Advertisement putting effect intelligent evaluation method and system

Abstract

The invention is suitable for the technical field of advertisement putting evaluation, and provides an intelligent evaluation method and system for an advertisement putting effect, and the method comprises the steps: obtaining a current advertisement audio rhythm value of voice advertisement content, and adjusting the audio rhythm of the voice advertisement content when the current advertisement audio rhythm value is not in an optimal interval, and playing the adjusted advertisement audio rhythm value after the adjusted advertisement audio rhythm value falls into the optimal interval. Compared with the prior art, the method has the advantages that the influence of the advertisement is quantitatively evaluated from the aspect of movement behavior change, accurate optimization of the voice advertisement rhythm in the movement scene is realized, the matching degree between the advertisement content and the movement rhythm of the target user can be improved, and the user experience is improved. And the acceptability of the user to the advertisement information can be improved while the motion interference is reduced, so that the advertisement putting effect is effectively improved.

Description

Technical Field

[0001] This invention belongs to the field of advertising placement evaluation technology, and in particular relates to an intelligent evaluation method and system for advertising placement effectiveness. Background Technology

[0002] With the development of mobile internet technology, advertising delivery methods are gradually shifting from traditional static displays to intelligent delivery based on user behavior characteristics. In existing technologies, advertising delivery systems typically pre-evaluate advertisements before playback and adjust the content appropriately based on user profiles, device types, usage environments, and historical interaction data to improve advertising effectiveness. For example, existing technologies can adjust parameters such as ad duration, playback volume, voice playback speed, display brightness, playback timing, and playback frequency before playback, making the ad content more suitable for the user's usage scenario. Furthermore, in audio content platforms or sports applications, voice ads are often inserted as part of the audio content playback process. By pre-evaluating the ad content and adjusting appropriate parameters, ad reach and user acceptance can be improved to some extent.

[0003] However, existing technologies typically rely heavily on outcome metrics such as click-through rate, dwell time, and conversion rate for advertising effectiveness evaluation, or simply perform statistical analysis based on historical user behavior data. They pay little attention to real-time behavioral changes that occur during ad playback. This is particularly true in scenarios involving running and listening to audio through headphones. When users receive audio ads during exercise, their attention may shift between the activity and the ad content, impacting their performance, such as changes in cadence, pace, or rhythm. Current technologies often fail to fully utilize this behavioral change information for ad evaluation and lack mechanisms to assess real-time changes in user performance during exercise. This makes it difficult to effectively measure the match between the ad's rhythm and the user's current exercise rhythm.

[0004] Furthermore, those skilled in the art have discovered in their research that when users listen to audio advertisements through headphones while running, there is often a certain coupling relationship between the rhythm of the advertisement audio and the user's current exercise rhythm. When the rhythm of the advertisement audio does not match the user's exercise rhythm, the user's stride frequency may be significantly disturbed, and a certain recovery process will occur after the advertisement ends. However, existing technologies generally do not quantify this disturbance recovery process, nor do they have technical solutions to assess the degree of advertising impact from the perspective of stride frequency recovery, making it difficult to identify the adaptation relationship between the advertisement audio rhythm and the user's exercise rhythm. As a result, when delivering audio advertisements in exercise scenarios, the advertisement rhythm often cannot be dynamically optimized according to the user's current exercise state, thus affecting the user's listening experience and the effectiveness of advertising. Therefore, how to intelligently evaluate the advertising effect based on the characteristics of changes in user exercise rhythm in exercise scenarios, and optimize and adjust the advertisement audio rhythm accordingly, has become a technical problem that urgently needs to be solved in this field. Summary of the Invention

[0005] The purpose of this invention is to provide an intelligent evaluation method and system for advertising effectiveness, aiming to solve the problems mentioned in the background art.

[0006] The present invention is implemented as follows: an intelligent evaluation method for advertising delivery effect, the method includes: S1, when it is detected that voice advertising content is about to be delivered to a target user and the target user is in a state of listening with headphones and in motion, the target user's motion rhythm information is obtained and the target user's current motion rhythm value is determined;

[0007] S2. Based on the current movement rhythm value, select reference samples from historical advertising playback data. The background information of the advertising content corresponding to the reference sample is consistent with the current one, and the historical movement rhythm value corresponding to the reference sample is within the rhythm range corresponding to the current movement rhythm value. Furthermore, there is an identifiable step frequency disturbance and disturbance recovery process during the advertising playback.

[0008] S3. Obtain the advertising audio rhythm value and step frequency perturbation recovery value corresponding to each reference sample;

[0009] S4. Based on the correspondence between the advertising audio rhythm value and the step frequency disturbance recovery value, analyze the trend of the step frequency disturbance recovery value with the advertising audio rhythm value, and determine the optimal range of the advertising audio rhythm when there is an extreme point in the trend.

[0010] S5. Obtain the current audio rhythm value of the voice advertisement content. If it is not within the optimal range, adjust the audio rhythm of the voice advertisement content so that the adjusted audio rhythm value falls into the optimal range before playing it.

[0011] As a further limitation of the technical solution of this embodiment of the invention, the current movement rhythm value is calculated based on the movement rhythm information, which includes at least one of the following: step frequency information, movement cycle information, and acceleration change information during the target user's movement.

[0012] As a further limitation of the technical solution of the present invention, the background information of the advertising content corresponding to the reference sample being consistent with the background information of the current voice advertising content means that the advertising content corresponding to the reference sample and the voice advertising content are the same as or within a preset allowable range in at least one of the following: advertising type, advertising product category, advertising playback duration, and advertising voice expression form.

[0013] As a further limitation of the technical solution of the present invention, the advertising audio rhythm value is calculated based on the beat speed, rhythm intensity and audio energy of the advertising content audio. The beat speed, rhythm intensity and audio energy are normalized respectively, and weighted according to preset weights to obtain the advertising audio rhythm value.

