Multimedia server intended to be installed on board an aircraft, electronic entertainment system, method for generating additional content and computer program associated therewith

Abstract

The invention relates to a multimedia server (20) intended to be installed on board an aircraft (10) and connected to at least one entertainment terminal via an onboard local network, comprising: - a module (30) for generating additional content intended for the at least one entertainment terminal; - a knowledge base (34) storing a set of entertainment content elements; - a request module (32) for receiving a control instruction from the at least one entertainment terminal and for requesting, from the generating module (30), on the basis of the received control instruction, generation of additional content, the generating module (30) comprising a generative artificial intelligence model configured to generate, from the set of entertainment content elements stored in the knowledge base (34), the additional content intended for the at least one entertainment terminal.

Description

[0001] Multimedia server intended for installation on board an aircraft, electronic entertainment system, method for generating additional content and associated computer program

[0002] The present invention relates to a multimedia server intended to be installed on board an aircraft and connected to at least one on-board entertainment terminal on said aircraft via an on-board local network.

[0003] The invention also relates to an electronic entertainment system intended to be carried on board an aircraft, the system comprising at least one such entertainment terminal and one such multimedia server.

[0004] The invention also relates to a method for generating additional content, the method being implemented by the multimedia server.

[0005] The invention also relates to a computer program comprising software instructions which, when executed by a computer, implement such a method of generating additional content.

[0006] The invention relates to the field of aircraft entertainment systems, also known as in-flight entertainment systems or IFE (from the English In-Flight Entertainment), which are mainly found on board long-haul commercial aircraft.

[0007] As is typical, such an IFE system comprises a plurality of entertainment terminals, also called user terminals, generally in the form of individual screens or tablets. Each entertainment terminal is integrated, for example, into the passenger seat or the seat in front of it.

[0008] Arranged in this way, these terminals allow passengers to access multimedia content during the flight (for example, movies, TV shows, games or music) and to be informed about the progress of the flight (altitude, speed, current position, progress, etc.).

[0009] In some cases, these terminals also allow for the provision of certain practical information concerning, for example, the arrival airport and the broadcasting of announcements made by the crew in audio and / or video format.

[0010] Entertainment terminals are typically connected via a wired or wireless network to an embedded media server that is also part of the IFE system. Depending on the various possible IFE system implementations, each terminal accesses the embedded media server via the network or uses its own storage space to access computer content and perform its functions.

[0011] Some systems also offer entertainment terminals that can implement at least some of its functions from their storage space and some others from the embedded media server.

[0012] Thus, in order to be accessible, the computer content is first loaded onto the embedded multimedia server and then possibly distributed to the entertainment terminals via the embedded local network to be stored in their storage space.

[0013] However, such a process of pre-loading computer content onto the embedded media server and then storing it in the entertainment terminals' storage space is generally cumbersome and expensive, especially with regard to updating the computer content. This process consumes significant computing resources and bandwidth during distribution to the entertainment terminals.

[0014] The aim of the invention is therefore to provide a multimedia server intended to be carried on board an aircraft and connected to at least one entertainment terminal, enabling the generation of computer content for said at least one entertainment terminal, while reducing the consumption of computer resources and bandwidth during the storage of the content and its distribution to said at least one entertainment terminal.

[0015] To this end, the invention relates to a multimedia server intended to be installed on board an aircraft and connected to at least one onboard entertainment terminal on said aircraft via an onboard local area network, the multimedia server comprising:

[0016] - a generation module configured to generate additional content for at least one entertainment terminal,

[0017] - a knowledge base storing a set of entertainment content elements,

[0018] - a request module configured to receive a command instruction from said at least one entertainment terminal and to request from the generation module, on the basis of said received command instruction, the generation module comprising a generative artificial intelligence model, said generative artificial intelligence model being configured to generate, from the set of entertainment content elements stored in the knowledge base, additional content for said at least one entertainment terminal, each additional content generated by the generative artificial intelligence model being content distinct from the content elements included in the knowledge base.

