Content generation system and method

An automated content generation system uses multiple modules to aggregate and weight feedback for optimizing audio-visual content, addressing inefficiencies and bias in traditional methods, ensuring high user engagement and adherence to boundaries.

WO2026024181A1PCT designated stage Publication Date: 2026-01-29HEINEKEN SUPPLY CHAIN BV
View PDF 4 Cites 0 Cited by

Patent Information

Application Number
PCT/NL2025/050361
Authority / Receiving Office
WO · WO
Patent Type
Applications
Current Assignee / Owner
Priority Date
2024-07-25
Filing Date
2025-07-24
Publication Date
2026-01-29

AI Technical Summary

Technical Problem

The generation of audio-visual content that appeals to consumers is time-consuming and resource-intensive, often involving significant market research and is prone to human bias.

Method used

An automated electronic content generation system that operates automatically or nearly automatically, utilizing multiple content generation modules, aggregating feedback data to adjust parameters and share best practices, and incorporating weightings to prioritize positive feedback for improved content creation.

Benefits of technology

Facilitates efficient and bias-reduced content generation by leveraging feedback to iteratively improve content creation, ensuring high user engagement while adhering to boundaries and audience preferences.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure NL2025050361_29012026_PF_FP_ABST
    Figure NL2025050361_29012026_PF_FP_ABST
Patent Text Reader

Abstract

A method is provided by generating audio-visual content data, for example advertisements, by multiple electronically coded content generation modules. The audio-visual data objects are generated based on an initial dataset, with parameters having pre-defined values. The values may be changed in the course of time. Feedback may be received from an audience on appreciation of the content item. Based on the feedback, parameters of one or more content generation modules may be changed. In this way, best practices may be shared among multiple content generation modules, by updating parameters based on parameters of content generation modules that receive the highest feedback scores. As such, the content generation modules may operate as a swarm.
Need to check novelty before this filing date? Find Prior Art

Description

[0001] Title: Content generation system and method

[0002] TECHNICAL FIELD

[0003] The various aspects and examples thereof relate to distributed generation of content, based on feedback provided by an audience.

[0004] BACKGROUND

[0005] Generation of audio-visual content that is appreciated by consumers may be an elaborate task. For example, this may required significant market research, of which in turn the results need to be interpreted for generating content. Firstly, such process costs a lot of time and resources. Second, executed by human beings, bias may be introduced.

[0006] SUMMARY

[0007] It is preferred to provide an automated electronic content generation system and method, that is able to operate automatically or almost fully automatic to generated electronic content, for example for advertisement campaigns on electronic media.

[0008] A first aspect provides, in a first electronic data processing system, a method of managing generation of audio-visual data. The method comprises operating, on the electronic data processing device, a first electronically coded content generation module configured to generate a first audio- visual data object based on a first set of parameters comprised by the first electronically coded content generation module, operating, on the electronic data processing device, a second electronically coded content generation module configured to generate a second audio- visual data object based on a second set of parameters comprised by the second electronically coded content generation module, receiving, by the electronic data processing device, a first electronic signal comprising feedback data on the first audio-visual data object, determining an adjustment of a first value of a first parameter of the first set of parameters to be adjusted in accordance with the feedback data, updating the first value of the first parameter based on the feedback data, sending, by means of a second electronic signal, data related to the first updated parameter to a second electronically coded content generation module, updating a second value of a second parameter comprised by the second set of parameters based on the updated first value of the first parameter, the second parameter corresponding to the first parameter; and generating the second audio- visual data object based on the second set of parameters the second parameter having the updated value.

[0009] Different content generation modules may operate in different ways and provide different audio- visual data objects and different audiovisual data objects may be presented to different audiences. As a result, feedback based on each of the audio-visual data objects may be different. This method provides sharing of best practices among electronically coded content generation modules.

[0010] An example of the first aspect further comprises operating, on the electronic data processing device, an electronically coded control module arranged to aggregate feedback data from multiple electronically coded content generation modules operated on electronic data processing device, wherein the data related to the first updated parameter is sent to the electronically coded control module; and the method further comprises generating aggregate feedback data based on the data related to the first updated parameter received from the first electronically coded content generation module and providing the aggregate feedback data on the first parameter to the second electronically coded content generation module.

