Broadcasting regulation method and system based on coupling of cognitive load and information complexity

Abstract

This application discloses a broadcast control method and system based on the coupling of cognitive load and information complexity. The method includes: acquiring the text to be broadcast and evaluating its information structure to obtain an information complexity index; acquiring interaction behavior data between the target user and the virtual interactive carrier, and evaluating cognitive load based on the interaction behavior data to determine the user's cognitive load index; determining a coupling coefficient based on the information complexity index and the user's cognitive load index; and determining a broadcast strategy corresponding to the text to be broadcast based on the coupling coefficient, and controlling the virtual interactive carrier to execute the corresponding broadcast operation according to the broadcast strategy. In the embodiments of this application, broadcast control is achieved by coupling information complexity and user cognitive load index, which can adaptively adjust the broadcast rhythm, alleviate cognitive overload when information is complex and user cognitive load is high, and optimize the broadcast rate and interaction efficiency when conditions are suitable, which can effectively improve the user's information reception effect and conversation experience.

Description

Technical Field

[0001] This application relates to the field of intelligent voice broadcasting technology, and in particular to a broadcasting control method and system based on the coupling of cognitive load and information complexity. Background Technology

[0002] With the development of virtual interactive carrier technology, digital human broadcasting is widely used in scenarios such as intelligent explanation, online consultation, and intelligent education. The broadcasting rhythm directly affects the user's information reception effect.

[0003] Current digital human broadcasting rhythm control solutions mainly fall into two categories: one is fixed-rhythm broadcasting technology, where the system uses a preset fixed speech rate and fixed pause intervals for broadcasting, without dynamically adjusting to the characteristics of the broadcast content or the user's real-time interaction status. This method has low implementation costs, but poor adaptability. When broadcasting low-complexity information, the rhythm is sluggish and the interaction efficiency is low; when broadcasting structured text containing multiple independent key information units and complex logical connections, it is easy to cause cognitive overload for users, making it difficult for them to effectively grasp and remember key information.

[0004] Another type is single-factor adaptive broadcasting technology, which adjusts the broadcasting pace based on a single-dimensional indicator. This includes two methods: adjustment based on information complexity and adjustment based on user emotional state. The information complexity-based approach only analyzes the characteristics of the broadcast text itself, adjusting the broadcasting speed through text length, keyword density, etc., without considering the user's real-time mental burden. The user emotional state-based approach only senses changes in user emotions, ignoring the structural complexity of the broadcast content itself. It is evident that these single-factor control methods lack synergistic consideration between content characteristics and user state, easily leading to mismatches in broadcasting pace adjustment and problems such as user cognitive overload or low interaction efficiency. Summary of the Invention

[0005] Therefore, it is necessary to provide a broadcast control method and system based on the coupling of cognitive load and information complexity to address the above-mentioned technical problems, so as to solve at least one of the problems existing in the prior art.

[0006] Firstly, a broadcast control method based on the coupling of cognitive load and information complexity is provided, including: Obtain the text to be broadcast and evaluate its information structure to determine the information complexity index. Acquire interaction behavior data between the target user and the virtual interactive carrier, and conduct cognitive load assessment based on the interaction behavior data to determine the user's cognitive load index; Based on the information complexity index and the user cognitive load index, a coupling coefficient is determined, which is used to characterize the matching state between the information complexity index and the user cognitive load index. Based on the coupling coefficient, the broadcasting strategy corresponding to the text to be broadcast is determined, and the virtual interactive carrier is controlled to perform the corresponding broadcasting operation according to the broadcasting strategy.

[0007] In one possible implementation, the step of evaluating the information structure of the text to be broadcast and determining the information complexity index includes: The text to be broadcast is subjected to structured parsing to obtain information structure features. The information structure features include at least one of the following: the number of information slots, the number of relationships between slots, and the information density value. The information density value is used to characterize the number of information slots per unit text length. Determine the semantic similarity between the text to be broadcast and historical dialogue records to obtain the information novelty; The information complexity index is obtained by fusing the information structure features and information novelty.

[0008] In one possible implementation, determining the user cognitive load index based on the interaction behavior data includes: Based on the interaction behavior data, interaction session features are extracted. These features include user response latency, number of request repetitions, dialogue turn density, and emotion recognition intensity. The emotion recognition intensity is used to quantify the significance of user emotion expression. The user cognitive load index is obtained by fusing the user response latency, request repetition count, dialogue turn density, and emotion recognition strength.

[0009] In one possible implementation, extracting interaction session features based on the interaction behavior data includes: Record the end timestamp of each round of text broadcast and the start timestamp of the target user initiating the next dialogue. Calculate the single-round response delay based on the time difference between the end timestamp and the start timestamp, and then calculate the average of the multiple round response delays as the user response delay. The number of repeated requests is determined based on the semantic similarity between the current user request session content and the historical request session content. The interaction behavior data is input into a preset emotion recognition model for emotion recognition to obtain the emotion recognition intensity; The total number of effective dialogue rounds is calculated based on the interaction behavior data, and then divided by the total duration of the session to obtain the number of dialogue rounds per unit duration, which is used as the dialogue round density.

[0010] In one possible implementation, determining the broadcasting strategy corresponding to the text to be broadcast based on the coupling coefficient includes: Determine the threshold range corresponding to the coupling coefficient; Based on the mapping relationship between the threshold interval and the broadcast rhythm parameters, the broadcast rhythm parameters corresponding to the text to be broadcast are determined. The broadcast rhythm parameters include at least one of the following: speech rate coefficient, slot interval, and information redundancy marker.

[0011] In one possible implementation, determining the broadcast rhythm parameters corresponding to the text to be broadcast based on the mapping relationship between the threshold interval and the broadcast rhythm parameters includes: When the coupling coefficient is greater than the first preset threshold, the speech rate is slow, the slot interval is the first duration, and the information redundancy is marked as enabled. When the coupling coefficient is less than or equal to the first preset threshold and greater than the second preset threshold, the speech rate is medium-slow, the slot interval is the second duration, and the information redundancy mark is turned off. When the coupling coefficient is less than or equal to the second preset threshold and greater than or equal to the third preset threshold, the speech rate is the standard speed, the slot interval is the third duration, and the information redundancy mark is turned off. When the coupling coefficient is less than the third preset threshold and greater than or equal to the fourth preset threshold, the speech rate is medium-fast, the slot interval is the fourth duration, and the information redundancy mark is turned off. When the coupling coefficient is less than the fourth preset threshold, the speech rate is fast, the slot interval is the fifth duration, and the information redundancy mark is turned off. Among them, the first preset threshold, the second preset threshold, the third preset threshold, and the fourth preset threshold decrease sequentially, and the first duration, the second duration, the third duration, the fourth duration, and the fifth duration decrease sequentially.

