater than or equal to the preset fast-forward times, simplifying the initial broadcast content based on the content simplification algorithm to obtain the target broadcast content; If the first line of information in the third information is less than the preset fast-forward times, the initial broadcast content is expanded based on the content expansion algorithm to obtain the target broadcast content; If the second behavior information in the third information is less than the preset number of times of viewing details, the initial broadcast content is simplified based on the content simplification algorithm to obtain the target broadcast content; If the first behavior information in the third information is greater than or equal to the preset number of times of viewing details, the initial broadcast content is expanded based on the content expansion algorithm to obtain the target broadcast content.

[0037] For example, taking smart glasses (the first device) as an example, if the initial broadcast content is determined as "recommending some popular movie commentaries" based on the user's voice command and user preference information (such as liking science fiction and action movies), "recommending popular science fiction and action movies such as "C" and "D" for you". During the voice broadcast process, the smart glasses monitor the user's behavior information: when "C" is mentioned, the user pauses the broadcast and views more detailed information about "C" through the touch operation of the glasses, and stays there for a long time; and when "D" is mentioned, the user quickly skips the relevant introduction. Based on this behavioral information, the smart glasses will adjust the initial broadcast content, and the target broadcast content may become "C is a highly acclaimed science fiction movie with wonderful special effects and profound plot. Many viewers have given it high praise. Next, explain in detail..."

[0038] In this embodiment, the third information may be user behavior information; In response to the third information being a first behavior, a first adjustment is made to the initial broadcast content to obtain target broadcast content; In response to the third information being a second behavior, performing a second adjustment on the initial broadcast content to obtain target broadcast content; The first behavior and the second behavior have different behavior modes, and the first adjustment and the second adjustment are handled differently.

[0039] The first behavior may include frequently searching for certain keywords, repeatedly broadcasting a certain broadcast content, clicking on certain recommended content, or broadcasting slowly; The second behavior may include pausing, fast-forwarding, skipping certain content, or ignoring certain recommended content during the voice broadcast.

[0040] The first adjustment may be to expand the initial broadcast content, and the second adjustment may be to reduce the initial broadcast content.

[0041] In this embodiment, the first behavior is a set of user behaviors that indicate a positive and in-depth exploration of the current broadcast content. For example, frequently searching for certain keywords indicates that the user is actively seeking more information on the same topic; repeatedly broadcasting a certain broadcast content indicates that the user wants to further understand and strengthen the memory of the content or is very interested in the content; clicking on certain recommended content indicates that the user is concerned about and interested in specific recommended information; and slow broadcasting indicates that the user wants to listen to the broadcast content of this type in detail.

[0042] The second behavior reflects that the user is dissatisfied with the current broadcast content, is not interested in it, or wants to skip some of the content. Pausing can be when the user needs to temporarily interrupt the broadcast to handle other matters or think; fast forwarding is to skip the part that is already known or not interested in; skipping certain content and ignoring certain recommended content clearly indicate that the user does not want to receive these specific broadcast information.

[0043] The first adjustment is to add more detailed information, relevant supplementary materials, interpretations from different angles, etc. on the basis of the initial broadcast content after judging from the first behavior that users have more needs for related content, so as to make the broadcast content richer and more comprehensive.

[0044] The second adjustment determines, based on the second behavior, that the user is not interested in some content or wants to quickly skim through it, and simplifies the initial broadcast content to remove redundant, minor or user-unnoticed parts, making the broadcast more compact and efficient.

[0045] The first device will continuously monitor the user's behavioral information (third information) during use, and dynamically adjust the amount of information of the initial broadcast content that has been determined based on these behaviors, and ultimately generate target broadcast content that is more in line with user needs.

[0046] For example, when a user is wearing smart glasses (the first device) and reporting on science and technology news, he frequently searches for the keyword "artificial intelligence" and repeatedly reports on a report about a new breakthrough in artificial intelligence. After the smart glasses detect this first behavior, they will expand the initial report content of the news. Originally, it simply reported the results of the breakthrough in artificial intelligence. After the expansion, it will add the research and development background of the technology, the introduction of the main research and development team, and the prediction of future application prospects, so as to obtain richer target broadcast content to meet the user's needs for in-depth understanding of artificial intelligence information.