[0011] This example allows for joint aggregation of feedback and, based on the feedback, provide joint learning to benefit all electronically coded content generation modules by sharing updated parameters.

[0012] In another example, the aggregated feedback data comprises weighting data providing an indication how the aggregated feedback data is to be weighted relative to at least one of feedback data on the audio-visual data object and the aggregated feedback data for adjusting at least one parameter of the first content generation module. Depending on various parameters, particular feedback may affect particular parameters more than others. Hence, based on feedback, particular weight may be assigned. Second, in some cases, updated parameter values may be used to provide new parameter values for the applicable electronically coded content generation module, in addition to existing parameter values within the applicable electronically coded content generation module. Weighing factors may be provided to attach a weight to internal and updated parameter values to determine new values to be used by the applicable electronically coded content generation module.

[0013] In a further example, the aggregated feedback data comprises one of a value of a parameter of the second data set that is to be adjusted and a feedback score. The processing to provide the new parameter values may be executed by the central entity or by the applicable electronically coded content generation module.

[0014] In again a further example, a further multitude of electronically coded content generation modules are operated on the electronic data processing device; and the electronically coded control module is arranged to determine which of the multiple content generation modules receives the highest feedback score. The feedback score comprises one or more values based on feedback of an audience on the audio- visual data object generated by a particular electronically coded content generation module. By determining which electronically coded content generation module receives the highest score, an electronically coded content generation module performing best with respect to content generation may be determined. Next, data related to that electronically coded content generation module may be shared. In yet a further example, feedback data of the further multitude of electronically coded content generation modules that receives the highest feedback score has a weight in the aggregated feedback data higher than feedback data of other content generation modules. This allows for fast and efficient iteration towards generation of audio- visual data objects that trigger high user engagement, which audio- visual data objects are generated by more electronically coded content generation module using updated data sets.

[0015] Yet another example further comprises obtaining boundary data providing boundary conditions for content of the audio-visual data, determining whether the content generated module, with parameter to be adjusted having an adjusted value, complies with the boundary conditions, and updating, if the content generation module complies with the boundary conditions with the parameter having the adjusted value, the at least one parameter of the content generation module based on the feedback data.

[0016] By merely relying on feedback data from an audience, content generation based on such feedback can go beyond boundaries of good manners and taste. This may be the case if audio- visual data objects are posted to media having a very diverse audience, like particular nonmainstream social media or other media, for example on-line newspapers and magazines. If an audio-visual data object with content touching any of the boundaries strongly appeals to a small group, such small group may, despite a small size, still have a big impact on feedback. Such feedback may result on parameters for generation of audio- visual data objects pushing the next generated audio- visual data objects over the edge of good manners - or even worse, relate to illegal content. This example aims to prevent that or at least reduce the risk of such scenario happening.

[0017] Again another example further comprises sending the first audiovisual data object to an electronic computing system hosting an electronic content publishing service, for publication of the first audio-visual data object to a first audience, wherein the first electronic signal is received from the electronic computing system and feedback data is based on feedback of the first audience on the first audio- visual data object. Such example is particularly, but not exclusively, useful for receiving feedback from social media platforms.

[0018] Again a further example further comprises sending the second audio- visual data object to an electronic computing system hosting an electronic content publishing service, for publication of the second audiovisual data object to a second audience. This example allows for obtaining parameter values for creating audio- visual data objects that are appreciated by multiple audiences. The first audience may be different from the second audience. Whereas they may be different, the audiences may still overlap.

[0019] In another example, the first electronically coded content generation module is arranged to receive central content data from a central content repository; and generate the audio- visual data object based on the central content data. In certain scenarios, values of parameters of the electronically coded content generation module may be reset. Such may be the case of a new advertisement campaign, but also to prevent content of audio- visual data objects to wander off too far from initially intended content, a central update may be advantageous.

[0020] In yet a further example, data related to the updated parameters is sent to the second electronically coded content generation module if a predetermined amount of feedback data on the audio-visual data object has been received. This may provide a more stable system.