[0012] In one possible implementation, after controlling the virtual interactive carrier to perform the corresponding broadcast operation according to the broadcast strategy, the method further includes: Obtain the target user's feedback behavior data regarding this broadcast operation; Based on the feedback behavior data, the user cognitive load index is adaptively optimized.

[0013] Secondly, a broadcast control system based on the coupling of cognitive load and information complexity is provided, including: The information complexity assessment module is used to acquire the text to be broadcast, evaluate the information structure of the text to be broadcast, and determine the information complexity index. The user cognitive load assessment module is used to acquire interaction behavior data between the target user and the virtual interactive carrier, and to conduct cognitive load assessment based on the interaction behavior data to determine the user cognitive load index. The coupling coefficient calculation module is used to determine the coupling coefficient based on the information complexity index and the user cognitive load index. The coupling coefficient is used to characterize the matching state between the information complexity index and the user cognitive load index. The broadcast rhythm generation and broadcast execution module is used to determine the broadcast strategy corresponding to the text to be broadcast based on the coupling coefficient, and control the virtual interactive carrier to perform the corresponding broadcast operation according to the broadcast strategy.

[0014] Thirdly, a computer device is provided, including a memory, a processor, and computer-readable instructions stored in the memory and running on the processor, wherein the processor, when executing the computer-readable instructions, implements the broadcast control method based on the coupling of cognitive load and information complexity as described above.

[0015] Fourthly, a readable storage medium is provided, on which computer-readable instructions are stored, which, when executed by a processor, implement the broadcast control method based on the coupling of cognitive load and information complexity as described above.

[0016] The aforementioned broadcast control method and system based on the coupling of cognitive load and information complexity includes the following steps: acquiring the text to be broadcast and evaluating its information structure to determine an information complexity index; acquiring interaction behavior data between the target user and the virtual interactive carrier and evaluating cognitive load based on the interaction behavior data to determine a user cognitive load index; determining a coupling coefficient based on the information complexity index and the user cognitive load index, whereby the coupling coefficient characterizes the matching state between the information complexity and the user cognitive load index; and determining a broadcast strategy corresponding to the text to be broadcast based on the coupling coefficient, and controlling the virtual interactive carrier to execute the corresponding broadcast operation according to the broadcast strategy. In this embodiment, broadcast control is achieved by coupling information complexity and user cognitive load, overcoming the shortcomings of fixed broadcast rhythms failing to adapt to scene changes and single-factor adjustments failing to consider both content characteristics and user status. On the one hand, it can adaptively adjust the broadcast rhythm according to the text construction complexity and the user's real-time cognitive load level, reducing user reception pressure and preventing cognitive overload in complex information and high cognitive load scenarios; on the other hand, it optimizes the broadcast rate when conditions are suitable, ensuring interaction efficiency. By collaboratively deciding on both content and user aspects, the broadcasting strategy is dynamically adapted. The appropriateness of the broadcasting pace depends on the match between the information complexity and the user's cognitive load. When the user's cognitive load is low, even if the information complexity is high, the broadcasting pace can be appropriately accelerated; conversely, when the user's cognitive load is high, even if the information complexity is low, the broadcasting pace can be slowed down to ensure the user can keep up. This effectively improves the user's information reception and human-computer interaction experience in digital human broadcasting scenarios. Attached Figure Description

[0017] To more clearly illustrate the technical solutions of the embodiments of this application, the drawings used in the description of the embodiments of this application will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0018] Figure 1 This is a schematic diagram of an application environment for a broadcast control method based on the coupling of cognitive load and information complexity in one embodiment of this application; Figure 2 This is a flowchart illustrating a broadcast control method based on the coupling of cognitive load and information complexity in one embodiment of this application. Figure 1 ; Figure 3 This is a flowchart illustrating a broadcast control method based on the coupling of cognitive load and information complexity in one embodiment of this application. Figure 2 ; Figure 4 This is a schematic diagram of an overall module layer of a broadcast control method system based on the coupling of cognitive load and information complexity in one embodiment of this application; Figure 5 This is a schematic diagram of a broadcast control method system based on the coupling of cognitive load and information complexity in one embodiment of this application; Figure 6 This is a schematic diagram of a computer device according to one embodiment of this application. Detailed Implementation

[0019] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0020] The broadcast control method based on the coupling of cognitive load and information complexity provided in this embodiment can be applied to applications such as... Figure 1 The application environment shown may specifically include an input and evaluation layer, a coupled decision layer, and an execution and feedback layer.

[0021] Specifically, the input and evaluation layer may include a text input and parsing layer, an information complexity evaluation layer, and a user cognitive load evaluation layer. The text input and parsing layer, as the system's data source entry point, receives the text to be broadcast and performs structured parsing, identifying information structure features such as information slots (e.g., flight numbers, gate numbers, time numbers), relationships between information slots (e.g., changes, delays, cancellations), and information density (the number of information slots per unit text length). The parsing results are then transmitted to the information complexity evaluation layer. The information complexity evaluation layer receives the parsing results from the text input and parsing layer and calculates the information complexity index (C_inf). This index integrates multiple dimensions such as the number of information slots, the number of relationships between slots, and information novelty, outputting a continuous value in the range of 0-1, with higher values ​​indicating more complex information. The calculation results are then transmitted to the coupling coefficient calculation layer in the coupling decision layer. The user cognitive load assessment layer is used to collect implicit signals during the user's interaction with the digital human in real time (such as user response latency, number of request repetitions, emotion recognition strength, dialogue turn density, etc.), and calculate the user cognitive load index (L_cog) based on these implicit signals, outputting a continuous value in the range of 0-1, with higher values ​​indicating a heavier current cognitive load on the user. The calculation results are transmitted to the coupling coefficient calculation layer in the coupling decision layer.

[0022] The coupling decision layer specifically comprises a coupling coefficient calculation layer and a broadcast rhythm generation layer. The coupling coefficient calculation layer receives the information complexity index and the user cognitive load index, establishes a coupling matching relationship between them, and calculates the coupling coefficient. The calculation result is transmitted to the broadcast rhythm generation layer. The broadcast rhythm generation layer, based on the threshold range of the coupling coefficient, selects the corresponding broadcast rhythm parameters (speech rate coefficient, slot interval duration, information redundancy marker) from a preset set of broadcast parameters to generate the final broadcast rhythm configuration. The configuration result is transmitted to the rhythm execution layer within the execution and feedback layer.