[0014] As a further limitation of the technical solution of the present invention, the identifiable cadence disturbance and disturbance recovery process refers to the target user's cadence being in a stable state within a preset time period before the voice advertisement is played, the target user's real-time cadence shifting more than a preset threshold relative to the baseline cadence corresponding to the stable state during the voice advertisement playback, and recovering to the allowable fluctuation range corresponding to the baseline cadence within a preset time period after the voice advertisement playback ends.

[0015] As a further limitation of the technical solution of this embodiment of the invention, the process of obtaining the cadence perturbation recovery value includes: obtaining the baseline cadence within a preset time period before the voice advertisement is played; after the voice advertisement is played, detecting the time required for the target user's real-time cadence to re-enter the allowable fluctuation range corresponding to the baseline cadence, and determining the time as the cadence perturbation recovery value.

[0016] As a further limitation of the technical solution of this embodiment of the invention, step S4 specifically includes:

[0017] Several reference samples are sorted from smallest to largest according to the rhythm value of the advertisement audio to obtain a sample sequence;

[0018] Obtain the step frequency perturbation recovery value corresponding to each reference sample in the sample sequence, and extract the trend of step frequency perturbation recovery value with the advertising audio rhythm value based on the correspondence between the advertising audio rhythm value and the advertising audio rhythm value in the sample sequence;

[0019] Determine whether the trend of change shows a trend of first decreasing and then increasing. When the trend of change shows a trend of first decreasing and then increasing, determine the extreme point corresponding to the trend of change, and determine the optimal range of advertising audio rhythm within the preset range of the advertising audio rhythm value corresponding to the extreme point.

[0020] An intelligent evaluation system for advertising performance, the system comprising:

[0021] The motion rhythm acquisition module is used to acquire the motion rhythm information of the target user and determine the target user's current motion rhythm value when it detects that voice advertising content is about to be delivered to the target user and the target user is listening with headphones and in motion.

[0022] The reference sample filtering module is used to filter reference samples from historical ad playback data based on the current movement rhythm value. The background information of the ad content corresponding to the reference sample is consistent with the current one, and the historical movement rhythm value corresponding to the reference sample is within the rhythm range corresponding to the current movement rhythm value. Furthermore, there is an identifiable step frequency disturbance and disturbance recovery process during ad playback.

[0023] The parameter acquisition module is used to obtain the advertising audio rhythm value and step frequency perturbation recovery value corresponding to each reference sample;

[0024] The optimal interval determination module is used to analyze the trend of the change of the step frequency disturbance recovery value with the advertising audio rhythm value based on the correspondence between the advertising audio rhythm value and the step frequency disturbance recovery value, and determine the corresponding optimal interval of the advertising audio rhythm when there is an extreme point in the trend.

[0025] The audio rhythm adjustment module is used to obtain the current audio rhythm value of the voice advertisement content. When it is not in the optimal range, the module adjusts the audio rhythm of the voice advertisement content so that the adjusted audio rhythm value falls into the optimal range before playback.

[0026] As a further limitation of the technical solution of this embodiment of the invention, the current movement rhythm value is calculated based on the movement rhythm information, which includes at least one of the following: step frequency information, movement cycle information, and acceleration change information during the target user's movement.

[0027] As a further limitation of the technical solution of the present invention, the background information of the advertising content corresponding to the reference sample being consistent with the background information of the current voice advertising content means that the advertising content corresponding to the reference sample and the voice advertising content are the same as or within a preset allowable range in at least one of the following: advertising type, advertising product category, advertising playback duration, and advertising voice expression form.

[0028] Compared with the prior art, the present invention has the following beneficial effects:

[0029] This invention addresses the lack of evaluation of changes in athletic performance during running activities and when listening to audio advertisements through headphones. It proposes an intelligent evaluation method for advertising effectiveness based on the characteristics of athletic rhythm response. By introducing a new quantitative indicator—the cadence perturbation recovery value—the method uses the recovery process required for a target user to return to the baseline cadence within the allowable fluctuation range after ad playback as the evaluation criterion. Furthermore, it establishes a relationship between the two by combining the cadence perturbation recovery value with the ad audio rhythm value, identifying the extreme value patterns formed by the cadence perturbation recovery value as a function of the ad audio rhythm value, thereby determining the optimal range of the ad audio rhythm.

[0030] Based on this, the audio rhythm of the voice advertisement content is adaptively adjusted before the advertisement is played, so that the advertisement audio rhythm value falls within the optimal range. Compared with the prior art, this invention quantifies the impact of advertising from the perspective of changes in motion behavior, and achieves precise optimization of the rhythm of voice advertisements in motion scenarios. This not only improves the matching degree between the advertisement content and the target user's motion rhythm, but also increases the user's acceptance of the advertisement information while reducing motion interference, thereby effectively improving the advertising delivery effect. Attached Figure Description

[0031] Figure 1 A flowchart of the method provided in the embodiments of the present invention;

[0032] Figure 2 This is a flowchart illustrating the process of determining the optimal interval of advertising audio rhythm in the method provided in this embodiment of the invention;

[0033] Figure 3 The application architecture diagram of the system provided in the embodiments of the present invention. Detailed Implementation

[0034] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the invention.

[0035] Figure 1 A flowchart of the method provided by an embodiment of the present invention is shown.

[0036] Specifically, an intelligent evaluation method for advertising effectiveness includes the following steps:

[0037] Step S1: When it is detected that voice advertising content is about to be delivered to the target user and the target user is listening with headphones and in motion, obtain the target user's motion rhythm information and determine the target user's current motion rhythm value.

[0038] The current movement rhythm value is calculated based on the movement rhythm information, which includes at least one of the following: step frequency information, movement cycle information, and acceleration change information during the target user's movement.

