In-flight entertainment server on an aircraft, providing linguistic translation of announcements

Abstract

IN-FLIGHT ENTERTAINMENT SERVER ON AN AIRCRAFT, PROVIDING A LANGUAGE TRANSLATION OF ANNOUNCEMENTS This IFE computer server (150) includes at least one network interface (260), at least one processor (270) connected to communicate via at least one network interface (260) with video display units and / or personal electronic devices (122); and at least one memory (280).At least one memory (280) includes a language conversion module (210) for translating an announcement comprising digitized speech and / or text from a first language into a set of available translated languages; a passenger preference module (200) for identifying a set of passenger preference languages ​​based on information obtained for passengers on board the aircraft; and an announcement broadcasting module (230) for receiving the announcement in the first language generated by a crew member or a control device, and for processing the announcement in the first language through the language conversion module (210) to produce a set of translated announcements. See Figure 2 for abbreviations.

Description

Title of the invention: In-flight entertainment server on an aircraft, providing linguistic translation of announcements. Technical field

[0001] This disclosure relates to the provision of notifications in the environment of an aircraft or other vehicle. BACKGROUND

[0002] Announcements made by airline crews to passengers in aircraft cabins are primarily based on sound, which makes them unintelligible to people who speak another language. Crew members attempt to manage this problem by repeating the announcements in several languages.

[0003] However, the availability of multilingual crew members on these flights and the amount of multilingual repetition that passengers and crew will consider acceptable are limited in practice. SUMMARY

[0004] The invention relates to an in-flight entertainment computer server on an aircraft, referred to as an IFE computer server, comprising:

[0005] at least one network interface;

[0006] at least one processor connected to communicate via at least one network interface with video display units, each referred to as VDU, and / or personal electronic devices, each referred to as PED; and

[0007] at least one memory comprising:

[0008] a language conversion module configured to translate an advertisement comprising digitized speech and / or text from a first language into a set of available translated languages;

[0009] a passenger preference language module configured to identify a set of passenger preference languages ​​based on information obtained for passengers on board an aircraft and managed by the IFE computer server; and

[0010] an ad serving module configured to:

[0011] receive the announcement in the first language generated by a crew member or by a control device;

[0012] process the announcement in the first language via the language conversion module in order to produce a set of translated announcements, each corresponding to a different language from among the set of passengers' preferred languages; and

[0013] for each passenger having a preferred language different from the first language of the announcement, identify a VDU and / or a PED associated with a passenger and route through an aircraft cabin network one of the announcements translated into a preferred language of the passenger to the VDU and / or the PED associated with the passenger.

[0014] According to other advantageous aspects of the invention, the IFE computer server comprises one or more of the following features, taken individually or in all technically possible combinations:

[0015] - the language conversion module includes a learning model automatic trained to associate input digitized speech and / or text in order to generate digitized speech and / or text in a plurality of languages ​​from the available set of translated languages;

[0016] - the machine learning model is trained using a set of pre-configured passenger announcements to translate any of the pre-configured passenger announcements from a first language in the translated language set to any of the other translated languages ​​in the translated language set;

[0017] - the machine learning model is further trained using a value of probability that indicates the likelihood that a preconfigured passenger announcement from the set of preconfigured passenger announcements will be received during one or more defined aircraft operating states in a set of defined aircraft operating states;

[0018] - the language conversion module is configured to access a repository storing a set of preconfigured passenger announcements, each of the preconfigured passenger announcements in the set is stored in a plurality of languages ​​corresponding to the set of translated languages, and each of the preconfigured passenger announcements in the set is stored with a probability value indicating the probability that the preconfigured passenger announcement will be received during one or more defined operational states of the aircraft in a set of defined operational states of the aircraft;

[0019] - the ad serving module is further configured to:

[0020] obtain aircraft status data from the aircraft cabin network indicating that the aircraft is currently in one of a set of defined operational states of the aircraft when the announcement is received, and

[0021] Match the received advertisement to one of the pre-configured passenger advertisements in the set by correlating a content model of the received advertisement to a content model of one of the pre-configured passenger advertisements in the set and correlating the probability value indicating the probability that the advertisement pre-configured passenger service occurs in the current operational state indicated for the aircraft;