[0019] Thanks to the incorporation of a generative artificial intelligence model configured in the additional content generation module, the multimedia server according to the invention makes it easier to generate additional content for each entertainment terminal, for the same given storage space. Furthermore, the multimedia server according to the invention reduces the computing resources consumed at the content storage level itself, while providing the same level of entertainment to the user of the entertainment terminal. Finally, the multimedia server according to the invention reduces the bandwidth used when downloading content to an electronic entertainment system equipped with the server.

[0020] A generative artificial intelligence model is defined as any artificial intelligence model capable of generating text, images, videos, or other media in response to requests, and equipped with one or more generative subsystem(s) producing multiple outputs.

[0021] According to other advantageous aspects of the invention, the multimedia server comprises one or more of the following features, taken individually or in all technically possible combinations:

[0022] - the multimedia server is made in the form of an electronic box comprising a backplane card and a plurality of electronic cards connected to the backplane card;

[0023] - the generative artificial intelligence model is implemented on a dedicated electronic board;

[0024] - the generation module is configured to generate additional personalized content based on a user of said at least one entertainment terminal;

[0025] - the generation module is configured to generate additional content adapted according to the characteristics of the aircraft flight and / or according to one or more editorial constraint(s) and / or according to one or more regulatory constraint(s);

[0026] - the generative artificial intelligence model includes a neural network based on a self-attentive model or a generative adversarial neural network;

[0027] - said neural network based on a self-attentive model is a large language model; and - the additional content is selected from the group consisting of: multimedia content such as film, television program, video game or audio content; textual content and photo content.

[0028] The invention also relates to an electronic entertainment system intended to be installed on board an aircraft, the system comprising:

[0029] - at least one entertainment terminal; and

[0030] - a multimedia server connected to each entertainment terminal via an embedded local network, the multimedia server being as defined above.

[0031] The invention also relates to a method, implemented by a multimedia server intended to be installed on board an aircraft and connected to at least one onboard entertainment terminal on said aircraft via an onboard local area network, for generating additional content for said at least one entertainment terminal, the multimedia server comprising an additional content generation module, a knowledge base storing a set of entertainment content elements, and a query module, the additional content generation module comprising a generative artificial intelligence model, said generative artificial intelligence model being configured to generate additional content from the set of entertainment content elements stored in the knowledge base, the method comprising the following steps:

[0032] - the reception, by the request module, of a command instruction from said at least one entertainment terminal,

[0033] - based on the received command instruction, the request module issues a request to generate additional content to the additional content generation module,

[0034] - the receipt, by the additional content generation module, of the request to generate additional content that was issued,

[0035] - based on the request for additional content generation received, the generation, by the additional content generation module, of additional content for said at least one entertainment terminal, each additional content generated by the generative artificial intelligence model being a content distinct from the content elements included in the knowledge base.

[0036] According to other advantageous aspects of the invention, the method for generating additional content comprises one or more of the following features, taken individually or in all technically possible combinations:

[0037] - the step of generating additional content for said at least one entertainment terminal is carried out on the fly; - the generative artificial intelligence model is trained on the ground before the multimedia server is loaded onto the aircraft, the generative artificial intelligence model making inferences from its training data when the multimedia server is loaded onto the aircraft;

[0038] - the generative artificial intelligence model is trained by automatic machine learning (from the English Auto Machine Learning) or by fine-tuning (from the English fine-tuning).

[0039] The invention also relates to a computer program comprising software instructions which, when executed by a computer, implement a method for generating additional content, as defined above.

[0040] These features and advantages of the invention will become clearer upon reading the following description, given solely by way of non-limiting example, and made with reference to the accompanying drawings, in which:

[0041] - Figure 1 is a schematic view of several aircraft, each equipped with an electronic entertainment system including entertainment terminals, not shown, and a multimedia server, each aircraft being near its parking spot in an airport;

[0042] - Figure 2 is a schematic view of an electronic enclosure forming the multimedia server of Figure 1, the enclosure comprising a backplane board and a plurality of electronic boards connected to the backplane board; and

[0043] - Figure 3 is a flowchart of a method for generating additional content according to the invention, the method being implemented by the multimedia server of Figure 1.