[0039] In this embodiment of the invention, existing advertising delivery technologies, which pre-evaluate the advertising content to be played and adjust parameters before playback, are already relatively mature processing methods. Specifically, before detecting that advertising content is about to be pushed or played, existing technologies typically pre-adjust the presentation parameters of the advertising content based on the target user's basic profile, terminal type, historical interaction data, playback environment, and advertising delivery strategy to improve the reach efficiency and acceptance effect of the advertising content. The adjustment methods may include trimming or extending the advertising playback duration, adjusting the advertising brightness, adjusting the advertising volume, adjusting the advertising speech rate, delaying or advancing the advertising playback timing, and adjusting the intensity of dynamic effects on the advertising interface. For voice advertising content, existing technologies can also adjust the voice broadcast volume, broadcast speed, and background sound intensity based on the terminal device status, environmental noise level, or user's past preferences. Therefore, evaluating and optimizing parameters before advertising content playback has a mature technical foundation, which is also the source of the technical theme of this invention.

[0040] However, those skilled in the art have found in their research that while existing technologies can adjust advertising content before playback, the evaluation and optimization of voice advertising content remains insufficiently precise in specific scenarios where the target user is running and listening to audio through headphones. In particular, existing technologies focus more on the playback parameters of the advertising content itself and outcome metrics such as ad clicks, dwell times, and conversions, while rarely analyzing the real-time changes in user performance during exercise. In reality, when target users receive voice advertising content while running, their attention often shifts to the advertising content, resulting in changes in their performance, such as stride fluctuations, short-term pace changes, changes in arm swing rhythm, and changes in breathing rhythm. Among these, a typical and easily quantifiable change is the alteration of cadence. Furthermore, the acquisition and analysis of cadence during running is relatively mature in existing technologies and can be stably achieved through various terminal devices and sensors.

[0041] Furthermore, those skilled in the art have discovered that when target users listen to audio advertisements, their cadence shifts relative to the baseline cadence corresponding to their original stable state due to a change in attention; this is an objectively existing phenomenon. Rather than simply focusing on whether the cadence changes, this invention further focuses on the recovery process of this cadence change back to the allowable fluctuation range corresponding to the original baseline cadence after the advertisement playback ends. Research on a large amount of historical advertisement playback data has revealed a significant correlation between the cadence perturbation recovery value and the advertisement audio rhythm value, and this correlation is not monotonic. In other words, a larger or smaller advertisement audio rhythm value is not necessarily better; rather, there exists an optimal range of advertisement audio rhythm that is more suitable for the target user under a specific movement rhythm state. When the current audio rhythm value of the ad content is within the optimal range, although the target user will still experience gait frequency disturbances due to the perception of the ad content during ad playback, the recovery process required to return to the baseline gait frequency within the allowable fluctuation range is better, and the corresponding gait frequency disturbance recovery value is more in line with the target optimization direction. This indicates that the voice ad content is more compatible with the target user's current movement state, and the target user has a better acceptance of the ad content, thereby enabling intelligent evaluation and optimization of ad placement effectiveness.

[0042] In this embodiment of the invention, the condition that "voice advertising content is about to be delivered to the target user while the target user is listening through headphones and in motion" can be achieved through existing terminal collaborative detection mechanisms. Specifically, the system or terminal can first determine whether the voice advertising content is about to be pushed or played based on an advertising scheduling queue, an audio playback queue, an advertising insertion strategy, or preset advertising delivery triggering conditions. For example, when the target user is listening to content through a sports application, audio application, video application, or other applications with voice advertising insertion capabilities, when the advertising scheduling module recognizes that the advertising content is about to be inserted into the current audio playback stream within a preset time window, it can be determined that the condition of "voice advertising content is about to be delivered to the target user" is met.

[0043] Meanwhile, the headphone listening status can be determined through connection information between the terminal and the headphones, as well as playback link information. For example, if the terminal detects that a Bluetooth headphone, wired headphone, or other wearable audio output device is connected and the current audio output channel is pointing to the headphone, it can determine that the target user is listening to the headphones. Furthermore, in some embodiments, confirmation can also be made by combining headphone wearing detection results, such as based on headphone in-ear detection sensors, contact sensors, or wearing status feedback, to confirm that the headphones are actually being worn, thereby improving the accuracy of the determination.

[0044] To identify whether a target user is in motion, motion sensing data acquired by terminal or wearable devices can be used. For example, data from accelerometers, gyroscopes, inertial measurement units, or pedometer modules in mobile phones, smartwatches, fitness trackers, or smart headphones can be used to identify whether the target user is currently in continuous running, jogging, sprinting, or other motion states with stable periodic characteristics. When periodic acceleration fluctuations, gait rhythms, and corresponding body movement characteristics are detected continuously within a preset time window, it can be determined that the target user is in motion. In a preferred embodiment of the present invention, the target user is in a running motion state.

[0045] The motion rhythm information can originate from at least one of terminal devices, wearable devices, and headphone devices. Specifically, the motion rhythm information can be collected by sensors built into a smartphone, or by a smartwatch, fitness tracker, or smart headphones and then transmitted to a terminal or server for processing. Since running has a distinct periodicity, acquiring information on the target user's acceleration changes, angular velocity changes, and gait cycle information during exercise using inertial sensors, and then extracting cadence and motion cycle information, is a relatively mature technique in this field. This invention does not focus on a specific sensor structure as its innovation point, but rather, based on existing mature motion detection technologies, further applies the acquired motion rhythm information to the pre-evaluation of voice advertising content and the adjustment of audio rhythm.