[0021] A second aspect provides an electronic data processing device comprising an electronic data processor arranged to execute a method of managing generation of audio-visual data. The method comprises operating, on the electronic data processing device, a first electronically coded content generation module configured to generate a first audio- visual data object based on a first set of parameters comprised by the first electronically coded content generation module, operating, on the electronic data processing device, a second electronically coded content generation module configured to generate a second audio-visual data object based on a second set of parameters comprised by the second electronically coded content generation module, receiving, by the electronic data processing device, a first electronic signal comprising feedback data on the first audio- visual data object, determining an adjustment of a first value of a first parameter of the first set of parameters to be adjusted in accordance with the feedback data, updating the first value of the first parameter based on the feedback data, sending, by means of a second electronic signal, data related to the first updated parameter to a second electronically coded content generation module, updating a second value of a second parameter comprised by the second set of parameters based on the updated first value of the first parameter, the second parameter corresponding to the first parameter; and generating the second audio- visual data object based on the second set of parameters the second parameter having the updated value.

[0022] A third aspect provides: a computer programme product comprising computer executable code causing an electronic processing unit, when programmed in accordance with the computer executable code, to execute a method. The method comprises operating, on an electronic data processing device, a first electronically coded content generation module configured to generate a first audio- visual data object based on a first set of parameters comprised by the first electronically coded content generation module, operating, on the electronic data processing device, a second electronically coded content generation module configured to generate a second audio- visual data object based on a second set of parameters comprised by the second electronically coded content generation module, receiving, by the electronic data processing device, a first electronic signal comprising feedback data on the first audio- visual data object, determining an adjustment of a first value of a first parameter of the first set of parameters to be adjusted in accordance with the feedback data, updating the first value of the first parameter based on the feedback data, sending, by means of a second electronic signal, data related to the first updated parameter to a second electronically coded content generation module, updating a second value of a second parameter comprised by the second set of parameters based on the updated first value of the first parameter, the second parameter corresponding to the first parameter, and generating the second audio- visual data object based on the second set of parameters the second parameter having the updated value.

[0023] BRIEF DESCRIPTION OF THE DRAWINGS

[0024] The various aspects and examples thereof will now be discussed in further detail in conjunction with drawings. In the drawings:

[0025] Figure 1: shows an electronic data processing system; and

[0026] Figure 2: shows an example of a method executed by the electronic data processing system.

[0027] DETAILED SUMMARY

[0028] Figure 1 shows an electronic data processing system 100. The electronic data processing system is arranged to generate audio-visual content, publish the generated audio-visual content, receive feedback based on the published audio-visual content and update at least parts of the electronic data processing system that generate audio-visual content, based on the received feedback. Audio-visual content may comprise written text, audio, still images, video sequences, other, or a combination of two or more thereof.

[0029] The various parts of the data processing system 100 may be implemented one a single piece of hardware or two or more pieces of hardware. The different pieces of hardware may be located adjacent to one another, near one another or be remotely located relative to one another. Different pieces of hardware may mean different integrated circuits, different computing devices, different sets of computer systems provided in a single rack in a data warehouse or differently. If the various parts are run on the same processing unit, each device may be implemented as one or more distinct virtual machines.

[0030] The system 100 comprises a control device 140, a social media publication server 170 and a content database server 1010. The social media publication server and the content database server 1010 are not necessarily part of the system 100 and may be provided outside the system, though still in operational connection with the further parts of the system 100.

[0031] The system 100 further comprises a first content generation device 110, a second content generation device 110' and a third content generation device 110". The three content generation devices are implemented in substantially the same way, comprising equivalent components indicated with the same reference signs, with prime and double prime for the second content generation device 110' and the third content generation device 110". Therefore and for the sake of clarity, only the first content generation device 110 will be discussed in detail.

[0032] The first content generation device 110 comprises a generator processing unit 112, a generator storage unit 114 and a storage communication unit 116. The generator processing unit 112 is arranged to control the first content generation device 110 and the various parts thereof. To that purpose, the generator processing unit 112 may be programmed using computer executable code that may be stored in the generator storage unit 114.