[0022] - the set of defined operational states of the aircraft comprises a plurality of elements including aircraft door closure, aircraft taxiing, takeoff and climb to cruising altitude, turbulence, in-flight emergencies, cruising altitude, descent approach for landing, ground emergencies and aircraft door opening;

[0023] - the information obtained for passengers on board the aircraft and managed by the IFE computer servers include at least one of the following pieces of information: the language settings a passenger has selected for a VDU and / or PED user interface associated with the passenger, information indicating a country of origin, and information indicating the passenger's nationality; and

[0024] - the ad serving module is further configured to route one translated announcements in the passenger's preferred language to the VDU and / or PED associated with the passenger, alongside a video via at least one network interface to the same VDU and / or PED for simultaneous display.

[0025] The invention also relates to a method implemented by an in-flight entertainment computer server on board an aircraft, referred to as an IFE computer server, the method comprising:

[0026] the reception of an announcement including speech and / or digitized text in a first language generated by a crew member or by a control device;

[0027] the processing of the announcement, by means of a language conversion module, in the first language in order to produce a set of translated announcements each corresponding to a different language from among the set of the passengers' preferred languages;

[0028] the translation, via the language conversion module, of the advertisement from a language to a set of available translated languages;

[0029] the identification of a set of passengers' preferred languages ​​based on information obtained for passengers on board an aircraft and managed by the IFE computer server; and

[0030] for each passenger having a preferred language different from the first language of the announcement, the identification of a video display unit, referred to as VDU, and / or a personal electronic device, referred to as PED, associated with a passenger and the routing through a network of the aircraft cabin of one of the announcements translated into a preferred language of the passenger to the VDU and / or the PED associated with the passenger.

[0031] According to other advantageous aspects of the invention, the method comprises one or more of the following features, taken individually or in all technically possible combinations:

[0032] - the method further comprises training a learning model automatic to associate input digitized speech and / or text in order to generate digitized speech and / or text in a plurality of languages ​​from the set of available translated languages;

[0033] - the machine learning model is trained using a set of pre-configured passenger announcements to translate any of the pre-configured passenger announcements from a first language in the translated language set to any of the other translated languages ​​in the translated language set;

[0034] - the machine learning model is further trained using a value probability which indicates a probability that a preconfigured passenger announcement from the set of preconfigured passenger announcements will occur in a current operational state of the aircraft;

[0035] - the method further comprises accessing a repository storing a set of preconfigured passenger announcements, each of the preconfigured passenger announcements in the set stored in a plurality of languages ​​corresponding to the set of translated languages, and each of the preconfigured passenger announcements in the set is stored with a probability value indicating the probability that the preconfigured passenger announcement will be received during one or more defined operational states of the aircraft in a set of defined operational states of the aircraft;

[0036] - the process further comprises:

[0037] obtaining aircraft state data from the aircraft cabin network indicating that the aircraft is currently in one of a set of defined operational states of the aircraft when the announcement is received, and

[0038] matching the received announcement with one of the pre-configured passenger announcements in the set by correlating a content model of the received announcement with a content model of one of the pre-configured passenger announcements in the set and correlating the probability value indicating the probability that the pre-configured passenger announcement will occur in the indicated current operational state of the aircraft;

[0039] - the set of defined operational states of the aircraft comprises a plurality of elements including aircraft door closure, aircraft taxiing, takeoff and climb to cruising altitude, turbulence, in-flight emergencies, altitude of cruise, the descent approach for landing, ground emergencies and the opening of the aircraft doors.

[0040] - the information obtained for passengers on board the aircraft and managed by the IFE computer servers include at least one of the following pieces of information: the language settings a passenger has selected for a VDU and / or PED user interface associated with the passenger, information indicating a country of origin, and information indicating the passenger's nationality; and

[0041] - the method further comprises the delivery of one of the translated advertisements in the passenger's preferred language to the VDU and / or PED associated with the passenger, alongside a video to the same VDU and / or PED for simultaneous display.