[0044] In Figure 1, several aircraft 10 are positioned near their respective parking positions at an airport (not shown). Each aircraft 10 is equipped with an electronic entertainment system 16 comprising entertainment terminals (not shown) and a media server 20.

[0045] Each aircraft 10 is preferably an airplane, in particular a commercial airplane, such as a long-haul aircraft. Each aircraft 10 is capable of carrying passengers, in particular a few dozen passengers, or even a few hundred passengers.

[0046] Each entertainment system 16, on board a respective aircraft 10, is configured to broadcast multimedia content to the passengers of the aircraft 10, particularly during the flight (e.g., movies, TV shows, games or music), and / or information on the progress of the flight (altitude, speed, current position, progress, etc.).

[0047] As an optional extra, each 16-seater entertainment system is configured to broadcast practical information concerning, for example, the arrival airport, for example via announcements in audio and / or video form.

[0048] Each entertainment terminal is known in itself, and is connected to the multimedia server 20 via a local network, not shown, on board the respective aircraft 10.

[0049] Each entertainment terminal is, for example, fixed or integrated into the passenger's seat itself, or fixed or integrated into the seatback in front of the passenger's seat. The seats are typically arranged in rows within the aircraft.10

[0050] Each entertainment terminal, also called a user terminal, comprises a human-machine interface (HMI), a processing module, an application module, and a telecommunications module. The HMI includes, for example, a touchscreen, or a screen and a keyboard. The application module comprises one or more application blocks. The telecommunications module includes a control block and one or more telecommunications interfaces adapted for transmitting and receiving data. The processing module is adapted to control and coordinate the operation of the application module, the HMI, and the telecommunications module. In one embodiment, it includes a processor and memory, for example, a hard drive or any other storage space suitable for storing computer content received, in particular, via the telecommunications module.

[0051] Each entertainment terminal is adapted to deliver the content offered by the in-flight entertainment system to passengers. For example, after a passenger selects an option on the screen of their entertainment terminal's HMI (Human-Machine Interface), the processing module is adapted to identify that a menu for selecting one of the applications, such as a list of movies, should then be displayed. Then, upon detection of a movie selection by the passenger, the processing module is adapted to trigger the movie's display on the screen. In addition to movies, the content delivered by the applications can be varied: multimedia content, such as movies, TV programs, games, or music; flight parameters (altitude, speed, etc.) and progress (for example, using a moving map); and announcements (audio and / or video) from the crew.The screen is capable of displaying images corresponding to computer content distributed from the multimedia server 20. More precisely, the computer content is received via the telecommunications module from the multimedia server 20 in the form of a broadcast stream (from the English word "streaming") or stored in the terminal's memory, after having been previously downloaded from the multimedia server 20 via said telecommunications module.

[0052] The multimedia server 20 is connected to at least one entertainment terminal, preferably to each entertainment terminal, via an embedded local network.

[0053] The multimedia server 20 includes an additional content generation module 30, and a request module 32 configured to request additional content generation from the generation module 30.

[0054] The multimedia server 20 includes a knowledge base 34 storing a set of entertainment content elements.

[0055] The multimedia server 20 also includes a module 40 for distributing computer content to each entertainment terminal.

[0056] The Generation 30 module includes a generative artificial intelligence model. Preferably, the generative artificial intelligence model comprises a neural network based on a self-attentive model or a generative adversarial neural network. When the generative artificial intelligence model includes a neural network based on a self-attentive model, the latter is, for example, a large language model (LLM).

[0057] In the example in Figure 1, the multimedia server 20 includes an information processing unit 45 formed for example of a memory 46 and a processor 48 associated with the memory 46.