[0023] The execution and feedback layer specifically comprises a rhythm execution layer and a feedback acquisition layer. The rhythm execution layer receives and configures the playback rhythm parameters, driving the digital human to perform voice playback and lip-syncing according to the configured speech rate, pause intervals, redundancy strategies, etc., outputting the final playback content. The feedback acquisition layer continuously collects the user's subsequent interactive behaviors after playback (such as whether repetition is requested, whether key information is correctly repeated, changes in response latency, etc.), feeding the collection results back to the user cognitive load assessment layer to update the user cognitive load index, achieving closed-loop optimization of cognitive load assessment. This ensures that the playback rhythm dynamically adapts to changes in the user's state.

[0024] In one embodiment, such as Figure 2 , Figure 3As shown, a broadcast control method based on the coupling of cognitive load and information complexity is provided, including the following steps: In step S110, the text to be broadcast is obtained, and the information structure of the text to be broadcast is evaluated to determine the information complexity index. Optionally, the system collects or receives the text to be played sent by the virtual interaction carrier during the current session, and determines the text type of the text to be played. If the text to be played is natural language text, semantic parsing is performed on the text to extract multi-dimensional information structure features; if the text to be played is predefined structured text, the built-in multi-dimensional information structure features of the text are directly read. A comprehensive processing is then performed through a preset evaluation model or collaborative computing method to determine the information complexity index C_inf corresponding to the text to be played. The information complexity index C_inf is used to characterize the structural complexity of the text itself. The higher the index value, the more information is included in the text to be played, the more complex the relationships between the information, the higher the information density, and the more cognitive resources the user needs to consume to receive and understand the text.

[0025] The virtual interactive carrier can be a virtual intelligent entity with voice broadcasting and visual interaction capabilities, including but not limited to virtual digital humans, virtual tour guides, intelligent virtual customer service representatives, virtual anchors, and AI interactive virtual avatars, used to perform voice broadcasting, information explanation, and human-computer interaction tasks for users. For example, in scenarios such as users querying flight information, gate changes, or baggage carousel information, the virtual interactive carrier can broadcast corresponding information to the user based on the user's request (e.g., the gate for flight CA1832 changes to C52, and the boarding time changes to 15:30, etc.).

[0026] In step S120, interactive behavior data between the target user and the virtual interactive carrier is acquired, and cognitive load is assessed based on the interactive behavior data to determine the user's cognitive load index. Optionally, during the real-time interaction between the virtual interactive platform and the user, the resulting interactive behavior data can be continuously collected. This data may include characteristics that characterize the user's real-time comprehension state and receiving pressure, such as user response time, frequency of interactive feedback, repeated questioning behavior, silence duration, and operation swiping rhythm. Then, based on preset cognitive load assessment rules or models, the user's cognitive load can be assessed to obtain the user's cognitive load index L_cog at the current moment.

[0027] The user cognitive load index is a dynamically updated quantitative indicator used to characterize the information reception burden, comprehension pressure, and mental exertion of users in the current interaction scenario. A higher index indicates greater information reception pressure and higher cognitive resource consumption, making users more prone to comprehension delays, information omissions, and cognitive overload.

[0028] Understandably, a user cognitive overload protection mechanism can be implemented. During real-time broadcast control, the dynamically updated user cognitive load index is continuously monitored. If this index exceeds a preset upper threshold, the user is determined to be in a state of high cognitive overload, with a significant decrease in information reception capacity. At this point, the system automatically activates the overload protection strategy, pausing the continuous output of subsequent broadcast content and confirming the user's current state and providing a buffer through simple inquiries, status confirmations, and pause prompts. Once the user provides feedback and their cognitive load index returns to the normal threshold range, the remaining content can continue to be broadcast. This effectively avoids the system continuously outputting complex information when the user is cognitively saturated, preventing the continuous accumulation of mental burden on the user, avoiding problems such as information omissions, comprehension failures, and decreased interactive experience, and improving user experience and information reception effectiveness.

[0029] In step S130, a coupling coefficient is determined based on the information complexity index and the user cognitive load index. The coupling coefficient is used to characterize the matching state between the information complexity index and the user cognitive load index. Optionally, after obtaining the information complexity index C_inf and the user cognitive load index L_cog corresponding to the text to be broadcast, the coupling coefficient K_couple can be determined by combining these two indicators. This coupling coefficient K_couple is used to characterize the matching coupling state between information complexity and user cognitive load, reflecting the cognitive interference effect generated by their superposition. Specifically, the coupling coefficient K_couple can be obtained by subtraction, such as calculating the difference between the normalized value of the information complexity index and the normalized value of the user cognitive load index, as shown below: K_couple = C_inf - L_cog; Wherein, C_inf-L_cog represents the difference between the information complexity index C_inf and the user cognitive load index L_cog.

[0030] The coupling coefficient K_couple can take values ​​in the range [-1, 1], and the meanings of each value interval are as follows: K_couple>0 indicates that the information complexity exceeds the user's cognitive load, and the user may have difficulty keeping up with the broadcast pace; K_couple≈0 indicates that the information complexity matches the user's cognitive load, and the current broadcast rhythm is appropriate; K_couple<0 indicates that the user's cognitive load is higher than the information complexity, and the broadcasting pace can be appropriately accelerated.

[0031] In addition, the coupling coefficient can be calculated using methods such as rule matching and weighted fusion. Taking rule matching as an example, the information complexity index and user cognitive load index can be pre-divided into level ranges. When both the information complexity index and the user cognitive load index are in the high range, the first-level coupling coefficient with a larger adjustment range is matched; when only one index is above the corresponding threshold, the second-level coupling coefficient with a smaller adjustment range is matched. Taking the weighted fusion method as an example, corresponding control weights can be configured for the information complexity index and the user cognitive load index respectively, and the coupling coefficient is obtained through weighted calculation. The weight ratio of the two indicators can be dynamically adjusted according to the interaction scenario type.

[0032] In step S140, based on the coupling coefficient, the broadcasting strategy corresponding to the text to be broadcast is determined, and the virtual interactive carrier is controlled to perform the corresponding broadcasting operation according to the broadcasting strategy.

[0033] Optionally, the broadcasting strategy corresponding to the coupling coefficient can be determined by methods such as mapping table lookup and mapping function calculation. The broadcasting strategy may specifically include one or more control parameters such as speech rate coefficient, slot interval duration, and information redundancy marker. Then, the system can control the virtual interactive carrier to perform corresponding broadcasting operations according to the matched broadcasting strategy. For example, the speech rate of the TTS speech synthesis can be adjusted according to the speech rate coefficient; a specified duration of silence interval can be inserted between information slots; if the information redundancy marker is enabled, the key information can be repeated in a simplified form immediately after broadcasting; the lip-syncing rhythm of the digital human can be adjusted synchronously to ensure lip-syncing and speech matching, thereby achieving dynamic adaptation of the broadcasting rhythm to content characteristics and user cognitive state.