[0033] The storage communication unit 116 is arranged to communicate with other entities within the system 100 over a network 900. The network 900 may be a wired network, a wireless network or a combination thereof. The network 900 may be combination of two or more different network structures, like a cellular communication network, another mobile communication network, a optical communication network or other.

[0034] In the generator storage unit 114, a content generation machine 120 is stored. The content generation machine 120 may comprise an artificial intelligence, trained to generate audio-visual data based on a particular prompt on one hand and training data on another hand. The training data may be annotated audio-visual data, the input may be audiovisual data that is to be processed by the content generation machine 120. Additionally or alternatively, the content generation machine 120 may comprise other content generation techniques, based on machine learning - for example comprising a genetic algorithm - or other techniques.

[0035] The content generation machine 120 has a first generation parameter 122, a second generation parameter 124 and a third generation parameter 126 associated with it. Each of these parameters has a value associated with it. The value is preferably numerical, but may also be alphanumerical or have another format. With different values of the parameters, the content generation machine 120 may handle input data in a different way, either for the purpose of training or for the purpose of generation of content.

[0036] The generator processing unit 112 comprises a machine execution subunit 132, a posting subunit 134 and an updating subunit 136. The machine execution subunit 132 is arranged to execute the content generation machine 120. The posting subunit 134 is arranged to generate a post comprising the generated content, for sending the post to the social media publication server 170 for publication of the generated content, via the storage communication unit 116. The updating subunit 136 is arranged to update a value of at least one of the first generation parameter 122, the second generation parameter 124 and the third generation parameter 126 and / or update the content generation machine 120 in its entirety. The control device 140 is arranged to control the first content generation device 110, the second content generation device 110' and the third content generation device 110". To that purpose, the control device comprises a control processing unit 142, a control storage unit 144 and a control communication unit 146. The control processing unit 142 is arranged to control the control device 140 and the various parts thereof. To that purpose, the control processing unit 142 may be programmed using computer executable code that may be stored in the control storage unit 144. The control communication unit 146 is arranged to communicate with other entities within the system 100 over the network 900.

[0037] In the control storage unit 144, a master machine 150 is stored. The master machine 150 is a machine like the content generation machine 110 and may be distributed to each of the first content generation device 110, the second content generation device 110' and the third content generation device 110" as a seed, as will be discussed below, to ensure that all content generation machines comprise the same content generation machine 120 with the same values at one moment in time.

[0038] The master machine 150 comprises a first master parameter 152, a second master parameter 154 and a third master parameter 156, controlling the operation of the master machine 150 when executed. The parameters may provide an indication of basic content items to be used for generation of audio-visual content items, indications of colours to be used, of types of people to be included in the audio-visual content items, of product to be included in the audio-visual content items, of a general atmosphere of the audio-visual content items to be generated, of a speed of scene change within the audio-visual content items to be generated, other, or a combination of two or more thereof.

[0039] The control processing unit 142 comprises a group management subunit 162, a master updater subunit 164 and a parameter processing subunit 166. The group management subunit 162 controls the operation of the first content generation device 110, the second content generation device 110' and the third content generation device 110". The master updater subunit 164 is arranged to update the master machine 150, either as a whole or by updated one or more values of the first master parameter 152, the second master parameter 154 and the third master parameter 156.

[0040] The parameter processing subunit 166 is arranged to receive updates from the first content generation device 110, the second content generation device 110' and the third content generation device 110" and feedback data that is received by the social media publication server 170 - either directly or via the first content generation device 110, the second content generation device 110' and the third content generation device 110" - and to process the data received to provide data to the master updater subunit 164 on how the master machine 150 may be updated.

[0041] The social media server 170 may be implemented as computer systems known in the art for hosting social media platforms. The details provided here provide merely an example how such platform may be implemented. The social media platform server 170 may for example comprise a social media processing unit 172, a social media storage unit 174 and a social media communication unit 176.

[0042] The social media processing unit 172 is arranged to control the social media device 170 and the various parts thereof. To that purpose, the social media processing unit 172 may be programmed using computer executable code that may be stored in the social media storage unit 177. The social media communication unit 176 is arranged to communicate with other entities within the system 100 over the network 900.