[0047] For another example, when the smart glasses play music recommendation information for the user, the user skips the introduction of some songs and fast-forwards the recommendation of some music styles during the voice broadcast. After the smart glasses recognize these secondary behaviors, they will reduce the initial broadcast content. Originally, there may be a detailed introduction of multiple music styles and many songs. After the reduction, it may only focus on one or two music styles that the user has not skipped and may be interested in, as well as a few popular songs, making the broadcast content more refined, avoiding wasting the user's time, and quickly conveying the core music information that the user may be interested in to the user.

[0048] This embodiment adjusts the amount of content of the initial broadcast content according to the user's behavioral information, so that the target broadcast content can be highly adapted to changes in user needs and enhance the accuracy and flexibility of interaction. Whether it is information acquisition or entertainment experience, it can reduce the user's energy consumption in screening information, optimize the use experience of smart devices, and improve user satisfaction and loyalty.

[0049] S104: Broadcast the target broadcast content.

[0050] In this embodiment, the text content of the target broadcast content is input into the speech synthesis model to obtain the speech signal of the target broadcast content; The voice signal is played through the broadcasting device of the first device.

[0051] Broadcasting refers to the process of converting prepared text information into voice information through the voice output device of the first device (such as the speaker of smart glasses, the earpiece or external speaker of a smart phone, etc.) and conveying it to the user in the form of audible sound.

[0052] The final voice information (target broadcast content) determined and adjusted in the previous steps is output to the user in a suitable voice form. The first device will use the built-in speech synthesis technology to convert the text in the target broadcast content into natural and fluent voice, and set the appropriate volume, speech speed, timbre and other parameters for playback based on factors such as the user's preferences and the current usage environment of the first device.

[0053] For example, for smart glasses, if the target broadcast content is "There are three highly rated restaurants near your location, namely the 'La Meizi Restaurant' specializing in Sichuan cuisine, located 500 meters ahead; the 'Delicious Western Restaurant' serving Western food, 300 meters on the right street; and the 'Haiweixian Restaurant' serving seafood, 800 meters away." The smart glasses will use their speech synthesis engine to convert this text into speech. If the user has previously set a preference for a gentle female voice, a moderate speaking speed, and automatic adjustment of the volume according to ambient noise (for example, the current environment is quiet and the volume is moderate), then the smart glasses will use such voice settings to broadcast information about nearby restaurants to the user, so that the user can obtain useful information by listening to the voice without looking at the screen, which facilitates their decision-making, such as choosing a restaurant to dine at.

[0054] It can be concluded from the above that the present disclosure determines the initial broadcast content by integrating user voice information (first information) and preference information (second information), and can accurately locate user needs. Whether it is based on the detailed demand expression of the first instruction or the quick trigger of the second instruction, it can quickly provide content that meets the user's preferences, greatly improving the pertinence and efficiency of information acquisition. Secondly, the introduction of user behavior information adjusts the initial broadcast content to obtain the target broadcast content, so that the broadcast can be dynamically optimized according to the user's real-time feedback. This dynamic adaptability can effectively avoid information redundancy and mismatch, and continuously enhance the user experience. Furthermore, the overall process has a high degree of automation, from command reception, content determination to broadcasting, which is done in one go, reducing user manual operations, and is particularly suitable for wearable devices such as smart glasses or convenient interactive devices such as smart speakers, allowing users to conveniently and efficiently obtain personalized voice information services in multiple scenarios. Therefore, the present disclosure can improve the personalization and intelligence of voice broadcast control.

[0055] In one embodiment of the present disclosure, determining the initial broadcast content based on the first information and the second information includes: determining a plurality of speech features of the first information; determining a plurality of preference characteristics of the second information; fusing the plurality of speech features and the plurality of preference features to obtain at least one first feature; Initial broadcast content is determined based on at least one first feature.

[0056] In this embodiment, determining multiple voice features of the first information includes: Inputting the first information into the acoustic feature model to obtain a plurality of first speech features; Inputting the first information into a semantic extraction model to obtain a plurality of second speech features; A plurality of speech features are determined based on the plurality of first speech features and the plurality of second speech features.

[0057] The plurality of first speech features may include fundamental frequency (reflecting the pitch), intensity (the strength of the sound), timbre (the characteristics of the sound) and duration; The plurality of second voice features may include keywords, topics, and intentions.

[0058] Determine multiple preference characteristics of the second information, including: Based on the analysis of the user's option settings and configuration information on the first device, multiple preference features of the second information are obtained.