[0034] It should be noted that the higher the coupling coefficient, the greater the cognitive pressure resulting from the superposition of information complexity and user cognitive load. Correspondingly, the broadcast rate can be reduced, pause intervals can be increased, and information redundancy markers can be enabled to repeat key information, thereby improving user reception and conversation performance. If the coupling coefficient is low, the broadcast rate can be appropriately increased, pause intervals can be reduced, and information redundancy markers can be disabled to improve interaction efficiency.

[0035] Furthermore, this coupling coefficient can not only adjust the speech rate of TTS speech synthesis, but also synchronously regulate the visual performance parameters of the virtual interactive platform. These visual performance parameters include at least one of the following: digital human movement density, facial expression amplitude, and intensity of visual stimuli. For example, a higher coupling coefficient can reduce the frequency of digital human body movements, weaken the amplitude of facial expression fluctuations, reduce unnecessary visual information interference, and lower the user's information reception load; while a lower coupling coefficient can moderately restore the richness of visual performance and enhance the vividness of interaction.

[0036] This application provides a broadcast control method based on the coupling of cognitive load and information complexity, comprising: acquiring text to be broadcast and evaluating the information structure of the text to be broadcast to determine an information complexity index; acquiring interaction behavior data between a target user and a virtual interactive carrier and evaluating cognitive load based on the interaction behavior data to determine a user cognitive load index; determining a coupling coefficient based on the information complexity index and the user cognitive load index, the coupling coefficient being used to characterize the matching state between the information complexity and the user cognitive load index; determining a broadcast strategy corresponding to the text to be broadcast based on the coupling coefficient, and controlling the virtual interactive carrier to execute the corresponding broadcast operation according to the broadcast strategy. In this application embodiment, broadcast control is achieved by coupling information complexity and user cognitive load index, overcoming the shortcomings of fixed broadcast rhythms being unable to adapt to scene changes and single-factor adjustments being unable to simultaneously consider content characteristics and user states. On the one hand, it can adaptively adjust the broadcast rhythm according to the text construction complexity and the user's real-time cognitive load level, reducing the user's receiving pressure and preventing cognitive overload in complex information and high cognitive load scenarios; on the other hand, it optimizes the broadcast rate when conditions are suitable, ensuring interaction efficiency. By collaboratively deciding on both content and user aspects, the broadcasting strategy is dynamically adapted. The appropriateness of the broadcasting pace depends on the match between the information complexity and the user's cognitive load. When the user's cognitive load is low, even if the information complexity is high, the broadcasting pace can be appropriately accelerated; conversely, when the user's cognitive load is high, even if the information complexity is low, the broadcasting pace can be slowed down to ensure the user can keep up. This effectively improves the user's information reception and human-computer interaction experience in digital human broadcasting scenarios.

[0037] In one embodiment of this application, the step of evaluating the information structure of the text to be broadcast and determining the information complexity index includes: The text to be broadcast is subjected to structured parsing to obtain information structure features. The information structure features include at least one of the following: the number of information slots, the number of relationships between slots, and the information density value. The information density value is used to characterize the number of information slots per unit text length. Determine the semantic similarity between the text to be broadcast and historical dialogue records to obtain the information novelty; The information complexity index is obtained by fusing the information structure features and information novelty.

[0038] Optionally, the text information to be broadcast on the virtual interactive carrier is obtained through a text input and parsing layer, and the text is then structured and parsed. Specifically, word segmentation, entity recognition, and semantic role labeling are performed on the text first to identify information units (information slots) with independent semantics, such as flight numbers (CA1832, MU586), gate numbers (C52, D18), time numbers (15:30), baggage carousel numbers (carousel number 7), etc.; then, through semantic dependency analysis and logical relationship recognition, constraints, parallelism, temporal relationships, etc., between different information units are extracted. Relationships such as attribution (e.g., change relationship, "changed from C52 to D18"), delay relationship (e.g., delayed until 16:45"), and causal relationship (e.g., delay due to weather) are identified. Information density, i.e., the number of information slots per unit text length, is calculated. Finally, the number of information slots N_slot, the number of relationships between slots N_relation, and the information density value D_info are statistically analyzed and integrated to form an information slot list, a relationship list, and an information density value. This yields information structure features, which are used to characterize the complexity of the text's content organization.

[0039] Then, historical dialogue records are collected, such as the most recent N rounds of dialogue. The text to be broadcast is then semantically compared with the historical dialogue records. By calculating semantic similarity, duplicate and novel information are identified, and the information novelty is obtained. Specifically, the semantic feature vectors of the text to be broadcast and the text of the most recent N rounds of dialogue are extracted respectively. The semantic similarity between each information segment in the text to be broadcast and the content of the historical dialogue is calculated segment by segment. A similarity threshold is set. If the semantic similarity of a segment is higher than the similarity threshold, the segment is determined to be duplicate information that has been pushed before. If it is lower than the threshold, it is determined to be novel information. The information novelty N_novel is obtained by counting the number of new information in the text to be broadcast.

[0040] Finally, fusion methods such as weighted fusion and multi-dimensional feature mapping can be used to fuse the aforementioned information structure features and information novelty in multiple dimensions to obtain the information complexity index. Taking the weighted fusion method as an example, the information structure features and information novelty can be normalized separately, mapping each index to a standard numerical range of 0-1. Then, corresponding weights can be assigned to the normalized information structure features and information novelty. Each weight parameter can be obtained through extensive scenario experiments and user interaction data calibration. At the same time, it can be dynamically updated and adapted according to conditions such as interaction scenario, user group, and interaction stage. After completing the weight configuration, the two types of features can be weighted and fused to obtain the information complexity C_inf (a continuous value in the 0-1 range).

[0041] For example, taking the structural features as the number of information slots N_slot, the number of relationships between slots N_relation, and the information density value D_info, the information complexity C_inf can be calculated using the following formula: C_inf=α·N_slot+β·N_relation+γ·N_novel+δ·D_info; Among them, α, β, γ, and δ represent weights, and α+β+γ+δ=1.

[0042] In one embodiment of this application, determining the user's cognitive load index based on the interaction behavior data includes: Based on the interaction behavior data, interaction session features are extracted. These features include user response latency, number of request repetitions, dialogue turn density, and emotion recognition intensity. The emotion recognition intensity is used to quantify the significance of user emotion expression. The user cognitive load index is obtained by fusing the user response latency, request repetition count, dialogue turn density, and emotion recognition strength.

[0043] Optionally, the user cognitive load assessment layer can collect the interaction session characteristics between the user and the virtual interaction carrier in real time during the interaction session, and comprehensively calculate the user cognitive load index L_cog. Specifically, these interaction session characteristics may include: user response latency T_response, number of request repetitions N_repeat, dialogue turn density T_turn, and emotion recognition intensity S_emotion.