[0046] The movement rhythm information includes at least one of the following: cadence information, movement cycle information, and acceleration change information during the target user's movement. Cadence information can be obtained by counting the number of steps completed by the target user per unit time. For example, within a preset time window, gait peaks are detected, the number of steps within that time window is obtained, and this is converted to steps per minute based on the time duration, thus forming cadence information. Movement cycle information can be obtained by identifying the time interval between two adjacent consecutive gait feature points. For example, the time difference between two adjacent landing peaks, two adjacent swing peaks, or other periodic feature points is identified, and this time difference is used as a single movement cycle. Further, the average or median of multiple movement cycles can be calculated as movement cycle information characterizing the current movement state. Acceleration change information can be obtained by acquiring the target user's triaxial acceleration data during running and extracting its amplitude changes, periodic fluctuations, peak-valley distributions, or combined acceleration change curves within a preset time period. The methods for acquiring the aforementioned cadence information, motion cycle information, and acceleration change information are all existing mature technologies in the fields of motion detection, gait analysis, and wearable device data processing. Those skilled in the art can choose one or more of them and combine them according to specific implementation requirements.

[0047] The current movement rhythm value is calculated based on the aforementioned movement rhythm information. Specifically, the current movement rhythm value can be obtained by directly using the number of steps per unit time based solely on cadence information; alternatively, it can be modified based on a combination of cadence information and movement cycle information; furthermore, it can be combined with acceleration change information to perform stability verification or confidence correction on the current movement rhythm value. For example, when the target user's current cadence change is detected to be relatively stable and has obvious periodic characteristics, the value corresponding to the cadence information can be directly used as the current movement rhythm value; when local fluctuations in cadence information are detected, cadence data within multiple consecutive time windows can be smoothed, averaged, averaged, or weighted by combining movement cycle information and acceleration change information to obtain a more stable current movement rhythm value. Therefore, the calculation process of the current movement rhythm value can be implemented using existing mature algorithms in this field. This invention does not limit the specific algorithm model, but rather uses the current movement rhythm value as an important basis for selecting reference samples and determining the optimal interval for advertising audio rhythm.

[0048] Furthermore, since different target users exhibit varying rhythmic states at different running stages, the system can map the current rhythmic value to a preset rhythm range. For example, the current rhythmic value can be divided into slow-paced, medium-paced, and fast-paced running intervals, or further subdivided into multiple finer-grained rhythm segments. When selecting reference samples from historical ad playback data, only samples with historical rhythmic values ​​falling within the rhythm range corresponding to the current rhythmic value are chosen. This ensures comparability of the samples in terms of movement states and improves the accuracy of subsequent analysis of the correlation between ad audio rhythmic values ​​and cadence disturbance recovery values.

[0049] It should be noted that the acquisition of motion rhythm information, extraction of cadence information, identification of motion cycle information, and extraction of acceleration change information by the target user listening to headphones and in motion can all be achieved using existing mature technologies. The improvement of this invention does not lie in reconstructing a completely new motion detection method, but in discovering and utilizing the following pattern: In the scenario of running and listening to voice advertisements through headphones, there is a non-monotonic correlation between the advertisement audio rhythm value and the cadence perturbation recovery value. Based on this correlation, the optimal range of advertisement audio rhythm is determined, thereby adjusting the audio rhythm of the voice advertisement content before the advertisement is played, making the advertisement content more suitable for the target user's current motion state, and thus improving the intelligent evaluation and optimization level of the advertisement's effectiveness.

[0050] Furthermore, the intelligent evaluation method for advertising effectiveness also includes the following steps:

[0051] Step S2: Based on the current movement rhythm value, select reference samples from the historical advertising playback data. The background information of the advertising content corresponding to the reference sample is consistent with the current one, and the historical movement rhythm value corresponding to the reference sample is within the rhythm range corresponding to the current movement rhythm value. Furthermore, there is an identifiable step frequency disturbance and disturbance recovery process during the advertising playback.

[0052] The background information of the advertising content corresponding to the reference sample is consistent with the background information of the current voice advertising content, which means that the advertising content corresponding to the reference sample and the voice advertising content are the same as or within a preset allowable range in at least one of the following: advertising type, advertising product category, advertising playback duration, and advertising voice expression form.

[0053] The identifiable cadence disturbance and disturbance recovery process refers to the target user's cadence being in a stable state within a preset time period before the voice advertisement is played, the target user's real-time cadence shifting more than a preset threshold relative to the baseline cadence corresponding to the stable state during the voice advertisement playback, and recovering to the allowable fluctuation range corresponding to the baseline cadence within a preset time period after the voice advertisement playback ends.

[0054] In this embodiment of the invention, the main purpose of step S2 is to filter out reference samples from historical ad playback data that can form an effective comparative relationship with the current delivery scenario, so as to use them for subsequent analysis of the changing patterns between the ad audio rhythm value and the step frequency perturbation recovery value. Since the target user's movement performance varies significantly under different ad content, user states, and movement rhythm conditions, directly using all historical ad playback data as the analysis object can easily introduce a large amount of noise data unrelated to the current delivery context, thereby affecting the accuracy of identifying the optimal interval of the ad audio rhythm. Therefore, by filtering historical ad playback data and constructing a set of reference samples that matches the current voice ad content and the target user's movement state, the reliability of subsequent trend analysis and extreme value interval identification can be effectively improved.

[0055] Firstly, the background information of the advertisement content corresponding to the reference sample must be consistent with or within a preset allowable range of the background information of the current audio advertisement content. This is to ensure the comparability of advertisement content attributes between different samples. Factors such as the type of advertisement content, product category, playback duration, and voice expression style can all affect the target user's attention allocation and perception patterns when listening to advertisements. For example, different types of advertisements differ significantly in voice structure, speech rate, and information density, which may affect the target user's movement rhythm to varying degrees. Therefore, by limiting the background information of the advertisement content, ensuring that the reference sample and the current audio advertisement content are identical or within a preset allowable range in at least one of the following aspects—advertisement type, product category, playback duration, and voice expression style—interference caused by differences in advertisement content can be reduced. This allows subsequent analysis to focus more on the impact of the advertisement audio rhythm value itself on the step frequency perturbation recovery value.