[0059] Voice features refer to representative and analyzable feature information extracted from the user's voice information (first information), including but not limited to voice pitch (e.g., high pitch may indicate emphasis or excitement, low pitch may indicate calmness or seriousness), speech speed (fast speed may imply anxiety or habit of fast-paced communication, slow speed may indicate thinking or emphasis), timbre (different people's voices have unique timbre, which can be used to identify identity or be associated with specific emotions), frequency of occurrence of specific words or phrases (certain keywords can reflect the theme or intention), etc.

[0060] Preference features are key elements extracted from the user's preference information (secondary information). For example, in music, preference features can be music style (rock, classical, pop, etc.), singer or band preference, song era preference; in news, it can be news category (current affairs, sports, entertainment, etc.), news source preference; in voice broadcast style, there are speech speed preferences (fast, medium, slow), timbre preferences (male, female, children's voice, etc.), etc.

[0061] Determining the multiple voice features of the first information is to conduct an in-depth analysis of the voice information spoken by the user. For example, when the user says to the first device "I want to listen to some cheerful songs to relax", the first device will analyze the voice information and obtain multiple voice features: it is detected that the user's speech speed is moderate, the timbre is relatively soft, and keywords such as "cheerful" and "relaxation" appear frequently.

[0062] Determine multiple preference features of the second information: Assuming that the user has pre-set a preference for pop music in the first device, and likes a reporting style with a slightly faster speech speed and a gentle female voice, then the attributes related to these pop music styles, specific singer preferences, and voice reporting styles are the preference features.

[0063] In this embodiment, multiple voice features and multiple preference features are fused to obtain at least one first feature, including: Based on the target fusion strategy, multiple speech features and multiple preference features are fused to obtain at least one first feature; If the correlation between the plurality of speech features and the plurality of preference features is greater than or equal to a first preset value, the first fusion strategy is used as a target fusion strategy; If the correlation between the plurality of speech features and the plurality of preference features is less than the first preset value, the second fusion strategy is used as the target fusion strategy; The first fusion strategy and the second fusion strategy have different processing methods, and the first preset value can be dynamically adjusted according to actual conditions.

[0064] The multiple voice features may include a first pitch feature, a first speaking speed feature, a first timbre feature, and a first keyword feature; the multiple preference features may include a second style feature, a second interest feature, a second speaking speed feature, and a second timbre feature.

[0065] The calculation formula of correlation can be:

[0066] in, Indicates the relevance, Indicates the total number of categories of speech features, represents the total number of categories of preference features, Indicates The value of the speech feature after normalization. Indicates The value of a preference feature after normalization.

[0067] Indicates The normalization method can be different according to the type of speech feature. For example, for the frequency of keyword occurrence, the number of times the keyword occurs in the speech information can be divided by the total number of words in the speech information to get the proportional value. For speaking speed, it can be normalized according to a pre-set speaking speed range (such as slow, medium, and fast correspond to 0-0.3, 0.3-0.7, 0.7-1, etc.).

[0068] Indicates The value of a preference feature after normalization. For example, for music style preference, if a certain style is preferred, the corresponding value is 1, otherwise it is 0; the speech speed preference can also be normalized similarly.

[0069] When the calculated correlation is greater than or equal to the first preset value, the first fusion strategy is adopted, which can focus more on emphasizing the impact of voice features on the initial broadcast content, such as taking the keywords and intentions in the voice as the core, and fine-tuning in combination with preference features. For example, if the voice mentions "the latest technology news" and the user prefers fast broadcast, the initial broadcast content may first focus on the latest technology news searched, and then broadcast at a faster speed.

[0070] When the correlation is less than the first preset value, the second fusion strategy is adopted, which can pay more attention to the dominant role of the preference feature and use the voice feature as a supplement or screening condition. For example, if the user prefers classical music, and the voice command is vague and just says "play music", then classical music is used as the main direction, combined with other possible information in the voice (such as the emotional tone of the voice, if it is a soothing tone, softer tracks in classical music may be selected) to determine the initial broadcast content.

[0071] The initial broadcast content is determined based on the first feature obtained by fusion. Taking the first feature of "cheerful pop music" as an example, the initial broadcast content can be "recommending cheerful pop songs for you, hoping to make you happy." Then the first device can make subsequent broadcasts or further processes based on this initial broadcast content, such as preparing to play song resources or displaying song-related information.