[0044] Then, fusion methods such as weighted fusion and multi-dimensional feature mapping can be used to fuse user response latency, request repetition count, dialogue turn density, and emotion recognition intensity to obtain the user cognitive load index L_cog. Taking weighted fusion as an example, firstly, each feature of user response latency, request repetition count, dialogue turn density, and emotion recognition intensity is normalized and uniformly mapped to the 0-1 standard value range to eliminate the influence of differences in the dimensions of different indicators; then, corresponding weights are assigned to each feature, which can be calibrated based on interactive experimental data or dynamically adjusted according to the interaction scenario and user type; finally, the normalized features are multiplied by their corresponding weights and summed to obtain the user cognitive load index L_cog (a continuous value in the 0-1 range).

[0045] The user cognitive load index L_cog can be calculated using the following formula: L_cog=λ·T_response'+μ·N_repeat'+ν·S_emotion+ξ·T_turn'; Among them, λ, μ, ν, and ξ are weights, and λ+μ+ν+ξ=1.

[0046] In one embodiment of this application, extracting interaction session features based on the interaction behavior data includes: Record the end timestamp of each round of text broadcast and the start timestamp of the target user initiating the next dialogue. Calculate the single-round response delay based on the time difference between the end timestamp and the start timestamp, and then calculate the average of the multiple round response delays as the user response delay. The number of repeated requests is determined based on the semantic similarity between the current user request session content and the historical request session content. The interaction behavior data is input into a preset emotion recognition model for emotion recognition to obtain the emotion recognition intensity; The total number of effective dialogue rounds is calculated based on the interaction behavior data, and then divided by the total duration of the session to obtain the number of dialogue rounds per unit duration, which is used as the dialogue round density.

[0047] Optionally, based on the collected real-time user interaction behavior data, multiple interactive session features are extracted to characterize the user's cognitive state, specifically including user response delay T_response, number of request repetitions N_repeat, dialogue turn density T_turn, and emotion recognition intensity S_emotion.

[0048] The user response latency T_response reflects the time taken for a user to understand and respond to information in a single round. Specifically, it is calculated by recording the end timestamp of each round of text broadcast and the start timestamp of the user initiating the next dialogue, calculating the time difference between the two timestamps, and then averaging the single-round response latency of multiple consecutive rounds of interaction as the final user response latency T_response, which reflects the overall lag in the user's information reception.

[0049] The number of repeated requests, N_repeat, is used to characterize the interaction state where the user's understanding of information is ambiguous and the content is not digested. By comparing the semantic similarity between the current user request session content and the historical request session content, the frequency of the user's repeated inquiries and confirmations is identified, and the number of repeated requests, N_repeat, is obtained.

[0050] The emotion recognition intensity S_emotion is used to help judge the user's cognitive burden state from the emotional dimension. The collected interaction behavior data and user text conversation data can be input into a preset emotion recognition model (which can be at least one of the following: a text emotion classification model built based on a pre-trained language model, a lightweight semantic perception network, or a rule-based emotion dictionary model) to identify the activity level of negative emotions such as confusion, doubt, irritability, and fatigue of the user and output the corresponding emotion recognition intensity (0-1 range).

[0051] Dialogue turn density T_turn is used to characterize the interaction compactness and information input frequency of the current session. It is obtained by averaging the total number of effective dialogue turns in the current interactive session and combining it with the total duration of the session. The dialogue turns per unit time are the dialogue turn density T_turn.

[0052] In one embodiment of this application, determining the broadcasting strategy corresponding to the text to be broadcast based on the coupling coefficient includes: Determine the threshold range corresponding to the coupling coefficient; Based on the mapping relationship between the threshold interval and the broadcast rhythm parameters, the broadcast rhythm parameters corresponding to the text to be broadcast are determined. The broadcast rhythm parameters include at least one of the following: speech rate coefficient, slot interval, and information redundancy marker.

[0053] Optionally, multi-level threshold intervals corresponding to the coupling coefficients can be pre-defined and configured. Different threshold intervals correspond to different user cognitive loads and complex superposition states of text information. After obtaining the real-time coupling coefficients, their respective target threshold intervals are matched. Then, based on the one-to-one mapping relationship between the preset threshold intervals and the broadcast rhythm parameters, the broadcast rhythm parameters corresponding to the current text to be broadcast are determined. Through hierarchical threshold matching and parameter mapping, the broadcast rhythm parameters can be adaptively output in real time according to the dynamic changes in text complexity and user real-time cognitive state. This avoids the problems of poor information adaptability, user cognitive overload, or low interaction efficiency caused by a fixed broadcast rhythm, effectively improving the intelligence and adaptability of broadcast interaction.

[0054] The broadcast rhythm parameters include at least one of the following: speech rate coefficient, slot interval duration, and information redundancy marker. The speech rate coefficient is used to control the speed of TTS broadcasting on the virtual interactive carrier; the slot interval is the pause buffer duration between adjacent information slots and text segments, allowing users sufficient time to digest the information; the information redundancy marker is used to identify whether the current broadcast scenario needs to add auxiliary explanations, repetitive broadcasts, or other redundant information content.

[0055] In one embodiment of this application, determining the broadcast rhythm parameters corresponding to the text to be broadcast based on the mapping relationship between the threshold interval and the broadcast rhythm parameters includes: When the coupling coefficient is greater than the first preset threshold, the speech rate is slow, the slot interval is the first duration, and the information redundancy is marked as enabled. When the coupling coefficient is less than or equal to the first preset threshold and greater than the second preset threshold, the speech rate is medium-slow, the slot interval is the second duration, and the information redundancy mark is turned off. When the coupling coefficient is less than or equal to the second preset threshold and greater than or equal to the third preset threshold, the speech rate is the standard speed, the slot interval is the third duration, and the information redundancy mark is turned off. When the coupling coefficient is less than the third preset threshold and greater than or equal to the fourth preset threshold, the speech rate is medium-fast, the slot interval is the fourth duration, and the information redundancy mark is turned off. When the coupling coefficient is less than the fourth preset threshold, the speech rate is fast, the slot interval is the fifth duration, and the information redundancy mark is turned off. Among them, the first preset threshold, the second preset threshold, the third preset threshold, and the fourth preset threshold decrease sequentially, and the first duration, the second duration, the third duration, the fourth duration, and the fifth duration decrease sequentially.