[0056] In some implementations, to improve the effectiveness of the reference samples, stricter screening criteria can be employed to ensure a high degree of similarity between the reference samples and the current delivery context. For example, in addition to ensuring consistency or similarity in the background information of the advertising content, it can be further restricted that the user's movement rhythm value during historical ad playback must be within the rhythm range corresponding to the current movement rhythm value, thereby ensuring the comparability of the target user's movement state in different samples. Furthermore, in some implementations, the reference samples can be further screened by combining factors such as user movement type, user device type, advertising playback environment, and user's historical movement habits. For example, only historical ad playback records generated during running can be selected, or only historical ad playback records generated in a headphone audio playback environment can be selected. Through the above screening methods, the representativeness of the reference samples can be further improved, making the subsequent analysis of the relationship between advertising audio rhythm values ​​and step frequency perturbation recovery values ​​more accurate.

[0057] It should be noted that in practical applications, the background information of the reference samples does not necessarily need to be completely identical. In some cases, overly strict screening criteria may result in an insufficient number of usable samples, thus affecting the stability of subsequent trend analysis. Therefore, in this embodiment of the invention, it is permissible for the advertising content corresponding to the reference sample to differ somewhat from the current voice advertising content in some background information, as long as the difference is within a preset allowable range. For example, even slight differences in advertising playback duration, voice expression, or product category can still be used as reference samples. By introducing a preset allowable range, the comparability of samples can be ensured while also considering the number and diversity of samples, thereby improving the reliability of the overall analysis results.

[0058] Furthermore, in this embodiment of the invention, it is also required that the reference sample exhibits identifiable cadence perturbations and recovery processes during the advertisement playback. This is because not all advertisement playback records significantly impact the target user's movement rhythm. If, in certain historical records, the target user maintains a stable cadence before and after the advertisement playback, such samples cannot reflect the actual impact of the advertisement audio rhythm value on changes in movement rhythm, and therefore are unsuitable as reference samples. By selecting only historical records exhibiting identifiable cadence perturbations and recovery processes, it can be ensured that the selected reference samples truly reflect the impact of advertisement playback on the target user's movement rhythm, thereby improving the effectiveness of subsequent analysis.

[0059] Specifically, in this embodiment of the invention, the identifiable cadence disturbance and recovery process refers to the following: During a preset time period before the audio advertisement plays, the target user's cadence remains stable; during the audio advertisement plays, the target user's real-time cadence deviates from the baseline cadence corresponding to the stable state by more than a preset threshold; and within a preset time period after the audio advertisement plays, the cadence recovers to the allowable fluctuation range corresponding to the baseline cadence. The baseline cadence can be obtained by statistically analyzing cadence data within the preset time period before the audio advertisement plays, for example, using the average, median, or weighted average. The preset threshold is used to determine whether a significant disturbance has occurred in the cadence, while the allowable fluctuation range is used to determine whether the cadence has recovered to a stable state. Through this method, the cadence change process in historical advertisement playback records can be identified, thereby filtering out truly valuable sample data.

[0060] Through the screening process in step S2, a set of reference samples with high similarity to the current delivery scenario in terms of advertising content attributes, user movement rhythm state, and cadence perturbation characteristics can be obtained. This provides a reliable data foundation for subsequent analysis of the changing trend between advertising audio rhythm value and cadence perturbation recovery value, as well as for determining the optimal range of advertising audio rhythm.

[0061] Furthermore, the intelligent evaluation method for advertising effectiveness also includes the following steps:

[0062] Step S3: Obtain the advertising audio rhythm value and step frequency perturbation recovery value corresponding to each reference sample.

[0063] The advertising audio rhythm value is calculated based on the beat speed, rhythm intensity, and audio energy of the advertising content audio. The beat speed, rhythm intensity, and audio energy are normalized and then weighted according to preset weights to obtain the advertising audio rhythm value.

[0064] The process of obtaining the cadence perturbation recovery value includes: obtaining the baseline cadence within a preset time period before the voice advertisement is played; after the voice advertisement is played, detecting the time required for the target user's real-time cadence to re-enter the allowable fluctuation range corresponding to the baseline cadence, and determining the time as the cadence perturbation recovery value.

[0065] In this embodiment of the invention, the purpose of step S3 is to provide a quantitative indicator for subsequent analysis of the relationship between the advertising audio rhythm value and the cadence perturbation recovery value. Specifically, it is necessary to obtain the corresponding advertising audio rhythm value and cadence perturbation recovery value for each reference sample to form data pairs that can be used for trend analysis.

[0066] The calculation of the advertising audio rhythm value can be based on existing audio signal processing technologies. Specifically, audio feature parameters such as beat speed, rhythm intensity, and audio energy can be extracted from the advertising audio content. Beat speed can be obtained through beat detection algorithms, such as identifying periodic beat peaks in the audio signal to determine the number of beats per unit time; rhythm intensity can be obtained by analyzing the amplitude of audio rhythm changes or the prominence of beats; and audio energy can be obtained by calculating the energy amplitude of the audio signal within a preset time window. These parameters are all common feature indicators in the field of audio analysis. In this embodiment of the invention, beat speed, rhythm intensity, and audio energy can be normalized separately and weighted according to preset weights to obtain the corresponding advertising audio rhythm value. Since the above audio feature extraction and normalization methods are all existing mature technologies, this invention does not limit the specific audio feature extraction algorithm.