[0072] The user says to the smart car system (the first device), "I want to listen to something inspirational to motivate myself on the road." The first device analyzes the voice features and finds that the user's speaking speed is normal, the tone is firm, and the word "inspirational" is prominent. Combined with the user's preference information, such as preference for audiobooks, frequent listening to celebrity biography audiobooks, and preference for calm male voice broadcasts. The voice features and preference features are combined to obtain the first feature of "Inspirational celebrity biography audiobook recommendation." Based on this, the initial broadcast content can be "Recommend the audiobook "XXX" for you, a calm male voice tells you about Qiao's inspirational life and accompanies your journey."

[0073] From the above, it can be concluded that this embodiment accurately analyzes voice features and can deeply understand the user's immediate needs and emotional tendencies, such as sensing the user's urgency for information from the anxious speaking speed. Mining preference features can fully grasp the user's long-term preferences. The fusion of the two as the first feature can build a highly personalized starting point for broadcasting. This makes the initial broadcast content closely fit the user and reduces irrelevant information push. Whether it is entertainment resource recommendations or information query broadcasts, they can directly hit the user's interests, improve the accuracy and effectiveness of information delivery, enhance the satisfaction and fluency of user-device interaction, and make voice broadcast services more intelligent and intimate.

[0074] In one embodiment of the present disclosure, determining the initial broadcast content based on the second information includes: determining a plurality of preference characteristics of the second information; Screening multiple preferred features to obtain at least one second feature; The initial broadcast content is determined based on at least one second feature.

[0075] In this embodiment, the preference feature refers to the key elements extracted from the second information (user's preference information). For example, when the first device is a pair of smart glasses, it may include preferences for voice interaction, such as voice timbre, speech speed, and volume; and functional usage preferences, such as topics of interest when receiving information push (such as technology news, sports event results, etc.).

[0076] The plurality of preference features are screened to obtain at least one second feature, including: Determine the user's selection tendency based on the user information; Matching and screening the user's selection tendency with multiple preference features to obtain at least one second feature; The user information may include the scenario in which the user is located, the user's activity level over a period of time, or a set priority rule.

[0077] Screening refers to selecting more relevant and important features as the second feature from multiple identified preferred features based on preset rules, priorities, or current situational needs.

[0078] When the smart glasses receive the second instruction, they will filter these preference features according to the type of instruction and the current context. For example, when the second instruction is "show news", then among the many preference features, the news-related preference feature of "strong interest in science and technology news" will be selected as the second feature.

[0079] Taking the second instruction of "display news" as an example, based on the second feature of "strong interest in science and technology news", the initial broadcast content can be "pushing the latest science and technology news headlines for you, including [specific science and technology news title 1], [specific science and technology news title 2], etc."

[0080] From the above, it can be concluded that the present embodiment determines multiple preference features that can comprehensively cover the user's preference tendencies in different functional and experience dimensions; by screening to obtain the second feature, the key preferences can be highlighted in a specific instruction context to avoid information confusion and redundancy. For example, smart glasses filter out key preferences based on instructions to determine the initial broadcast content; when pushing news, interest preferences are locked to accurately provide information of interest. This greatly improves the pertinence and effectiveness of voice broadcasts, enhances the fit and satisfaction of user interaction with smart devices, and optimizes the overall user experience.

[0081] In one embodiment of the present disclosure, reporting target broadcast content includes: determining a plurality of broadcast features based on fourth information, where the fourth information includes basic information of the user; Determine a broadcast style based on multiple broadcast features; Broadcast target content based on broadcast style.

[0082] In one embodiment of the present disclosure, the plurality of broadcast features include the user's location, the user's level of understanding, and the user's emotions; The broadcast style is determined based on multiple broadcast characteristics, including: Determine the broadcast language based on the user's location; Determine the recommendation tendency of knowledge content based on the user's understanding level; Determine the broadcast mode based on the user's emotions; The broadcasting style is determined based on the broadcasting language, the recommended tendency of the knowledge content and the broadcasting mode.

[0083] In this embodiment, the fourth information (basic information of the user) refers to information related to the inherent attributes or basic status of the user. For example, the user's location reflects his / her geographical location and the corresponding language and cultural environment; the degree of understanding reflects the user's ability to accept and digest various types of knowledge or information; and emotions represent the user's current psychological and emotional state, such as happiness, frustration, calmness, etc.

[0084] Report features are key elements extracted from basic user information to determine the report style, including region, understanding level, and emotions. These features are interrelated and jointly affect the final report presentation method.

[0085] The broadcast style refers to the overall characteristics of the voice broadcast in terms of language, content depth and breadth, tone and intonation, etc. For example, different broadcast languages, the degree of detail of knowledge content, cheerful or calm broadcast modes, etc. all belong to the category of broadcast style.