[0056] Optionally, the broadcast rhythm parameters corresponding to the text to be broadcast are determined based on a preset mapping relationship between the threshold range and the broadcast rhythm parameters. Specifically, when the coupling coefficient is greater than the first preset threshold, the broadcast speed is configured to slow, the slot interval is set to the first duration, and the information redundancy flag is enabled; when the coupling coefficient is less than or equal to the first preset threshold and greater than the second preset threshold, the broadcast speed is configured to medium-slow, the slot interval is set to the second duration, and the information redundancy flag is disabled; when the coupling coefficient is less than or equal to the second preset threshold and greater than or equal to the third preset threshold, the broadcast speed is configured to standard, the slot interval is set to the third duration, and the information redundancy flag is disabled; when the coupling coefficient is less than or equal to the third preset threshold and greater than or equal to the fourth preset threshold, the broadcast speed is configured to medium-fast, the slot interval is set to the fourth duration, and the information redundancy flag is disabled; when the coupling coefficient is less than the fourth preset threshold, the broadcast speed is configured to fast, the slot interval is set to the fifth duration, and the information redundancy flag is disabled.

[0057] The preset thresholds are: first > second > third > fourth; and first duration > second duration > third duration > fourth duration > fifth duration. It is understood that a higher coupling coefficient indicates a stronger combined effect of information complexity and user cognitive load. This is mitigated by reducing the broadcast speed, lengthening the slot intervals, and enabling information redundancy supplementation as needed. Conversely, a lower coupling coefficient allows for a faster broadcast pace and shorter pauses, balancing interaction efficiency. All the aforementioned thresholds and duration parameters can be pre-calibrated according to the interaction scenario or adaptively adjusted accordingly.

[0058] For example, the mapping rules between the threshold range and the broadcast rhythm parameters are shown in Table 1 below: Table 1: Broadcast Parameter Mapping Table

[0059] As can be seen from the above Table 1, the relative strength relationship between the user's cognitive load and the information complexity can be distinguished according to the numerical interval of the coupling coefficient K_couple, and the corresponding broadcast rhythm parameters are bound. When K_couple > +0.5, the broadcast speech speed coefficient is configured as 0.7, the speech speed is slow, the slot interval is set to 800ms, and the information redundancy flag (repeating key information) is enabled; when +0.2 < K_couple ≤ +0.5, the broadcast speech speed coefficient is configured as 0.85, the speech speed is medium-slow, the slot interval is set to 500ms, and the information redundancy flag is disabled; when -0.2 ≤ K_couple ≤ +0.2, the broadcast speech speed coefficient is configured as 1.0, the speech speed is standard speed, the slot interval is set to 300ms, and the information redundancy flag is disabled; when -0.5 ≤ K_couple < -0.2, the broadcast speech speed coefficient is configured as 1.15, the speech speed is medium-fast, the slot interval is set to 200ms, and the information redundancy flag is disabled; when K_couple < -0.5, the broadcast speech speed is configured as fast, the slot interval is set to 100ms, and the information redundancy flag is disabled.

[0060] It can be seen that when the information complexity is high, the information itself will bring a significant understanding burden to the user. At this time, the broadcast speech speed should be reduced, the slot interval should be extended, and the information redundancy flag should be enabled when necessary, so as to give the user sufficient time to digest the information and avoid cognitive overload; on the contrary, when the information complexity is relatively low and the cognitive pressure mainly comes from the user side, there is no need to reserve a long buffer time. By moderately increasing the speed and reducing pauses, the information push efficiency is improved within the controllable cognitive load range, so as to realize intelligent differentiated broadcast regulation.

[0061] In an embodiment of the present application, after controlling the virtual interaction carrier to perform the corresponding broadcast operation according to the broadcast strategy, the method further includes: Acquire the feedback behavior data of the target user for this broadcast operation; Adaptively optimize the user cognitive load index based on the feedback behavior data.

[0062] Optionally, after the virtual interactive platform executes the broadcasting operation according to the broadcasting strategy, it can continuously monitor the user's subsequent interactive behavior and collect feedback signals. For example, it can collect data such as whether the user requests repeated broadcasting (yes / no); the trend of the user's response delay (increase / decrease compared to the historical average); and whether the user correctly repeats key information in subsequent interactions. Then, based on the collected feedback signals, an interaction effect evaluation result can be formed, and the weight parameters corresponding to each feature used to solve the user's cognitive load index L_cog can be iteratively updated according to this evaluation result, achieving adaptive optimization of the weights. For example, if the user repeatedly fails to repeat key information or frequently requests repeated broadcasting, it indicates a deviation in the current cognitive load assessment. The weight ratio of features such as user response delay and emotion recognition strength can be appropriately increased to optimize the prediction accuracy of the cognitive load index in the next stage, ensuring that the broadcasting control strategy continuously adapts to individual user characteristics.

[0063] In the next round of interaction, the optimized weight parameters can be used to calculate the user load index, thereby generating a cognitive load index that better reflects the user's actual cognitive state. Based on the optimized user cognitive load index, and combined with the information complexity index of the text to be broadcast in the current round, the coupling coefficient is calculated to dynamically determine the broadcast strategy for the text to be broadcast in the current round. This closed-loop iterative approach allows the cognitive load assessment model to adapt to individual differences and changes in user state. Furthermore, as session data accumulates with each round of interaction, the personalization and long-term adaptability of the broadcast control scheme are gradually improved.

[0064] In this embodiment, broadcast control is achieved by coupling information complexity with the user's cognitive load index, overcoming the shortcomings of fixed broadcast rhythms that cannot adapt to scene changes and single-factor adjustments that cannot take into account both content characteristics and user status. On the one hand, the broadcast rhythm can be adaptively adjusted according to the complexity of text construction and the user's real-time cognitive load level, reducing the user's receiving pressure and preventing cognitive overload in complex information and high cognitive load scenarios; on the other hand, the broadcast rate is optimized when conditions are suitable to ensure interaction efficiency. Through collaborative decision-making between content and user, the broadcast strategy is dynamically adapted. The rationality of the broadcast rhythm depends on the degree of matching between information complexity and user cognitive load status. When the user's cognitive load is low, even if the information complexity is high, the broadcast rhythm can be appropriately accelerated; when the user's cognitive load is high, even if the information complexity is low, the broadcast rhythm can be slowed down to ensure that the user can keep up. This effectively improves the user's information reception effect and human-computer interaction experience in digital human broadcast scenarios.

[0065] It should be understood that the sequence number of each step in the above embodiments does not imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of this application.