[0042] The invention also relates to a computer-readable, non-transient storage medium on which a computer program is stored, comprising instructions which, when executed on at least one processor, cause the at least one processor to perform operations comprising:

[0043] the reception of an announcement including speech and / or digitized text in a first language generated by a crew member or by a control device;

[0044] the processing of the announcement, by means of a language conversion module, in the first language in order to produce a set of translated announcements each corresponding to a different language from among the set of the passengers' preferred languages;

[0045] the translation, via the language conversion module, of the advertisement from a language to a set of available translated languages;

[0046] the identification of a set of passengers' preferred languages ​​based on information obtained for passengers on board an aircraft and managed by an on-board in-flight entertainment computer server, referred to as an IFE computer server; and

[0047] for each passenger having a preferred language different from the first language of the announcement, the identification of a video display unit, referred to as a VDU, and / or a personal electronic device, referred to as a PED, associated with a passenger and the routing through an aircraft cabin network of one of the announcements translated into a passenger's preferred language to the VDU and / or PED associated with the passenger. Brief description of the drawings

[0048] The accompanying drawings, which are included to provide a better understanding of the disclosure and are incorporated into and form part of this application, illustrate certain non-limiting embodiments of the invention. In the drawings:

[0049] [Fig.1] Fig.1 illustrates a simplified functional diagram of an IFE system comprising an IFE server configured to provide announcements translated into a passenger's preferred language, according to certain embodiments;

[0050] [Fig.2] Fig.2 illustrates a simplified functional diagram of the IFE server of Fig.1, configured to process an announcement in the form of digitized speech and / or text input, translate the announcement into a set of translated languages, and route at least one of the translated announcements to a passenger's VDU and / or PED, according to certain embodiments; and

[0051] [Fig.3] Fig.3 illustrates an example of a machine learning model trained to associate input digitized speech and / or text in order to generate digitized speech and / or text in a plurality of languages ​​from the set of available translated languages, according to certain embodiments. DETAILED DESCRIPTION

[0052] In the detailed description that follows, numerous specific details are set forth to enable a thorough understanding of certain aspects of the invention. However, those skilled in the art will understand that the present invention can be implemented without these specific details. In other instances, well-known methods, procedures, components, and circuits have not been described in detail so as not to obscure the present invention.

[0053] Although various embodiments are explained here in the context of translating an announcement and routing a translated announcement to a passenger in an aircraft in-flight entertainment (IFE) environment for display to passengers, other embodiments are not limited to these and can be used with other types of vehicles, including, without limitation, ships (e.g., cruise ships), buses, and trains, and in non-vehicle-related environments (e.g., television, streaming video services, audio announcement systems, etc.). Therefore, the embodiments are not limited to use with passengers but can be used for anyone with a hearing impairment or who prefers to see announcements in a language other than that used in the original announcement.The terms "announcements" and "notifications" are used interchangeably and may have the same meaning in this document.

[0054] Certain embodiments of this disclosure aim to solve a long-standing problem related to the broadcasting of passenger announcements in the cabin of an aircraft and how to make these announcements accessible and acceptable to passengers who speak another language or who are deaf. The term "deaf passengers" is used here to refer to any person with a hearing impairment that makes Hearing the announcements is more difficult. The technical challenge lies in the lack of effective communication and situational awareness for flight passengers who do not understand the announcements due to their inability to understand the language(s) of the in-flight announcements.

[0055] Currently, the majority of passenger announcements in aircraft cabins are delivered via audio messages broadcast through the aircraft's loudspeakers and headphone interfaces. These announcements fall into several categories, including passenger announcements (PAs), prerecorded announcements (PRAMs), and video announcements (generally for safety videos). Announcements can cover a wide range of topics, from superficial information such as duty-free shopping to messages that may impact the passenger travel experience (connecting flight information, delay notifications) or even passenger safety (turbulence notifications).Passengers who do not speak the language of the announcement and deaf passengers, however, receive little information from these announcements, generally relying on visual cues such as cabin signs or video announcements with subtitles in one or more written languages, if available.

[0056] The use of live audio for these messages may have major disadvantages, including the following: it is not suitable for passengers with hearing impairments, for passengers who do not understand the language of the announcement and may be missed by passengers in many situations (e.g. if the passenger is asleep, if they are distracted by a conversation with another passenger or if they are distracted by viewing content).