[0058] Memory 46 of multimedia server 20 is then capable of storing generation software, query software, generative artificial intelligence software, and optionally, distribution software. Processor 48 is then capable of executing each of the following software programs: generation software, query software, generative artificial intelligence software, and optionally, distribution software. Memory 46 of multimedia server 20 is also capable of storing a knowledge base.

[0059] In an unrepresented variant, the generation module 30, the query module 32, the generative artificial intelligence model, and as an optional complement the distribution module 40, are each implemented as a programmable logic component, such as an FPGA (Field Programmable Gate Array), or as a dedicated integrated circuit, such as an ASIC (Application Specific Integrated Circuit).

[0060] When the multimedia server 20 is implemented as one or more software programs, that is, as a computer program, it is also capable of being stored on a computer-readable medium (not shown). A computer-readable medium is, for example, a medium capable of storing electronic instructions and being connected to a bus of a computer system. Examples of such a readable medium include an optical disc, a magneto-optical disc, ROM, RAM, any type of non-volatile memory (e.g., EPROM, EEPROM, FLASH, NVRAM), a magnetic card, or an optical card. A computer program containing software instructions is then stored on this readable medium.

[0061] The multimedia server 20 is, for example, implemented as an electronic enclosure 50 comprising a backplane card 52 and a plurality of electronic cards 54 and one or more power supply cards 55. Each electronic card 54, and each power supply card 55 respectively, is connected to the backplane card 52 via a respective backplane connector 56, as shown in Figure 2. The electronic enclosure 50 further comprises a protective housing 58 inside which are housed the backplane card 52 and the plurality of electronic cards 54 and power supply card(s) 55, and external connectors 60 arranged around the periphery of the housing 58. The external connectors 60 are intended, in particular, to allow the multimedia server 20 to be connected to the onboard local area network, as well as to an onboard power supply network on board the aircraft 10.

[0062] The local network on board aircraft 10 is a wired-only network, or a partially wired and partially wireless network, or a completely wireless network.

[0063] When the onboard local area network is partly wired and partly wireless, the multimedia server 20 is for example wired to an entertainment terminal in each row of seats, and said terminal wired to the multimedia server 20 is then wirelessly connected to the other terminals in the row of seats.

[0064] Each wireless link is, for example, a radio link, such as a link conforming to at least one version of the IEEE 802.11 standard, also known as the Wi-Fi™ standard, or a link conforming to the Li-Fi standard, for example to the IEEE 802.15.7 standard.

[0065] The generation module 30 is configured to generate additional content for each entertainment terminal. For example, the generation module 30 is configured to generate personalized additional content based on a specific user of the entertainment terminal. Alternatively, or in addition, the generation module 30 is configured to generate additional content adapted to the flight characteristics of aircraft 10 and / or based on one or more editorial constraints and / or based on one or more regulatory constraints. The request module 32 is configured to receive a command instruction from the entertainment terminal and to request the generation module 30 to generate additional content based on the received command instruction.

[0066] The distribution module 40 is configured to distribute the corresponding computer content to each entertainment terminal. For example, the distribution module 40 is configured to transmit the corresponding computer content to each entertainment terminal as a streaming stream. In this case, the user of the entertainment terminal can view the content as it is being transmitted as a streaming stream.

[0067] Alternatively or in addition, the distribution module 40 is configured to load the corresponding computer content into the local memory of each entertainment terminal. Those skilled in the art will understand that this content loading must be performed beforehand to allow the entertainment terminal user to subsequently access this content.

[0068] The generative artificial intelligence model is configured to generate additional content for each entertainment terminal from the set of entertainment content elements stored in knowledge base 34. Preferably, the generative artificial intelligence model is implemented on a dedicated electronic board, specifically on one of the electronic boards 54.

[0069] Each additional piece of content generated by the generative artificial intelligence model is new content distinct from the content elements included in the knowledge base.

[0070] Advantageously, each additional piece of content is distinct from a linear combination of content elements included in the knowledge base.