[0043] The social media processing unit 172 comprises a publication subunit 192 controlling publication of posts submitted to the platform social media server 170. With the publication of posts, audio-visual data is presented to users of the social media service hosted by the social media server 170. The social media processing unit 172 further comprises a feedback subunit 194 for processing feedback that members of the public that use the social media service - followers - submit to the social media platform in response to a post.

[0044] The feedback may comprise a selection of pre- defined feedback - for example 'like', 'love', 'support', 'funny', 'dislike', 'hate', 'approve' -, free text feedback, an audio-visual data item, or a combination of two or more thereof. The feedback may be classified in a particular category expressing a sentiment or in general, per feedback item or in general for a post and in particular for an audio-visual data item comprised by the post. For example in such case, but also in other cases, the feedback may be provided with a numerical value indicating an approval rating in general or a rating in a classification in particular.

[0045] In another example, the generated audio-visual content is published to and audience as a commercial message embedded in other content, like an online newspaper. In such scenario, no direct feedback may be provided. Rather, engagement of the audience may be measured in other ways. Such engagement may be measured in whether a user has moved a pointing device like a mouse pointer or a finger to or over the content item, a time for how long a user hovers a mouse or other pointing device over the content item as displayed, how fast a user pauses display of the content item, how long the content item has bee displayed to a user or whether a user has dismissed display of the content item. Hence feedback is broader then 'likes' or direct free text comment.

[0046] The social media storage unit 174 is arranged to store one or more social media posts 180. In this example, the social media post 180 is generated by one of the first content generation device 110, the second content generation device 110' and the third content generation device 110" and sent to the social media server 170 via the network 190. The social media post 180 comprises an audio-visual content item 182, account information 184 indicating an account of an entity - person - to which the post 180 is related and feedback data 186, indicating the feedback as discussed above.

[0047] The system 100 further comprises a content database server 1010. The content database server 1010 is arranged for storing basic content data for training content generation machines - the content generation machine 120, the master machine 150 or both - or for input for generation of content items.

[0048] The content database server 1010 may comprise a content database processing unit 1012, a content database storage unit 1014 and a content database communication unit 1016. The content database processing unit 1012 is arranged to content database the content database device 1010 and the various parts thereof. To that purpose, the content database processing unit 1012 may be programmed using computer executable code that may be stored in the content database storage unit 1017. The content database communication unit 1016 is arranged to communicate with other entities within the system 100 over the network 900.

[0049] The database processing unit 1012 comprises a database management subunit 1032 for managing a database for a first content item 1022, a second content item 1024 and a third content item 1026 that are stored in the content database storage unit 1014. The database processing unit 1012 further comprises a content distribution subunit 134 for distribution content to the control server 140, the first content generation device 110, the second content generation device 110' and the third content generation device 110"

[0050] The functionality of the various parts of the electronic data processing system 100 will now be discussed in further detail in conjunction with a flowchart 200 as shown by Figure 2. The flowchart 200 depicts an example of the first aspect. The parts of the flowchart are briefly summarised below: 202 start procedure

[0051] 204 generate content using model

[0052] 206 post to social media

[0053] 208 receive feedback

[0054] 210 modify parameter value

[0055] 212 generate content

[0056] 214 post to social media (update post)

[0057] 216 receive feedback

[0058] 218 process feedback received (compare)

[0059] 220 determine new values

[0060] 222 time to share?

[0061] 224 send feedback data to control server

[0062] 226 aggregate feedback data

[0063] 228 structural change?

[0064] 230 generate new model

[0065] 232 send new model

[0066] 234 update model

[0067] 236 generate new parameter set

[0068] 238 send new parameter set

[0069] 240 update parameter set

[0070] The procedure starts in a terminator 202, in which various parts of the electronic data processing system 100 are initialised. Such initialisation may comprise initialising the first content generation device 110, the second content generation device 110' and the third content generation device 110" by distributing the master machine 150 to each of the first content generation device 110, the second content generation device 110' and the third content generation device 110" to be set in each content generation device as the content generation machine 120 in the generator storage unit 114. Additionally or alternatively, the initialisation may comprise setting up the content database in the content database server 1010. Additionally or alternatively, other parts of the electronic data processing system 100 may be set to a known, optionally pre-determined, state.