[0086] When the smart glasses (first device) are preparing to broadcast the target content, they first obtain the user's basic information (fourth information) and analyze the broadcast characteristics such as the location, understanding level and emotions. For example, through the positioning function, it is known that the user is located in Sichuan, and from the user's past feedback on different knowledge contents, it is judged that the user's understanding level is medium, and through voice intonation analysis or other sensor data, it is judged that the user's current emotion is relatively calm.

[0087] Next, the broadcast style is determined based on these broadcast characteristics. For example, based on the user's location in Sichuan, the broadcast language can be determined to be Sichuan dialect; because the user's comprehension level is medium, the knowledge content recommendation tends to choose medium-difficulty, relatively easy-to-understand content; in view of the user's calm mood, it is determined to adopt a smooth and soothing broadcast mode.

[0088] If it is detected that the user is in a low mood, the smart glasses may use a soft, soothing tone and intonation to broadcast, and the speaking speed may be slowed down appropriately; if the user is excited, the broadcast mode may be more passionate and the speaking speed may be slightly faster, so as to better match the user's emotional state and enhance the user's acceptance of the broadcast content and emotional resonance.

[0089] The user is wearing smart glasses (the first device) and traveling in Italy. He is in a good mood and has a certain understanding of historical and cultural knowledge. When the smart glasses broadcast the introduction of local attractions, they determine that the broadcast language is Italian. The knowledge content recommendation tends to introduce the historical and cultural stories and unique features behind the attractions in depth. The broadcast mode adopts a cheerful and vivid tone, and introduces to the user in a passionate tone and moderate speed: "Benvenuti al Colosseo di Roma! Questoincredibile anfiteatro romano ha una storia lunga e affascinante..." (Welcome to the Colosseum! This incredible Roman amphitheater has a long and fascinating history...).

[0090] From the above, it can be concluded that this embodiment determines the broadcast characteristics and forms the style based on the user's basic information (the fourth information), which can achieve highly personalized broadcasts. For example, the language is determined according to the region, so that users in a foreign country can receive information without obstacles, enhancing the universality and friendliness of the device. Adjusting the knowledge content according to the degree of understanding can make the information reach accurately, whether beginners or senior enthusiasts can get what they need, and improve the efficiency of knowledge dissemination. Considering the emotion setting broadcast mode, it can give warm comfort when the user is low, and share the passion when the user is high, greatly shortening the emotional distance with the user, and upgrading the voice broadcast from a simple information transmission to intimate companionship, comprehensively optimizing the user experience, and improving the user's use of the first device.

[0091] In one embodiment of the present disclosure, when the target broadcast content is broadcasted, it also includes: Collect multiple voice segments of the user; taking a plurality of speech segments as a first speech set; Based on the first voice set, it is determined whether the current broadcast needs to be interrupted.

[0092] In this embodiment, judging whether the current broadcast needs to be interrupted based on the first voice set includes: If the first speech set is a speech segment set related to pause, the current broadcast needs to be interrupted; If the first speech set is a speech segment set related to the adjustment, the current broadcast does not need to be interrupted.

[0093] A speech segment refers to an audio segment that is captured from a user's speech and has relatively complete semantics or analyzable features. For example, a sentence, a phrase, or several consecutive words spoken by a user can be considered a speech segment.

[0094] The first voice set is a set of multiple voice segments collected. This set will be used as the basis for subsequent judgment of whether the broadcast is interrupted, and the user's intention can be inferred through comprehensive analysis of the voice segments in the set.

[0095] The first device may be equipped with a sensitive microphone, which can continuously monitor the surrounding sounds and recognize the user's voice during the announcement. For example, "pause for a moment" or "speak later" will be collected by the first device.

[0096] The collected multiple voice segments are aggregated to form a first voice set, which provides a basis for subsequent judgment.

[0097] The first device uses speech recognition technology to analyze each speech segment in the first speech set. If a keyword or semantic expression with the intention of interrupting the broadcast, such as "pause", "stop", "wait a moment", etc., is detected in the speech set, it is determined that the current broadcast needs to be interrupted.