[0057] Various embodiments aim to improve communication and situational awareness for passengers through an IFE server that translates announcements into digital speech and / or text into one or more languages ​​preferred by the passengers and routes the translated announcements to the passengers for them to view in text form.

[0058] In some embodiments, the IFE server uses an aircraft acquisition module and an announcement management module (described in more detail with reference to [Fig. 2]) and connects them to a live, AI-based speech-to-text (STT) service to generate one or more translated announcements. All translated announcements are then stored and presented to passengers via interfaces such as a REST (Representational State Transfer) application programming interface (API). In some embodiments, this is done live while the announcement is being projected through a loudspeaker for the other passengers. In other embodiments, the translation and / or routing of The translated announcement to the passenger(s) is subject to a predefined delay, or the translated announcements are provided to the passengers at their request, using their VDU and / or their PED.

[0059] Announcements may be used to inform passengers of events and messages initiated by the crew, relating in particular to the following topics: purchases, promotion of an airline credit card, closing of aircraft doors; opening of aircraft doors; start of taxiing and takeoff procedure; start of approach and landing procedure; start of taxiing and takeoff procedure; start of approach and landing procedure; activation of a seatbelt fastening signal; forecast of turbulence; deactivation of the seatbelt fastening signal; depressurization of the aircraft cabin; etc.

[0060] Fig. 1 illustrates a simplified functional diagram of an IFE 100 system configured to provide announcements translated into a passenger's preferred language, according to certain embodiments.

[0061] With reference to [Fig. 1], the IFE system 100 comprises video display units (VDUs) 120 and VDUs and / or personal electronic devices (PEDs) 122 that communicate via communication distribution components 110 with an IFE server 150 and a communication terminal 130. Some VDUs communicate via wired networks (e.g., Ethernet) through seat electronic boxes 112, and other VDUs and / or PEDs 122 communicate via one or more wireless routers 114, such as a WiFi access point and / or a cellular access point (e.g., a pico-cellular radio base station) of distribution components 110. The VDUs may include, but are not limited to, seat displays that are mounted on individual seats, e.g. on seat backs or armrests.Examples of PED 122 devices that can be transported on board an aircraft may include, but are not limited to, smartphones, tablets, laptops, gaming computers, etc.

[0062] The PEDs can be communicatively connected to the IFE 150 server, for example by using WiFi communications routed through the wireless router 114 and / or by using Bluetooth communications routed through Bluetooth pairings between the PEDs and the VDUs 120.

[0063] The IFE server 150 can provide on-demand broadcasting and / or distribution of content requested by passengers. The IFE server 150 can receive announcements from a crew interface 160, cockpit equipment, and / or an aircraft notification interface (e.g., connected to an aircraft interface converter) via network 140. The communication terminal 130 can communicate via satellite networks with one or more radio access networks. on the ground 172 (e.g., ground relays) and can, via direct air-to-ground communications (DA2GC), communicate with 3GPP NR 5G cellular base stations (e.g., eNodeB, gNodeB, etc.), etc. In some optional embodiments, the IFE system 100 communicates via wireless networks 170 with an IFE system operating platform 105 that controls access to the content of an IFE content provider platform 104 and / or other networked content servers such as Internet content servers 174.

[0064] VDUs and / or PEDs can send requests to the IFE server 150 via the distribution components 110 and receive forwarded and translated announcements from the IFE server via the distribution components 110.

[0065] [Fig.2] illustrates a simplified functional diagram of the IFE 150 server of [Fig.1], configured to process an announcement in the form of digitized speech and / or text input, translate the announcement into a set of translated languages ​​and route at least one of the translated announcements to a VDU and / or a PED of a passenger 250, according to certain embodiments.