[0071] In step, 204, the content generation machine 120 is operated by the machine execution subunit 132 in the first content generation device 110. The audio-visual data item may relate to advertisement of goods, services or particular brand covering such goods or services. This step and other steps executed by the first content generation device 110 are or may be also executed by the second content generation device 110' and the third content generation device 110", unless discussed otherwise. This execution may be done simultaneously or subsequently, synchronised with respect to frequency or not.

[0072] In step 206, a social media post comprising the generated audiovisual content item is generated and send to the social media platform server 170 by the posting subunit 134. The post is published by means of the publication subunit 192 and feedback on the post is received by the feedback subunit 194. Next, the feedback is sent to the first content generation device 110 via the network 190 by means of the social media communication unit 176 and the storage communication unit 116, which received the feedback data in step 208. The feedback data may be aggregated data or raw data, providing an indication of each individual feedback received at each time.

[0073] In step 210, at least one value of at least one parameter of the content generation machine 120 is modified. The parameter modified and the amount by which the value is modified may be random or predetermined, optionally within particular boundaries. In step 212, the content generation machine 120 is operated by the machine execution subunit 132 in the first content generation device 110 again and the newly generated audio-visual content item is submitted to the social media platform server 170 as discussed above in step 214. In step 216, feedback is received from the social media platform server 170 as discussed above, based on the newly generated content item.

[0074] In step 218, the feedback data received for the firstly generated audio-visual content item and the secondly generated audio-visual content item is compared. Optionally, more than two audio-visual content items may be generated based on modified parameter values modified as in step 210 and feedback on all or at least most of the generated audio-visual content items is compared in step 218.

[0075] Based on the comparing step 218, in step 220, new values for the parameters of the content generation machine 120 are determined. Values with which audio-visual content has been generated with which a more positive feedback has been received may be used to update values of the parameters of the content generation machine 120. Optionally, new values or content generated these new values are checked against boundary conditions. Such boundary conditions may indicate boundaries of the audiovisual content; a boundary may be that any content generated does not have explicit language in it or demonstrate or otherwise promote illegal or illicit behaviour, like underage drinking or drinking and driving.

[0076] In step 222 is checked whether feedback may be sent to the control device 140 for aggregation of feedback data. If this is not the case, the parameters of the content generation machine 120 are updated locally by assigning values for which more positive feedback has been received in step 210 again. If it is time for aggregation of feedback received in response to content, feedback data is sent to the control device 140 in step 224. Optionally, in such case, also local updating of the content generation machine 120 may be performed as well, in step 210.

[0077] In step 226, feedback data from the first content generation device 110, the second content generation device 110' and the third content generation device 110" is aggregated. It is noted that audio- visual content items generated by the first content generation device 110, the second content generation device 110' and the third content generation device 110" may have been published on different platforms and published to different audiences.

[0078] The different audiences may be reached through various platforms. The first content generation device 110 has posted a first audiovisual content item it generated to a social media platform in Europe, the second content generation device 110' has provided a second audio-visual content item it generated to a Mexican newspaper as an advertisement and the third content generation device 110" has provided a third audio-visual content item it generated to a gaming platform in Japan, to be shown in a watch to play video game environment. In another example, the audiences are the same or overlap.

[0079] The aggregation may comprise processing of feedback data to determined which values of particular parameters comprised by the content generation machine 120, the second content generation machine 120' and the third content generation machine 120" result in generation of audiovisual content that, when published by means of the social media platform server 170, attract the highest feedback scores on a positive scale.

[0080] The aggregation may be a result of averaging on general or particular weight may be attributed to particular parameters. Additionally or alternatively, values of parameters resulting in significantly higher feedback scores may be given higher weight - so the aggregation does not necessarily have to be a linear operation.