[0098] From the above, it can be concluded that the first device can collect and analyze user voice segments in real time and accurately capture user intentions. When the voice set indicates that it is related to pause content, the broadcast is interrupted in time to effectively avoid information redundancy and respect the current needs of users. For example, the information broadcast can be paused when the user is busy. As for the uninterrupted broadcast of the set of voice segments related to adjustment, the broadcast experience can be optimized based on user adjustment instructions (such as volume and speech speed adjustment) without interfering with the continuity of information. This method greatly enhances the flexibility and intelligence of voice broadcast interaction, makes the broadcast service fit the user's immediate situation, improves the user's satisfaction and control of the device's voice function, and optimizes the overall usage experience.

[0099] In one embodiment of the present disclosure, when the target broadcast content is broadcasted, it also includes: Collect user's environmental information; Based on the environmental information, determine whether the current broadcast needs to be interrupted.

[0100] In this embodiment, judging whether the current broadcast needs to be interrupted based on the environmental information includes: determining environmental characteristics based on the environmental information; If the environmental feature is the first feature, the current broadcast needs to be interrupted; If the environmental feature is the second feature, the current broadcast does not need to be interrupted; Among them, the amount of information of the first feature is greater than the amount of information of the second feature. The first feature can be a theater environment feature, a conversation environment feature, and a meeting environment feature, and the second feature can be a home environment feature.

[0101] Environmental information refers to the data about the user's surrounding environment collected by the first device (smart glasses) through various sensors, including sound intensity, sound frequency distribution, light intensity, surrounding object recognition and other information. This information can reflect the type and characteristics of the user's environment.

[0102] Environmental features are representative attributes extracted from the collected environmental information and are used to determine the type of environment, such as cinema environment features, home environment features, etc. Different environmental features will affect the voice broadcast strategy.

[0103] The microphone, light sensor and other modules on the smart glasses (the first device) start working to collect data such as the sound and light of the surrounding environment. For example, in a movie theater, the microphone will collect the sound of movie playing in a low-light environment, the slight conversation of the audience, etc.; in a home environment, it can collect information about the normal light intensity in the room and the occasional sound of home appliances running.

[0104] The first device analyzes and processes the collected environmental information. If the sound information detects a continuous characteristic sound of a movie audio channel, and the light information is low-brightness and relatively stable, it is determined to be a cinema environment feature; if the sound is relatively messy but not high in intensity, and the light information conforms to the characteristics of indoor home lighting, it is determined to be a home environment feature.

[0105] When it is determined to be cinema environment feature information, the voice broadcast of smart glasses will be interrupted because cinemas require a quiet viewing environment. For example, a user wears smart glasses to check the movie showtime information and starts watching a movie. Once the user enters the viewing state, the smart glasses detect the cinema environment features, and other unfinished voice broadcasts (such as nearby restaurant recommendations) will stop to avoid disturbing others.

[0106] If it is determined to be a home environment feature, the voice broadcast of smart glasses can proceed normally. For example, when a user listens to news broadcast by smart glasses at home, even if there is some slight ambient noise around, due to the home environment feature, the broadcast will continue and will not be interrupted by these normal home environment interferences, making it convenient for users to receive information at any time.

[0107] From the above, it can be concluded that, on the one hand, this embodiment can effectively prevent the voice broadcast from interfering with a specific environment. For example, in a cinema or conference room, the first device automatically interrupts the broadcast after identifying the corresponding environmental characteristics, thereby maintaining the quietness and order of the environment. On the other hand, maintaining the broadcast in a relatively relaxed environment such as home ensures the continuity of information transmission and does not affect the user's normal reception of information or enjoyment of services. This intelligent interruption judgment mechanism greatly improves the adaptability and flexibility of the voice broadcast function, enabling it to dynamically adjust according to different environments and enhance the user experience.

[0108] Corresponding to the voice broadcast control method of the above embodiment, Figure 2 This is a structural block diagram of a voice broadcast control device provided by an embodiment of the present disclosure. For ease of description, only the parts related to the embodiment of the present disclosure are shown. Figure 2 The voice broadcast control device 20 is applied to the first device, and includes: a first broadcast content determination module 21, a second broadcast content determination module 22, an adjustment module 23 and a broadcast module 24.

[0109] The first broadcast content determination module 21 is used to determine the initial broadcast content based on the first information and the second information in response to receiving the first instruction; the first information includes the user's voice information, and the second information includes the user's preference information; A second broadcast content determination module 22, configured to determine the initial broadcast content based on the second information in response to receiving the second instruction; An adjustment module 23, configured to adjust the content volume of the initial broadcast content based on third information to obtain target broadcast content, wherein the third information includes user behavior information; The broadcast module 24 is used to broadcast target broadcast content.