[0067] Unlike the audio tempo value of an advertisement, the cadence perturbation recovery value characterizes the recovery process of a target user's movement rhythm back to its original stable state after being affected by the advertisement content during ad playback. This indicator quantifies the effectiveness of ad placement from the perspective of changes in movement behavior. In this embodiment of the invention, the cadence data of the target user within a preset time period before the audio advertisement playback is first acquired, and a baseline cadence is obtained through statistical processing to characterize the stable movement rhythm of the target user when not disturbed by the advertisement. Subsequently, after the audio advertisement playback ends, the real-time cadence changes of the target user are continuously monitored. When the real-time cadence re-enters the allowable fluctuation range corresponding to the baseline cadence, the time elapsed from the end of the advertisement playback to the recovery to this allowable fluctuation range is recorded, and this time is determined as the cadence perturbation recovery value.

[0068] The above method quantifies the recovery process required for target users to return to a stable exercise rhythm after an advertisement is played. Compared with traditional analysis methods that only focus on whether cadence changes, this invention introduces the cadence perturbation recovery value as an indicator to characterize the impact of advertising on the user's exercise state from the perspective of cadence recovery process. This measurement method is rarely used in advertising effectiveness evaluation scenarios in the field, and can more objectively reflect the impact of advertising audio rhythm on the user's attention shift and recovery process, providing a new analytical dimension for subsequent identification of the optimal range of advertising audio rhythm.

[0069] Furthermore, the intelligent evaluation method for advertising effectiveness also includes the following steps:

[0070] Step S4: Based on the correspondence between the advertising audio rhythm value and the step frequency disturbance recovery value, analyze the trend of the step frequency disturbance recovery value with the advertising audio rhythm value, and determine the optimal range of the advertising audio rhythm when there is an extreme point in the trend.

[0071] S5. Obtain the current audio rhythm value of the voice advertisement content. If it is not within the optimal range, adjust the audio rhythm of the voice advertisement content so that the adjusted audio rhythm value falls into the optimal range before playing it.

[0072] Specifically, Figure 2 A flowchart is shown to determine the optimal interval for the advertising audio rhythm.

[0073] Step S4 specifically includes the following steps:

[0074] Step S51: Sort several reference samples according to the advertising audio rhythm value from small to large to obtain a sample sequence;

[0075] Step S52: Obtain the step frequency perturbation recovery value corresponding to each reference sample in the sample sequence, and extract the trend of step frequency perturbation recovery value with the advertising audio rhythm value according to the correspondence between the advertising audio rhythm value and the step frequency perturbation recovery value in the sample sequence.

[0076] Step S53: Determine whether the change trend shows a trend of first decreasing and then increasing. When the change trend shows a trend of first decreasing and then increasing, determine the extreme point corresponding to the change trend, and determine the optimal range of advertising audio rhythm within the preset range of the advertising audio rhythm value corresponding to the extreme point.

[0077] In this embodiment of the invention, step S4 serves to analyze the relationship between the advertising audio rhythm value and the step frequency perturbation recovery value based on the acquired reference sample data, thereby identifying the advertising audio rhythm range that is more suitable for the target user in their current movement rhythm state. Through this step, the influence of the advertising audio rhythm value on the user's movement rhythm recovery can be extracted from historical samples, thus providing a basis for optimizing the audio rhythm of subsequent voice advertising content.

[0078] Specifically, in step S51, several reference samples are sorted from smallest to largest according to the advertising audio rhythm value to obtain a sample sequence. Through this sorting process, a sample arrangement with the advertising audio rhythm value as the horizontal variable can be formed, so that different reference samples show a continuous changing relationship in the dimension of advertising audio rhythm value, thereby facilitating subsequent analysis of the overall trend of the step frequency perturbation recovery value changing with the advertising audio rhythm value.

[0079] In step S52, the cadence perturbation recovery value corresponding to each reference sample in the sample sequence is obtained, and the trend of the cadence perturbation recovery value changing with the advertising audio rhythm value is extracted based on the correspondence between the advertising audio rhythm value and the cadence perturbation recovery value in the sample sequence. Through this step, a functional relationship or change relationship between the advertising audio rhythm value and the cadence perturbation recovery value can be established, which can be used to observe the changing pattern of the target user's movement rhythm recovery under different advertising audio rhythm conditions.

[0080] In step S53, it is determined whether the trend of change shows a decreasing-then-increasing trend. When the trend shows a decreasing-then-increasing trend, it indicates that as the rhythm value of the advertising audio changes, the step frequency disturbance recovery value will first decrease and then increase, thus forming an extreme point at a certain position. The advertising audio rhythm value corresponding to this extreme point usually represents the rhythm position where the advertising has the least impact on the target user's movement rhythm or the best recovery effect under the current movement rhythm state. Therefore, the optimal range of advertising audio rhythm is determined within the preset range of the advertising audio rhythm value corresponding to the extreme point. In this way, the range of advertising audio rhythms that are more suitable for the target user under the current movement rhythm conditions can be identified.

[0081] The processing of step S4 and its sub-steps described above actually verifies the aforementioned research findings, namely, that when target users are running and listening to audio advertisements through headphones, there is a non-monotonic relationship between their cadence recovery process and the rhythm of the advertisement audio. By sorting and trend analysis of historical reference samples, the extreme points in this non-monotonic relationship can be identified, thereby determining a more suitable advertisement audio rhythm range for the current exercise state. This process not only verifies the target user's response pattern to the advertisement rhythm during exercise but also provides data support for subsequent advertisement placement optimization.

[0082] In this embodiment of the invention, determining the optimal interval for the advertising audio rhythm is of great significance. By identifying this optimal interval, the audio advertising content to be delivered can be rhythmically adapted to the target user's current movement rhythm before playback. When the advertising audio rhythm value falls within this optimal interval, it indicates that the advertising rhythm has a high degree of matching with the target user's current movement rhythm, resulting in minimal interference with the target user's movement rhythm during playback. Users can maintain a good movement state while listening to the advertising content and effectively focus on the advertising information, thereby improving advertising acceptance and overall delivery effectiveness.