[0081] In step 228, the control processing unit 142 checks whether the result of the aggregation provides parameter values for a potentially better feedback score are within or beyond a particular boundary. If the changes are outside a particular boundary, the change represents a structural change. In case of such structural change, the process continues to step 230, in which the master machine 150 is retrained, for example by the master updater subunit 164. For training of the master machine 150, content may be used as stored in the content database server 1010.

[0082] In step 232, the newly trained master machine 150 is sent to the first content generation device 110, the second content generation device 110' and the third content generation device 110" by the group management subunit 162 and the applicable communication units, via the network 900. The newly trained master machine 150 is to replace the first content generation machine 120, the second content generation machine 120' and the third content generation machine 120". Subsequently, the procedure continues to step 204.

[0083] If the update of parameters, based on aggregation of parameter data and feedback data, with an aim to attract higher feedback data, compared to boundary conditions indicate that update of parameters values indicates a minor change, the procedure continues from step 228 to step 236. In step 236, the parameter processing subunit 166 determines new parameter values for use by the first content generation machine 120, the second content generation machine 120' and the third content generation machine 120".

[0084] The use of the parameter values thus provided by the parameter processing subunit 166 and sent to the content generation devices in step 238 may in one example be provided to fully replace parameter values in the first content generation machine 120, the second content generation machine 120' and the third content generation machine 120".

[0085] In another example, the parameter processing subunit 166 may provide parameter values to be sued in each of the content generation machine for updating in combination with values used in each of the content generation machines - preferably locally determined optimal values -, with a weight connected to each of the parameter for determining a weighted average value of the updated value provided by the control device 140 and the locally determined optimal values. In either example or other implementations as variations thereof, local parameter sets are updated in step 240. Subsequently, the procedure continues to step 204.

[0086] In summary, a method is provided by generating audio-visual content data, for example advertisements, by multiple electronically coded content generation modules. The audio-visual data objects are generated based on an initial dataset, with parameters having pre-defined values. The values may be changed in the course of time. Feedback may be received from an audience on appreciation of the content item. Based on the feedback, parameters of one or more content generation modules may be changed. In this way, best practices may be shared among multiple content generation modules, by updating parameters based on parameters of content generation modules that receive the highest feedback scores. As such, the content generation modules may operate as a swarm.

Claims

Claims1. In a first electronic data processing system, a method of managing generation of audio-visual data, the method comprising: operating, on the electronic data processing device, a first electronically coded content generation module configured to generate a first audio- visual data object based on a first set of parameters comprised by the first electronically coded content generation module; operating, on the electronic data processing device, a second electronically coded content generation module configured to generate a second audio- visual data object based on a second set of parameters comprised by the second electronically coded content generation module; receiving, by the electronic data processing device, a first electronic signal comprising feedback data on the first audio- visual data object; determining an adjustment of a first value of a first parameter of the first set of parameters to be adjusted in accordance with the feedback data; updating the first value of the first parameter based on the feedback data; sending, by means of a second electronic signal, data related to the first updated parameter to a second electronically coded content generation module; updating a second value of a second parameter comprised by the second set of parameters based on the updated first value of the first parameter, the second parameter corresponding to the first parameter; and generating the second audio- visual data object based on the second set of parameters the second parameter having the updated value.

2. The method according to claim 1, further comprisingoperating, on the electronic data processing device, an electronically coded control module arranged to aggregate feedback data from multiple electronically coded content generation modules operated on electronic data processing device; wherein the data related to the first updated parameter is sent to the electronically coded control module; and the method further comprising: generating aggregate feedback data based on the data related to the first updated parameter received from the first electronically coded content generation module; providing the aggregate feedback data on the first parameter to the second electronically coded content generation module.

3. The method according to claim 2, wherein the aggregated feedback data comprises weighting data providing an indication how the aggregated feedback data is to be weighted relative to at least one of feedback data on the audio- visual data object and the aggregated feedback data for adjusting at least one parameter of the first content generation module.

4. The method according to claim 2 or 3, wherein the aggregated feedback data comprises one of a value of a parameter of the second data set that is to be adjusted and a feedback score.

5. The method according to any one of claim 2 to claim 4, wherein: a further multitude of electronically coded content generation module are operated on the electronic data processing device; and the electronically coded control module is arranged to determine which of the multiple content generation modules receives the highest feedback score.