[0110] In one embodiment of the present disclosure, the first broadcast content determination module 21 is specifically configured to: determining a plurality of speech features of the first information; determining a plurality of preference characteristics of the second information; fusing the plurality of speech features and the plurality of preference features to obtain at least one first feature; Initial broadcast content is determined based on at least one first feature.

[0111] In one embodiment of the present disclosure, the second broadcast content determination module 22 is specifically configured to: determining a plurality of preference characteristics of the second information; Screening multiple preferred features to obtain at least one second feature; The initial broadcast content is determined based on at least one second feature.

[0112] In one embodiment of the present disclosure, the broadcast module 24 is specifically used to: determining a plurality of broadcast features based on fourth information, where the fourth information includes basic information of the user; Determine a broadcast style based on multiple broadcast features; Broadcast target content based on broadcast style.

[0113] In one embodiment of the present disclosure, the plurality of broadcast features include the user's location, the user's level of understanding, and the user's emotions; The broadcast module 24 is further specifically used for: Determine the broadcast language based on the user's location; Determine the recommendation tendency of knowledge content based on the user's understanding level; Determine the broadcast mode based on the user's emotions; The broadcasting style is determined based on the broadcasting language, the recommended tendency of the knowledge content and the broadcasting mode.

[0114] In one embodiment of the present disclosure, the broadcast module 24 is further configured to: Collect multiple voice segments of the user; taking a plurality of speech segments as a first speech set; Based on the first voice set, it is determined whether the current broadcast needs to be interrupted.

[0115] In one embodiment of the present disclosure, the broadcast module 24 is further configured to: Collect user's environmental information; Based on the environmental information, determine whether the current broadcast needs to be interrupted.

[0116] See also Figure 3 , Figure 3 A schematic block diagram of a wearable device provided by an embodiment of the present disclosure. Figure 3 The wearable device 300 in the embodiment shown may include: one or more processors 301, one or more input devices 302, one or more output devices 303 and one or more memories 304. The processors 301, input devices 302, output devices 303 and memories 304 communicate with each other via a communication bus 305. The memory 304 is used to store computer programs, which include program instructions. The processor 301 is used to execute the program instructions stored in the memory 304. The processor 301 is configured to call the program instructions to execute the functions of each module / unit in the above-mentioned device embodiments, such as Figure 2 The functions of modules 21 to 24 are shown.

[0117] It should be understood that in the embodiment of the present disclosure, the processor 301 may be a central processing unit (CPU), and the processor may also be other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor may be a microprocessor or the processor may also be any conventional processor, etc.

[0118] The input device 302 may include a touch panel, a fingerprint collection sensor (for collecting the user's fingerprint information and fingerprint direction information), a microphone, etc., and the output device 303 may include a display (LCD, etc.), a speaker, etc.

[0119] The memory 304 may include a read-only memory and a random access memory, and provide instructions and data to the processor 301. A portion of the memory 304 may also include a non-volatile random access memory. For example, the memory 304 may also store information about the device type.

[0120] In a specific implementation, the processor 301, input device 302, and output device 303 described in the embodiments of the present disclosure can execute the implementation methods described in the first and second embodiments of the voice broadcast control method provided in the embodiments of the present disclosure, and can also execute the implementation methods of the wearable device described in the embodiments of the present disclosure, which will not be repeated here.

[0121] In another embodiment of the present disclosure, a computer-readable storage medium is provided, wherein the computer-readable storage medium stores a computer program, wherein the computer program includes program instructions, and when the program instructions are executed by the processor, all or part of the processes in the above-mentioned embodiment method are implemented, and the computer program can also be completed by instructing the relevant hardware through the computer program. The computer program can be stored in a computer-readable storage medium, and when the computer program is executed by the processor, the steps of each of the above-mentioned method embodiments can be implemented. Among them, the computer program includes computer program code, and the computer program code can be in source code form, object code form, executable file or some intermediate form, etc. The computer-readable medium can include: any entity or device capable of carrying computer program code, recording medium, U disk, mobile hard disk, magnetic disk, optical disk, computer memory, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), electric carrier signal, telecommunication signal and software distribution medium, etc.

[0122] The computer-readable storage medium may be an internal storage unit of the wearable device of any of the aforementioned embodiments, such as a hard disk or memory of the wearable device. The computer-readable storage medium may also be an external storage device of the wearable device, such as a plug-in hard disk, a smart memory card (Smart Media Card, SMC), a secure digital (Secure Digital, SD) card, a flash card (Flash Card), etc. equipped on the wearable device. Furthermore, the computer-readable storage medium may also include both an internal storage unit of the wearable device and an external storage device. The computer-readable storage medium is used to store computer programs and other programs and data required by the wearable device. The computer-readable storage medium may also be used to temporarily store data that has been output or is to be output.