[0066] The IFE system can trigger notifications and / or announcements to passengers via an interface with an aircraft notification system (using various types and protocols of messaging) or via a crew public address system 240. When a notification or announcement is triggered by the aircraft notification system or the crew public address system 240, the crew interface 160 interprets the action to be taken and broadcasts the corresponding sound to the spaced loudspeakers 290 in the cabin. In addition, or alternatively, the crew public address system 240 can be used by a crew member to send a digital text to all or part of the aircraft's VDUs and / or PEDs. For example, a crew member can type a message or select a pre-configured passenger announcement on a control device communicatively coupled to the crew interface 160.The typed message or pre-configured announcement selected for passengers may include text to be sent to one or more VDUs and / or PEDs or be converted into an audio message to be broadcast through the speakers 290. In the context of this innovation, these audio messages (i.e. digitized speech) and / or texts are also sent to the IFE server 150.

[0067] A crew member can make an announcement using the crew public address system 240, and the crew interface 160 can broadcast an audio signal and a discrete signal in accordance with ARINC-628P3. The audio signal can be transmitted to the loudspeaker 290, and both the audio and discrete signals can be routed to the aircraft acquisition module 232 of the IFE server 150.

[0068] The IFE server 150 includes at least one network interface 260; at least one processor 270 connected to communicate via the at least one network interface 260 with video display units (VDUs) and / or personal electronic devices (PEDs) (e.g., VDU 120 or VDU and / or PED 122); and at least one memory 280. The at least one memory 280 includes a language conversion module 210. The language conversion module 210 enables the translation of an announcement (e.g., the audio signal sent to the aircraft acquisition module 232) comprising digitized speech and / or text from a first language (i.e., the language used in the announcement) into an available set of translated languages. For example, the 210 language conversion module can translate an advertisement from English into a plurality of other languages, including French, Spanish, Italian, Dutch, etc.

[0069] The memory 280 further includes a passenger preference language module 200 which makes it possible to identify a set of passenger preference languages ​​based on information obtained for passengers on board an aircraft and managed by the IFE computer server. In some embodiments, the information on preference languages ​​obtained for passengers on board the aircraft and managed by the IFE computer server includes at least one of the following: information indicating the language settings that a passenger has selected for a user interface of the VDU and / or the PED associated with the passenger, information indicating the preferred language defined by the passenger, information indicating a country of origin, information indicating the passenger's nationality, etc.For example, if a passenger has chosen Spanish as the text display language on their PED, the passenger's preferred language can then include Spanish, and the passenger's preferred language module can add Spanish to the passenger's set of preferred languages.

[0070] The Passenger Preferred Language Module 200 can identify a preferred language for each passenger on board the aircraft or for a subset of passengers on board the aircraft. For example, the Passenger Preferred Language Module 200 can be configured to identify only one preferred language for passengers (e.g., passenger 250) who have a VDU and / or a PED associated with them. The identification of a VDU and / or a PED associated with a passenger is discussed in more detail below.

[0071] The memory 280 further includes an announcement broadcast module 230. The announcement broadcast module 230 is used to receive an announcement in a first language generated by a crew member or by a control device. For example, the aircraft acquisition module 232 can receive the announcement and forward it to the announcement management module 234 and the speech transcription module. 236. In some embodiments, the ad management module 234 of the ad delivery module 230 is used to detect the start and end of the ad.

[0072] The announcement broadcasting module 230 further serves to process the announcement in the first language via the language conversion module 210 in order to produce a set of translated announcements, each corresponding to a different language from among the set of passengers' preferred languages. For example, if it is established that the passengers on board have 10 different language preferences, the set of passengers' preferred languages ​​can then contain all the languages ​​supported by the language conversion module 210. The language conversion module 210 can translate the announcement in the first language into each of the passengers' preferred languages ​​in the set that is different from the first language (it can, for example, produce the announcement in 9 translated languages ​​plus the original language of the announcement if the first language is part of the set of 10 passengers' preferred languages).

[0073] In cases where conversion and routing of the announcement are carried out live, the announcement delivery module 230 can start and stop live speech transcription processing based on the detected start and end of the live audio stream forming the announcement.

[0074] The processing (e.g., translation) of the announcement may result in the production of translated announcements in the form of transcripts. In some embodiments, after speech transcription processing has stopped, the announcement management module 234 may retrieve the transcript(s) for local storage on the IFE server 150 or elsewhere on board the aircraft. However, the transcripts may be stored remotely from the aircraft to reduce the aircraft's required storage capacity. Storing the transcripts on the aircraft may be advantageous so that the IFE server 150 can access them more quickly.