[0083] In step S5, the current audio rhythm value of the voice advertisement content is first obtained, and it is determined whether the audio rhythm value is within the optimal range. If the current audio rhythm value is already within the optimal range, the original audio content can be played directly without any additional adjustments. If the current audio rhythm value is not within the optimal range, the audio rhythm of the voice advertisement content needs to be adjusted so that the adjusted audio rhythm value falls within the optimal range before playback.

[0084] In practice, adjusting the audio rhythm can be achieved by modifying various audio features that affect the advertising audio rhythm value. For example, the tempo of the advertising audio can be appropriately increased or decreased by changing the speech speed or background tempo to bring the overall tempo closer to the optimal range. The rhythm intensity can also be adjusted by modifying the prominence of the audio beats or the amplitude of rhythm changes to make the rhythm performance closer to the optimal rhythm range. Furthermore, the audio energy can be appropriately adjusted, such as by adjusting the energy of the background music or the volume structure of the speech, to change the overall sense of rhythm. When the advertising audio rhythm value is higher than the optimal range, it can be reduced by appropriately decreasing the tempo, weakening the rhythm intensity, or lowering the audio energy. When the advertising audio rhythm value is lower than the optimal range, it can be compensated for by appropriately increasing the tempo, strengthening the rhythm intensity, or increasing the audio energy. By adjusting the above audio parameters, the advertising audio rhythm value can gradually approach the optimal range near the extreme point, thereby achieving adaptive optimization of the advertising rhythm.

[0085] Through the above technical solution, this invention enables intelligent evaluation and optimization of the rhythm of advertising audio for the specific application scenario of running while listening to voice advertisements through headphones. Compared to traditional methods that only evaluate based on ad click-through rate or user dwell time, this invention introduces a new evaluation index—step frequency perturbation recovery value—from the perspective of the exercise rhythm recovery process, and combines it with the advertising audio rhythm value for trend analysis. This allows for a more objective reflection of the impact of advertising content on the user's attention shift and recovery process. By identifying the optimal range of advertising audio rhythm and dynamically adjusting the advertising audio rhythm, the advertising content can be better adapted to the user's current exercise state, increasing the user's acceptance of the advertising content while reducing the interference of advertising playback on the user's exercise experience.

[0086] Therefore, this invention not only verifies and utilizes the regularity exhibited by the changes and recovery process of cadence in sports users while listening to voice advertisements, but also enables intelligent optimization of advertisement content based on this regularity. This effectively addresses the core issue raised in the aforementioned research: how to more reasonably evaluate and optimize the effectiveness of voice advertising in the scenario of running while listening through headphones. This technical solution has promising application prospects and can be widely used in sports application platforms, smart wearable device platforms, audio content platforms, and mobile advertising systems, improving both advertising effectiveness and the overall user experience for sports users.

[0087] Furthermore, Figure 3 An application architecture diagram of the system provided in an embodiment of the present invention is shown.

[0088] In another preferred embodiment of the present invention, an intelligent evaluation system for advertising effectiveness includes:

[0089] The motion rhythm acquisition module 100 is used to acquire the motion rhythm information of the target user and determine the current motion rhythm value of the target user when it is detected that voice advertising content is about to be delivered to the target user and the target user is listening with headphones and in motion.

[0090] The movement rhythm value is calculated based on the movement rhythm information, which includes at least one of the following: step frequency information, movement cycle information, and acceleration change information during the target user's movement.

[0091] Furthermore, the intelligent evaluation system for advertising performance also includes:

[0092] The reference sample filtering module 200 is used to filter reference samples from historical advertising playback data based on the current movement rhythm value. The background information of the advertising content corresponding to the reference sample is consistent with the current one, and the historical movement rhythm value corresponding to the reference sample is within the rhythm range corresponding to the current movement rhythm value. Furthermore, there is an identifiable step frequency disturbance and disturbance recovery process during the advertising playback.

[0093] The background information of the advertising content corresponding to the reference sample is consistent with the background information of the current voice advertising content, which means that the advertising content corresponding to the reference sample and the voice advertising content are the same as or within a preset allowable range in at least one of the following: advertising type, advertising product category, advertising playback duration, and advertising voice expression form.

[0094] Furthermore, the intelligent evaluation system for advertising performance also includes:

[0095] The parameter acquisition module 300 is used to acquire the advertising audio rhythm value and step frequency perturbation recovery value corresponding to each reference sample.

[0096] Furthermore, the intelligent evaluation system for advertising performance also includes:

[0097] The optimal interval determination module 400 is used to analyze the trend of the change of the step frequency disturbance recovery value with the advertising audio rhythm value based on the correspondence between the advertising audio rhythm value and the step frequency disturbance recovery value, and to determine the corresponding optimal interval of the advertising audio rhythm when there is an extreme point in the trend.

[0098] Furthermore, the intelligent evaluation system for advertising performance also includes:

[0099] The audio rhythm adjustment module 500 is used to obtain the current audio rhythm value of the voice advertisement content. When it is not in the optimal range, the audio rhythm of the voice advertisement content is adjusted so that the adjusted audio rhythm value falls into the optimal range before playback.

[0100] It should be understood that although the steps in the flowcharts of the various embodiments of the present invention are shown sequentially according to the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless explicitly stated herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least some steps in the various embodiments may include multiple sub-steps or multiple stages. These sub-steps or stages are not necessarily completed at the same time, but can be executed at different times. The execution order of these sub-steps or stages is not necessarily sequential, but can be performed alternately or in turn with other steps or at least a portion of the sub-steps or stages of other steps.

[0101] 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 a computer program instructing related hardware. The program can be stored in a non-volatile computer-readable storage medium, and when executed, it can include the processes of the embodiments of the above methods. 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 can include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory. Volatile memory can include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in various 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.

[0102] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0103] The embodiments described above are merely illustrative of several implementations of the present invention, and while the descriptions are specific and detailed, they should not be construed as limiting the scope of the present invention. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of the present invention, and these modifications and improvements all fall within the scope of protection of the present invention. Therefore, the scope of protection of this patent should be determined by the appended claims.