[0066] In one embodiment, a broadcast control system based on the coupling of cognitive load and information complexity is provided, which corresponds one-to-one with the broadcast control system based on the coupling of cognitive load and information complexity in the above embodiments. For example... Figure 4 , Figure 5 As shown, the broadcast control system based on the coupling of cognitive load and information complexity includes an information complexity assessment module 10, a user cognitive load assessment module 20, a coupling coefficient calculation module 30, and a broadcast rhythm generation and broadcast execution module 40. Detailed descriptions of each functional module are as follows: The information complexity assessment module 10 is used to acquire the text to be broadcast and to assess the information structure of the text to be broadcast in order to determine the information complexity index. The user cognitive load assessment module 20 is used to acquire interaction behavior data between the target user and the virtual interactive carrier, and to perform cognitive load assessment based on the interaction behavior data to determine the user cognitive load index. The coupling coefficient calculation module 30 is used to determine the coupling coefficient based on the information complexity index and the user cognitive load index, wherein the coupling coefficient is used to characterize the matching state between the information complexity index and the user cognitive load index. The broadcast rhythm generation and broadcast execution module 40 is used to determine the broadcast strategy corresponding to the text to be broadcast based on the coupling coefficient, and control the virtual interactive carrier to perform the corresponding broadcast operation according to the broadcast strategy.

[0067] In one embodiment of this application, the information complexity evaluation module 10 is further configured to: The text to be broadcast is subjected to structured parsing to obtain information structure features. The information structure features include at least one of the following: the number of information slots, the number of relationships between slots, and the information density value. The information density value is used to characterize the number of information slots per unit text length. Determine the semantic similarity between the text to be broadcast and historical dialogue records to obtain the information novelty; The information complexity index is obtained by fusing the information structure features and information novelty.

[0068] In one embodiment of this application, the user cognitive load assessment module 20 is further configured to: Based on the interaction behavior data, interaction session features are extracted. These features include user response latency, number of request repetitions, dialogue turn density, and emotion recognition intensity. The emotion recognition intensity is used to quantify the significance of user emotion expression. The user cognitive load index is obtained by fusing the user response latency, request repetition count, dialogue turn density, and emotion recognition strength.

[0069] In one embodiment of this application, the user cognitive load assessment module 20 is further configured to: Record the end timestamp of each round of text broadcast and the start timestamp of the target user initiating the next dialogue. Calculate the single-round response delay based on the time difference between the end timestamp and the start timestamp, and then calculate the average of the multiple round response delays as the user response delay. The number of repeated requests is determined based on the semantic similarity between the current user request session content and the historical request session content. The interaction behavior data is input into a preset emotion recognition model for emotion recognition to obtain the emotion recognition intensity; The total number of effective dialogue rounds is calculated based on the interaction behavior data, and then divided by the total duration of the session to obtain the number of dialogue rounds per unit duration, which is used as the dialogue round density.

[0070] In one embodiment of this application, the broadcast rhythm generation and broadcast execution module 40 is further configured to: Determine the threshold range corresponding to the coupling coefficient; Based on the mapping relationship between the threshold interval and the broadcast rhythm parameters, the broadcast rhythm parameters corresponding to the text to be broadcast are determined. The broadcast rhythm parameters include at least one of the following: speech rate coefficient, slot interval, and information redundancy marker.

[0071] In one embodiment of this application, the broadcast rhythm generation and broadcast execution module 40 is further configured to: When the coupling coefficient is greater than the first preset threshold, the speech rate is slow, the slot interval is the first duration, and the information redundancy is marked as enabled. When the coupling coefficient is less than or equal to the first preset threshold and greater than the second preset threshold, the speech rate is medium-slow, the slot interval is the second duration, and the information redundancy mark is turned off. When the coupling coefficient is less than or equal to the second preset threshold and greater than or equal to the third preset threshold, the speech rate is the standard speed, the slot interval is the third duration, and the information redundancy mark is turned off. When the coupling coefficient is less than the third preset threshold and greater than or equal to the fourth preset threshold, the speech rate is medium-fast, the slot interval is the fourth duration, and the information redundancy mark is turned off. When the coupling coefficient is less than the fourth preset threshold, the speech rate is fast, the slot interval is the fifth duration, and the information redundancy mark is turned off. Among them, the first preset threshold, the second preset threshold, the third preset threshold, and the fourth preset threshold decrease sequentially, and the first duration, the second duration, the third duration, the fourth duration, and the fifth duration decrease sequentially.

[0072] In one embodiment of this application, the system further includes a feedback acquisition module, used for: Obtain the target user's feedback behavior data regarding this broadcast operation; Based on the feedback behavior data, the user cognitive load index is adaptively optimized.

[0073] In this embodiment, broadcast control is achieved by coupling information complexity with the user's cognitive load index, overcoming the shortcomings of fixed broadcast rhythms that cannot adapt to scene changes and single-factor adjustments that cannot take into account both content characteristics and user status. On the one hand, the broadcast rhythm can be adaptively adjusted according to the complexity of text construction and the user's real-time cognitive load level, reducing the user's receiving pressure and preventing cognitive overload in complex information and high cognitive load scenarios; on the other hand, the broadcast rate is optimized when conditions are suitable to ensure interaction efficiency. Through collaborative decision-making between content and user, the broadcast strategy is dynamically adapted. The rationality of the broadcast rhythm depends on the degree of matching between information complexity and user cognitive load status. When the user's cognitive load is low, even if the information complexity is high, the broadcast rhythm can be appropriately accelerated; when the user's cognitive load is high, even if the information complexity is low, the broadcast rhythm can be slowed down to ensure that the user can keep up. This effectively improves the user's information reception effect and human-computer interaction experience in digital human broadcast scenarios.

[0074] Specific limitations regarding the broadcast control system based on the coupling of cognitive load and information complexity can be found in the limitations of the broadcast control method based on the coupling of cognitive load and information complexity mentioned above, and will not be repeated here. Each module in the aforementioned broadcast control system based on the coupling of cognitive load and information complexity can be implemented entirely or partially through software, hardware, or a combination thereof. These modules can be embedded in or independent of the processor in a computer device, or stored in the memory of a computer device as software, so that the processor can call and execute the corresponding operations of each module.

[0075] In one embodiment, a computer device is provided, which may be a terminal device, and its internal structure diagram may be as follows: Figure 6 As shown, the computer device includes a processor, memory, and network interface connected via a system bus. The processor provides computing and control capabilities. The memory includes a readable storage medium storing computer-readable instructions. The network interface communicates with external terminals via a network connection. When executed by the processor, the computer-readable instructions implement a broadcast control system based on the coupling of cognitive load and information complexity. The readable storage medium provided in this embodiment includes both non-volatile and volatile readable storage media.

[0076] In this application embodiment, a computer device is provided, including a memory, a processor, and computer-readable instructions stored in the memory and executable on the processor. When the processor executes the computer-readable instructions, it implements the steps of the broadcast control method based on the coupling of cognitive load and information complexity as described above.

[0077] In this embodiment of the application, a readable storage medium is provided, which stores computer-readable instructions. When the computer-readable instructions are executed by a processor, they implement the steps of the broadcast control method based on the coupling of cognitive load and information complexity as described above.