[0071] Advantageously, each additional content is distinct from a frame interpolation from a content element in the knowledge base.

[0072] Advantageously, each additional piece of content is entirely generated by the generative artificial intelligence model.

[0073] Advantageously, each additional piece of content is usable by at least one entertainment terminal independently of the content elements included in the knowledge base.

[0074] The neural network incorporated into the generative artificial intelligence model is typically a deep neural network. This neural network comprises an ordered succession of layers of neurons, each taking its inputs from the outputs of the preceding layer.

[0075] More specifically, each layer comprises neurons taking their inputs from the outputs of the neurons in the previous layer, or from the input variables for the first layer.

[0076] Alternatively, more complex neural network structures can be considered with a layer that can be linked to a layer further away than the immediately preceding layer.

[0077] Each neuron is also associated with an operation, that is, a type of processing, to be carried out by said neuron within the corresponding processing layer.

[0078] The operation of the multimedia server 20 will now be explained, particularly with the aid of Figure 3, which shows a flowchart of the additional content generation process according to the invention, the process being implemented by the multimedia server 20. Prior to this, the generative artificial intelligence model was trained using one or more training datasets. The training data typically includes data relating to multimedia content such as films, television programs, video games, or audio content (music, sound information); textual content; or even photo content (images, photographs). Preferably, the generative artificial intelligence model is trained on the ground before the multimedia server 20 is loaded onto the aircraft 10.The generative artificial intelligence model is trained, for example, by automatic machine learning or fine-tuning, either on board the aircraft or on the ground. The generative artificial intelligence model then makes inferences from its training data when the multimedia server 20 is installed on board the aircraft 10.

[0079] During an initial step 100, the multimedia server 20 receives, via its request module 32, a command instruction from one of the entertainment terminals. This command instruction takes, for example, the form of a notification sent by the entertainment terminal to the multimedia server 20 via the onboard local network, for instance, following a passenger's selection of a field displayed on the terminal's HMI screen. This command instruction incorporates a request to generate additional content; such additional content can typically take the form of multimedia content such as movies (films, animations, short sequences, documentaries created from scratch), television programs, video games, or audio content; text content; or even photo content.For example, such a request to generate additional content might correspond to a passenger selecting a field on their device specifying the generation and delivery of a video stream of a predetermined duration (which the passenger can optionally specify), incorporating a predetermined type of content (which the passenger can also optionally specify), or conversely, omitting certain content. The generated content may be complete, adapted or added to content already available on board, or provided by the passenger themselves.

[0080] The request module 32 of the media server 20 then sends, during step 110 and on the basis of the received command instruction, a request to generate additional content to the additional content generation module 30. During a subsequent step 120, the additional content generation module 30 receives the request to generate additional content issued by the request module 32.

[0081] The additional content generation module 30 then responds to this request in a subsequent step 130 by generating additional content for the entertainment terminal. More specifically, the generative artificial intelligence model makes inferences by taking as input one or more elements provided in the command instruction and then transmitted via the additional content generation request (such elements typically being a content type, content duration, etc.). The generative artificial intelligence model then generates the requested additional content. Following this generation step 130, the distribution module 40 of the media server 20 distributes the generated additional content to the entertainment terminal. The user of the entertainment terminal can then view and / or listen to this content on their terminal.

[0082] Preferably, step 130 of generating additional content is performed on the fly. In this case, the user of the entertainment terminal can view the content as it is transmitted in a broadcast stream (streaming mode).

[0083] Thus, by generating additional computer content via a generative artificial intelligence model, the multimedia server 20 according to the invention makes it easier to generate additional content for each entertainment terminal, for the same given storage space, as well as to reduce data storage within the entertainment system 16. The multimedia server 20 according to the invention is also less restrictive in terms of content updates (and therefore less expensive), compared to multimedia servers of the prior art.

[0084] It is thus understood that the multimedia server 20 according to the invention makes it possible to generate additional computer content for each entertainment terminal, while reducing the consumption of computer resources and bandwidth when storing the content in the entertainment system 16 and distributing it to the entertainment terminals.