[0075] The announcement broadcast module 230 further enables, for each passenger with a preferred language different from the first language of the announcement, the identification of a VDU and / or PED (for example, by their network address) associated with a passenger and the routing, through an aircraft cabin network, of one of the announcements translated into the passenger's preferred language to the VDU and / or PED associated with the passenger. The announcement broadcast module 230 can identify a VDU and / or PED associated with a passenger based on the passenger's seat assignment (for example, in cases where the VDU is a seatback VDU oriented towards the passenger's assigned seat), information indicating a PED registered for the passenger (for example, a passenger's phone that includes information indicating a phone owner or user), etc. After receiving the translated announcement, The VDU and / or the PED may display the translated announcement in text for the passenger to read.

[0076] In certain embodiments, the ad delivery module 230 is further used to route one of the translated ads in the passenger's preferred language to the passenger's associated VDU and / or PED, concurrently with a video via at least one network interface to the same VDU and / or PED for simultaneous display. The translated ad can be displayed, for example, in a pop-up window or superimposed on other displayed video content. This can allow a passenger watching a movie to visually see the translated ad without interrupting the movie.

[0077] Passengers' preferred languages ​​can include any number of languages. Identifying all passenger preferred languages ​​can include identifying a hearing-impaired passenger. If a passenger is identified as hearing-impaired, but their preferred language is the same as the first language of the announcement, the announcement delivery module 230 can further treat the hearing-impaired passenger as having the same preferred language as the announcement if the announcement is to be displayed as text for that passenger. However, if the announcement is to be broadcast as audio through the cabin public address system or the seat speaker connections, the announcement delivery module 230 can consider that the hearing-impaired passenger has a preferred language different from the first language.For example, if the passenger's preferred language is Spanish and the first language of the announcement is Spanish, but the announcement needs to be broadcast over a loudspeaker to a hearing-impaired passenger, the 230 announcement delivery module can then identify a VDU and / or PED associated with the hearing-impaired passenger and route one of the announcements translated into the passenger's preferred language (e.g., the announcement text in Spanish or another preferred language) through an aircraft cabin network to the VDU and / or PED associated with the passenger.

[0078] In this document, the term “speech” encompasses all forms of human vocalization, including speaking or singing.

[0079] In some embodiments, the language conversion module 210 further serves to access a repository storing a set of preconfigured passenger announcements, each of the preconfigured passenger announcements in the set is stored in a plurality of languages ​​corresponding to the set of translated languages, and each of the preconfigured passenger announcements in the set is stored with a probability value indicating the probability that the preconfigured passenger announcement will be received during one or more defined operational states of the aircraft in a set of defined operational states of the aircraft.

[0080] Preconfigured passenger announcements may include all types of planned announcements that can be broadcast to passengers in any aircraft operating state. Preconfigured passenger announcements may include, for example, more than 100 different types of preconfigured passenger announcements, with corresponding indications of the defined aircraft operating states in which they may respectively occur. The defined aircraft operating states may include aircraft door closure, aircraft taxiing, takeoff and climb to cruising altitude, turbulence, in-flight emergencies, cruising altitude, descent approach for landing, ground emergencies, and aircraft door opening.For example, if the aircraft is currently experiencing or will soon experience turbulence, the probability value will be higher for pre-configured passenger announcements related to fastening seat belts, informing passengers of turbulence, and asking them to return to their seats, and it will be lower for pre-configured passenger announcements related to duty-free shopping at the destination airport.

[0081] The announcement broadcasting module can be used to obtain aircraft state data from the aircraft cabin network, indicating that the aircraft is currently in one of a set of defined operational states at the time the announcement is received, and to match the received announcement to one of the preconfigured passenger announcements in the set. This is achieved by correlating a content model of the received announcement with a content model of one of the preconfigured passenger announcements in the set, and by correlating the probability value indicating the likelihood that the preconfigured passenger announcement will occur in the specified current operational state of the aircraft. The model matching can include matching the spectral frequency envelopes and the frame rate between the announcement and one of the preconfigured announcements in the set.

[0082] The repository can be stored locally in the aircraft (for example, in memory 280 or elsewhere) and can be accessed via at least one network interface 260.