[0123] Those of ordinary skill in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of the two. In order to clearly illustrate the interchangeability of hardware and software, the composition and steps of each example have been generally described in terms of function in the above description. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of this disclosure.

[0124] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working process of the wearable device and unit described above can refer to the corresponding process in the aforementioned method embodiment, and will not be repeated here.

[0125] In the several embodiments provided in the present application, it should be understood that the disclosed wearable devices and methods can be implemented in other ways. For example, the device embodiments described above are only schematic. For example, the division of units is only a logical function division. There may be other division methods in actual implementation. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. In addition, the mutual coupling or direct coupling or communication connection shown or discussed can be an indirect coupling or communication connection through some interfaces or units, or it can be an electrical, mechanical or other form of connection.

[0126] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the embodiments of the present disclosure.

[0127] In addition, each functional unit in each embodiment of the present disclosure may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0128] The above are only specific embodiments of the present disclosure, but the protection scope of the present disclosure is not limited thereto. Any technician familiar with the technical field can easily think of various equivalent modifications or replacements within the technical scope disclosed in the present disclosure, and these modifications or replacements should be included in the protection scope of the present disclosure. Therefore, the protection scope of the present disclosure should be based on the protection scope of the claims.

Claims

1. A voice broadcast control method, characterized in that: Applied to a first device, comprising: In response to receiving the first instruction, determining the initial broadcast content based on first information and second information; the first information includes the user's voice information, and the second information includes the user's preference information; In response to receiving the second instruction, determining initial broadcast content based on the second information; Adjusting the content volume of the initial broadcast content based on third information to obtain target broadcast content, wherein the third information includes user behavior information; The target broadcast content is broadcasted.

2. The voice broadcast control method according to claim 1, characterized in that: The determining of the initial broadcast content based on the first information and the second information includes: determining a plurality of speech features of the first information; determining a plurality of preference characteristics of the second information; fusing the plurality of speech features and the plurality of preference features to obtain at least one first feature; Initial broadcast content is determined based on the at least one first feature.

3. The voice broadcast control method according to claim 1, characterized in that: The determining of the initial broadcast content based on the second information includes: determining a plurality of preference characteristics of the second information; Screening the plurality of preference features to obtain at least one second feature; Initial broadcast content is determined based on the at least one second feature.

4. The voice broadcast control method according to claim 1, characterized in that: The broadcasting of the target broadcast content includes: determining a plurality of broadcast features based on fourth information, wherein the fourth information includes basic information of the user; Determining a broadcast style based on the multiple broadcast features; The target broadcast content is broadcast based on the broadcast style.

5. The voice broadcast control method according to claim 4, characterized in that: The multiple broadcast features include the user's location, the user's understanding level and the user's emotions; The determining of the broadcast style based on the multiple broadcast features includes: Determining a broadcast language based on the user's location; Determining a recommendation tendency of knowledge content based on the user's understanding level; Determining a broadcast mode based on the user's emotions; The announcement style is determined based on the announcement language, the recommendation tendency of the knowledge content and the announcement mode.

6. The voice broadcast control method according to claim 1, characterized in that: When broadcasting the target broadcast content, it also includes: collecting a plurality of voice segments of the user; taking the plurality of speech segments as a first speech set; Based on the first voice set, determine whether the current broadcast needs to be interrupted.

7. The voice broadcast control method according to claim 1, characterized in that: When broadcasting the target broadcast content, it also includes: Collecting environmental information of the user; Based on the environmental information, it is determined whether the current broadcast needs to be interrupted.

8. A voice broadcast control device, characterized in that: Applied to a first device, comprising: A first broadcast content determination module, configured to determine initial broadcast content based on first information and second information in response to receiving a first instruction; the first information includes user voice information, and the second information includes user preference information; A second broadcast content determination module, configured to determine initial broadcast content based on second information in response to receiving a second instruction; An adjustment module, configured to adjust the content volume of the initial broadcast content based on third information to obtain target broadcast content, wherein the third information includes user behavior information; The broadcast module is used to broadcast the target broadcast content.

9. A wearable device comprising a memory, a processor, and a computer program stored in the memory and running on the processor, characterized in that: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 7 are implemented.

Images