Claims

DEMANDS 1. A multimedia server (20) intended to be installed on board an aircraft (10) and connected to at least one onboard entertainment terminal on said aircraft (10) via an onboard local area network, the multimedia server (20) comprising: - a generation module (30) configured to generate additional content for said at least one entertainment terminal, - a knowledge base (34) storing a set of entertainment content elements, - a request module (32) configured to receive a command instruction from said at least one entertainment terminal and to request from the generation module (30), on the basis of said received command instruction, the generation module (30) comprising a generative artificial intelligence model, said generative artificial intelligence model being configured to generate, from the set of entertainment content elements stored in the knowledge base (34), additional content for said at least one entertainment terminal, each additional content generated by the generative artificial intelligence model being content distinct from the content elements included in the knowledge base.

2. Multimedia server (20) according to the preceding claim, in which the multimedia server (20) is made in the form of an electronic box (50) comprising a backplane card (52) and a plurality of electronic cards (54) connected to the backplane card (52).

3. Multimedia server (20) according to the preceding claim, in which the generative artificial intelligence model is implemented on a dedicated electronic card (54).

4. Multimedia server (20) according to any one of the preceding claims, wherein the generation module (30) is configured to generate additional customized content based on a user of said at least one entertainment terminal.

5. Multimedia server (20) according to any one of the preceding claims, wherein the generation module (30) is configured to generate additional content adapted according to the flight characteristics of the aircraft (10) and / or according to one or more editorial constraint(s) and / or according to one or more regulatory constraint(s).

6. Multimedia server (20) according to any one of the preceding claims, wherein the generative artificial intelligence model comprises a neural network based on a self-attentive model or a generative adversarial neural network.

7. Multimedia server (20) according to the preceding claim, wherein said self-attentive model-based neural network is a large language model.

8. Multimedia server (20) according to any one of the preceding claims, wherein the additional content is selected from the group consisting of: multimedia content such as film, television program, video game or audio content; text content and photo content.

9. Electronic entertainment system (16) intended to be installed on board an aircraft (10), the system (16) comprising: - at least one entertainment terminal; and - a multimedia server (20) connected to each entertainment terminal via an embedded local network, the multimedia server (20) being according to any one of the preceding claims.

10. A method, implemented by a multimedia server (20) intended to be installed on board an aircraft (10) and connected to at least one onboard entertainment terminal on board said aircraft (10) via an onboard local area network, for generating additional content for said at least one entertainment terminal, the multimedia server (20) comprising an additional content generation module (30), a knowledge base (34) storing a set of entertainment content elements, and a query module (32), the additional content generation module (30) comprising a generative artificial intelligence model, said generative artificial intelligence model being configured to generate additional content from the set of entertainment content elements stored in the knowledge base (34), 15. The process includes the following steps: - the reception (100), by the request module (32), of a command instruction from said at least one entertainment terminal, - based on the received command instruction, the emission (110), by the request module (32), of a request to generate additional content destined for the additional content generation module (30), - the reception (120), by the additional content generation module (30), of the additional content generation request issued, - on the basis of the additional content generation request received, the generation (130), by the additional content generation module (30), of additional content for said at least one entertainment terminal, each additional content generated by the generative artificial intelligence model being a content distinct from the content elements included in the knowledge base.

11. Method according to the preceding claim, wherein the step (130) of generating additional content for said at least one entertainment terminal is performed on the fly.

12. Method according to claim 10 or 11, wherein the generative artificial intelligence model is trained on the ground before the multimedia server (20) is loaded onto the aircraft (10), the generative artificial intelligence model making inferences from its training data when the multimedia server (20) is loaded onto the aircraft (10).

13. A method according to any one of claims 10 to 12, wherein the generative artificial intelligence model is trained by automatic machine or by fine-tuning.

14. Computer program comprising software instructions which, when executed by a computer, implement a method according to any one of claims 10 to 13.

Images