6. The method according to claim 5, wherein feedback data of the further multitude of electronically coded content generation modules that receives the highest feedback score has a weight in the aggregated feedback data higher than feedback data of other content generation modules.

7. the method according to any of the preceding claims, further comprising: obtaining boundary data providing boundary conditions for content of the audio-visual data; determining whether the content generated module, with parameter to be adjusted having an adjusted value, complies with the boundary conditions; updating, if the content generation module complies with the boundary conditions with the parameter having the adjusted value, the at least one parameter of the content generation module based on the feedback data.

8. The method according to any of the preceding claims, further comprising: sending the first audio- visual data object to an electronic computing system hosting an electronic content publishing service, for publication of the first audio-visual data object to a first audience; wherein the first electronic signal is received from the electronic computing system and feedback data is based on feedback of the first audience on the first audio-visual data object.

9. The method according to claim 8, further comprising: sending the second audio- visual data object to an electronic computing system hosting an electronic content publishing service, forpublication of the second audio- visual data object to a second audience.

10. The method according to claim 9, wherein the first audience is different from the second audience.

11. The method according to any of the preceding claims, wherein the feedback data comprises at least one of the following: a time period during which the audio- visual data object has been displayed to one or more members; whether one or more members have actively interrupted display of the audio-visual data object; whether one or more members have moved a pointing device over display of the audio-visual data object; a time period during which one or more members have positioned a pointing device over display of the audio-visual data object.

12. The method according to any of the preceding claims, wherein the first electronically coded content generation module is arranged to receive central content data from a central content repository; and generate the audio-visual data object based on the central content data.

13. The method according to any of the preceding claims, wherein data related to the updated parameters is sent to the second electronically coded content generation module if a predetermined amount of feedback data on the audio-visual data object has been received.

14. An electronic data processing device comprising an electronic data processor arranged to execute a method of managing generation of audio-visual data, the method comprising:operating, on the electronic data processing device, a first electronically coded content generation module configured to generate a first audio- visual data object based on a first set of parameters comprised by the first electronically coded content generation module; operating, on the electronic data processing device, a second electronically coded content generation module configured to generate a second audio- visual data object based on a second set of parameters comprised by the second electronically coded content generation module; receiving, by the electronic data processing device, a first electronic signal comprising feedback data on the first audio- visual data object; determining an adjustment of a first value of a first parameter of the first set of parameters to be adjusted in accordance with the feedback data; updating the first value of the first parameter based on the feedback data; sending, by means of a second electronic signal, data related to the first updated parameter to a second electronically coded content generation module; updating a second value of a second parameter comprised by the second set of parameters based on the updated first value of the first parameter, the second parameter corresponding to the first parameter; and generating the second audio- visual data object based on the second set of parameters the second parameter having the updated value.

15. A computer programme product comprising computer executable code causing an electronic processing unit, when programmed in accordance with the computer executable code, to execute a method comprising:operating, on an electronic data processing device, a first electronically coded content generation module configured to generate a first audio- visual data object based on a first set of parameters comprised by the first electronically coded content generation module; operating, on the electronic data processing device, a second electronically coded content generation module configured to generate a second audio- visual data object based on a second set of parameters comprised by the second electronically coded content generation module; receiving, by the electronic data processing device, a first electronic signal comprising feedback data on the first audio- visual data object; determining an adjustment of a first value of a first parameter of the first set of parameters to be adjusted in accordance with the feedback data; updating the first value of the first parameter based on the feedback data; sending, by means of a second electronic signal, data related to the first updated parameter to a second electronically coded content generation module; updating a second value of a second parameter comprised by the second set of parameters based on the updated first value of the first parameter, the second parameter corresponding to the first parameter; and generating the second audio- visual data object based on the second set of parameters the second parameter having the updated value.

Citation Information

Patent Citations

  • Low floor type motor driven assist- vehicle for field work

    KR1020240077079A

  • Automatically generating enhancements to AV content

    US20210368222A1

  • Partially local federated learning

    US20220398500A1

  • Method and system for generating content data

    WO2020162833A1