[0078] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by instructing related hardware with computer-readable instructions. These computer-readable instructions can be stored in a non-volatile readable storage medium or a volatile readable storage medium. When executed, these computer-readable instructions can include the processes of the embodiments of the methods described above. Any references to memory, storage, databases, or other media used in the embodiments provided in this application can include non-volatile and / or volatile memory. Non-volatile memory may include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory. Volatile memory may include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in a variety of forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), dual data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link DRAM (SLDRAM), RAMbus direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM), etc.

[0079] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the above-described division of functional units and modules is used as an example. In practical applications, the above functions can be assigned to different functional units and modules as needed, that is, the internal structure of the device can be divided into different functional units or modules to complete all or part of the functions described above.

[0080] The above embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of this application, and should all be included within the protection scope of this application.

Claims

1. A broadcast control method based on the coupling of cognitive load and information complexity, characterized in that, The method includes: Obtain the text to be broadcast and evaluate its information structure to determine the information complexity index. Acquire interaction behavior data between the target user and the virtual interactive carrier, and conduct cognitive load assessment based on the interaction behavior data to determine the user's cognitive load index; Based on the information complexity index and the user cognitive load index, a coupling coefficient is determined, which is used to characterize the matching state between the information complexity index and the user cognitive load index. Based on the coupling coefficient, the broadcasting strategy corresponding to the text to be broadcast is determined, and the virtual interactive carrier is controlled to perform the corresponding broadcasting operation according to the broadcasting strategy.

2. The broadcast control method based on the coupling of cognitive load and information complexity as described in claim 1, characterized in that, The step of evaluating the information structure of the text to be broadcast and determining the information complexity index includes: The text to be broadcast is subjected to structured parsing to obtain information structure features. The information structure features include at least one of the following: the number of information slots, the number of relationships between slots, and the information density value. The information density value is used to characterize the number of information slots per unit text length. Determine the semantic similarity between the text to be broadcast and historical dialogue records to obtain the information novelty; The information complexity index is obtained by fusing the information structure features and information novelty.

3. The broadcast control method based on the coupling of cognitive load and information complexity as described in claim 1, characterized in that, Determining the user's cognitive load index based on the interaction behavior data includes: Based on the interaction behavior data, interaction session features are extracted. These features include user response latency, number of request repetitions, dialogue turn density, and emotion recognition intensity. The emotion recognition intensity is used to quantify the significance of user emotion expression. The user cognitive load index is obtained by fusing the user response latency, request repetition count, dialogue turn density, and emotion recognition strength.

4. The broadcast control method based on the coupling of cognitive load and information complexity as described in claim 3, characterized in that, The step of extracting interaction session features based on the interaction behavior data includes: Record the end timestamp of each round of text broadcast and the start timestamp of the target user initiating the next dialogue. Calculate the single-round response delay based on the time difference between the end timestamp and the start timestamp, and then calculate the average of the multiple round response delays as the user response delay. The number of repeated requests is determined based on the semantic similarity between the current user request session content and the historical request session content. The interaction behavior data is input into a preset emotion recognition model for emotion recognition to obtain the emotion recognition intensity; The total number of effective dialogue rounds is calculated based on the interaction behavior data, and then divided by the total duration of the session to obtain the number of dialogue rounds per unit duration, which is used as the dialogue round density.

5. The broadcast control method based on the coupling of cognitive load and information complexity as described in claim 1, characterized in that, The step of determining the broadcasting strategy corresponding to the text to be broadcast based on the coupling coefficient includes: Determine the threshold range corresponding to the coupling coefficient; Based on the mapping relationship between the threshold interval and the broadcast rhythm parameters, the broadcast rhythm parameters corresponding to the text to be broadcast are determined. The broadcast rhythm parameters include at least one of the following: speech rate coefficient, slot interval, and information redundancy marker.

6. The broadcast control method based on the coupling of cognitive load and information complexity as described in claim 5, characterized in that, The step of determining the broadcast rhythm parameters corresponding to the text to be broadcast based on the mapping relationship between the threshold interval and the broadcast rhythm parameters includes: When the coupling coefficient is greater than the first preset threshold, the speech rate is slow, the slot interval is the first duration, and the information redundancy is marked as enabled. When the coupling coefficient is less than or equal to the first preset threshold and greater than the second preset threshold, the speech rate is medium-slow, the slot interval is the second duration, and the information redundancy mark is turned off. When the coupling coefficient is less than or equal to the second preset threshold and greater than or equal to the third preset threshold, the speech rate is the standard speed, the slot interval is the third duration, and the information redundancy mark is turned off. When the coupling coefficient is less than the third preset threshold and greater than or equal to the fourth preset threshold, the speech rate is medium-fast, the slot interval is the fourth duration, and the information redundancy mark is turned off. When the coupling coefficient is less than the fourth preset threshold, the speech rate is fast, the slot interval is the fifth duration, and the information redundancy mark is turned off. Among them, the first preset threshold, the second preset threshold, the third preset threshold, and the fourth preset threshold decrease sequentially, and the first duration, the second duration, the third duration, the fourth duration, and the fifth duration decrease sequentially.

7. The broadcast control method based on the coupling of cognitive load and information complexity as described in any one of claims 1-6, characterized in that, After controlling the virtual interactive carrier to execute the corresponding broadcast operation according to the broadcast strategy, the method further includes: Obtain the target user's feedback behavior data regarding this broadcast operation; Based on the feedback behavior data, the user cognitive load index is adaptively optimized.

8. A broadcast control system based on the coupling of cognitive load and information complexity, characterized in that, The system includes: The information complexity assessment module is used to acquire the text to be broadcast, evaluate the information structure of the text to be broadcast, and determine the information complexity index. The user cognitive load assessment module is used to acquire interaction behavior data between the target user and the virtual interactive carrier, and to conduct cognitive load assessment based on the interaction behavior data to determine the user cognitive load index. The coupling coefficient calculation module is used to determine the coupling coefficient based on the information complexity index and the user cognitive load index. The coupling coefficient is used to characterize the matching state between the information complexity index and the user cognitive load index. The broadcast rhythm generation and broadcast execution module is used to determine the broadcast strategy corresponding to the text to be broadcast based on the coupling coefficient, and control the virtual interactive carrier to perform the corresponding broadcast operation according to the broadcast strategy.

9. A computer device comprising a memory, a processor, and computer-readable instructions stored in the memory and running on the processor, characterized in that, When the processor executes the computer-readable instructions, it implements the broadcast control method based on the coupling of cognitive load and information complexity as described in any one of claims 1 to 7.

10. A readable storage medium having computer-readable instructions stored thereon, characterized in that, When the computer-readable instructions are executed by the processor, they implement the broadcast control method based on the coupling of cognitive load and information complexity as described in any one of claims 1 to 7.

Images