[0104] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A method for intelligently evaluating the effectiveness of advertising campaigns, characterized in that, The method includes: S1. When it is detected that voice advertising content is about to be delivered to the target user and the target user is listening with headphones and in motion, obtain the target user's motion rhythm information and determine the target user's current motion rhythm value. S2. Based on the current movement rhythm value, select reference samples from historical advertising playback data. The background information of the advertising content corresponding to the reference sample is consistent with the current one, and the historical movement rhythm value corresponding to the reference sample is within the rhythm range corresponding to the current movement rhythm value. Furthermore, there is an identifiable step frequency disturbance and disturbance recovery process during the advertising playback. S3. Obtain the advertising audio rhythm value and step frequency perturbation recovery value corresponding to each reference sample; S4. Based on the correspondence between the advertising audio rhythm value and the step frequency disturbance recovery value, analyze the trend of the step frequency disturbance recovery value with the advertising audio rhythm value, and determine the optimal range of the advertising audio rhythm when there is an extreme point in the trend. S5. Obtain the current audio rhythm value of the voice advertisement content. If it is not within the optimal range, adjust the audio rhythm of the voice advertisement content so that the adjusted audio rhythm value falls into the optimal range before playing it.

2. The intelligent evaluation method for advertising effectiveness according to claim 1, characterized in that, The current movement rhythm value is calculated based on the movement rhythm information, which includes at least one of the following: step frequency information, movement cycle information, and acceleration change information during the target user's movement.

3. The intelligent evaluation method for advertising effectiveness according to claim 1, characterized in that, The background information of the advertising content corresponding to the reference sample is consistent with the background information of the current voice advertising content, which means that the advertising content corresponding to the reference sample and the voice advertising content are the same as or within a preset allowable range in at least one of the following: advertising type, advertising product category, advertising playback duration, and advertising voice expression form.

4. The intelligent evaluation method for advertising effectiveness according to claim 1, characterized in that, The advertising audio rhythm value is calculated based on the beat speed, rhythm intensity, and audio energy of the advertising content audio. The beat speed, rhythm intensity, and audio energy are normalized and then weighted according to preset weights to obtain the advertising audio rhythm value.

5. The intelligent evaluation method for advertising effectiveness according to claim 1, characterized in that, The identifiable cadence disturbance and disturbance recovery process refers to the target user's cadence being in a stable state within a preset time period before the voice advertisement is played, the target user's real-time cadence shifting more than a preset threshold relative to the baseline cadence corresponding to the stable state during the voice advertisement playback, and recovering to the allowable fluctuation range corresponding to the baseline cadence within a preset time period after the voice advertisement playback ends.

6. The intelligent evaluation method for advertising effectiveness according to claim 5, characterized in that, The process of obtaining the cadence perturbation recovery value includes: obtaining the baseline cadence within a preset time period before the voice advertisement is played; after the voice advertisement is played, detecting the time required for the target user's real-time cadence to re-enter the allowable fluctuation range corresponding to the baseline cadence, and determining the time as the cadence perturbation recovery value.

7. The intelligent evaluation method for advertising effectiveness according to claim 1, characterized in that, Step S4 specifically includes: Several reference samples are sorted from smallest to largest according to the rhythm value of the advertisement audio to obtain a sample sequence; Obtain the step frequency perturbation recovery value corresponding to each reference sample in the sample sequence, and extract the trend of step frequency perturbation recovery value with the advertising audio rhythm value based on the correspondence between the advertising audio rhythm value and the advertising audio rhythm value in the sample sequence; Determine whether the trend of change shows a trend of first decreasing and then increasing. When the trend of change shows a trend of first decreasing and then increasing, determine the extreme point corresponding to the trend of change, and determine the optimal range of advertising audio rhythm within the preset range of the advertising audio rhythm value corresponding to the extreme point.

8. An intelligent evaluation system for advertising effectiveness, characterized in that, The system includes: The motion rhythm acquisition module is used to acquire the motion rhythm information of the target user and determine the target user's current motion rhythm value when it detects that voice advertising content is about to be delivered to the target user and the target user is listening with headphones and in motion. The reference sample filtering module is used to filter reference samples from historical ad playback data based on the current movement rhythm value. The background information of the ad content corresponding to the reference sample is consistent with the current one, and the historical movement rhythm value corresponding to the reference sample is within the rhythm range corresponding to the current movement rhythm value. Furthermore, there is an identifiable step frequency disturbance and disturbance recovery process during ad playback. The parameter acquisition module is used to obtain the advertising audio rhythm value and step frequency perturbation recovery value corresponding to each reference sample; The optimal interval determination module is used to analyze the trend of the change of the step frequency disturbance recovery value with the advertising audio rhythm value based on the correspondence between the advertising audio rhythm value and the step frequency disturbance recovery value, and determine the corresponding optimal interval of the advertising audio rhythm when there is an extreme point in the trend. The audio rhythm adjustment module is used to obtain the current audio rhythm value of the voice advertisement content. When it is not in the optimal range, the module adjusts the audio rhythm of the voice advertisement content so that the adjusted audio rhythm value falls into the optimal range before playback.

9. The intelligent evaluation system for advertising effectiveness according to claim 8, characterized in that, The movement rhythm value is calculated based on the movement rhythm information, which includes at least one of the following: step frequency information, movement cycle information, and acceleration change information during the target user's movement.

10. The intelligent evaluation system for advertising effectiveness according to claim 8, characterized in that, The background information of the advertising content corresponding to the reference sample is consistent with the background information of the current voice advertising content, which means that the advertising content corresponding to the reference sample and the voice advertising content are the same as or within a preset allowable range in at least one of the following: advertising type, advertising product category, advertising playback duration, and advertising voice expression form.

Images