[0083] In some embodiments, the language conversion module includes a machine learning model trained to associate input digitized speech and / or text in order to generate digitized speech and / or text in a plurality of languages ​​from the set of available translated languages.

[0084] Figure [Fig. 3] illustrates an example of a machine learning model (i.e. the ad model 500) which is trained to associate input digitized speech and / or text in order to generate digitized speech and / or text in a plurality of languages ​​from the set of translated languages ​​available, according to certain embodiments.

[0085] With reference to [Fig. 3], the language conversion module 210 may include a machine learning training module 510 that operates in training mode to train an announcement model 500 using a set of preconfigured passenger announcements 520. The set of preconfigured passenger announcements 520 may function as a repository storing a set of preconfigured passenger announcements that may include all types of planned announcements that may be broadcast to passengers in any operational state of the aircraft. Each of the preconfigured passenger announcements in the set 520 is stored in a plurality of languages ​​corresponding to a set of translated languages.The set of translated languages ​​can be selected based on information indicating which spoken languages ​​are expected to be preferred by passengers in the regions of the world where the aircraft is likely to be operated during its lifetime. Each of the preconfigured passenger announcements in set 520 can be stored with a probability value indicating the likelihood that the preconfigured passenger announcement will be received during one or more defined operational states of the aircraft within the set of defined operational states of the aircraft.

[0086] Accordingly, in certain embodiments, the announcement model 500 can be trained using a set of preconfigured passenger announcements 520 to translate any of the defined passenger announcements from a first language in the set of translated languages ​​into any of the other translated languages ​​in the set of translated languages. The announcement model 500 can also be trained using a probability value that indicates the probability that a defined passenger announcement from the set of defined passenger announcements will be received during one or more defined aircraft operating states within a set of defined aircraft operating states.

[0087] The set of defined operational states of the aircraft may include, for example, a plurality of elements among: the closing of the aircraft doors, the taxiing of the aircraft, the takeoff and climb to cruising altitude, turbulence, in-flight emergencies, cruising altitude, the descent approach for landing, ground emergencies and the opening of the aircraft doors.

[0088] In this way, the announcement model 500 can function as a reduced language model trained by supervised learning to recognize with high accuracy the live announcements that correspond to one or more preconfigured passenger announcements in the set 520. The set can include, for example, more than 100 different types of preconfigured passenger announcements, with corresponding indications of the defined operational states of the aircraft in which they can respectively occur. The announcement model 500 can to be used to translate, perhaps with less accuracy, live announcements that do not correspond to one or more pre-configured passenger announcements of set 520. The use of a reduced language model configured in this way can substantially reduce the memory and processing requirements of the computing platform operating as an IFE 150 server according to the embodiments described in this document.

[0089] In execution mode, a live announcement from the crew interface 160 is received by the announcement model 500 and transmitted by it, along with an indication of the aircraft's current operational status corresponding to one of the aircraft's defined operational statuses at the time the announcement was received, in order to produce a set of translated language versions of the live announcement. The announcement model 500 can match the live announcement to one of the preconfigured passenger announcements in set 520 by correlating a content model of the live announcement to a content model of one of the preconfigured passenger announcements in set 520 and by correlating the probability value of the preconfigured passenger announcement to the aircraft's status data.

[0090] The announcement model 500 may include a neural network model having hierarchical layers of summation nodes with weights and trigger thresholds that are trained by the machine learning training module 510 to perform matching and correlation. Alternatively or in addition, the announcement model 500 may include program instructions that logically match content models of the live announcement to content models of one of the preconfigured passenger announcements in the set 520.When the matching indicates that more than one of the pre-configured passenger announcements in set 520 are likely to match the live announcement, a selection among the candidates can be made based on the probability value indicating which has the highest probability of occurring in the current operational state of the aircraft compared to the probability of occurrence of the other candidates.

[0091] A 530 translated announcement routing module obtains information on each passenger's preferred language and determines the network addresses of the VDUs and / or PEDs of the seats providing service to each passenger. For each passenger, the 530 translated announcement routing module obtains a version from the set of translated language versions of the live announcement that corresponds to the passenger's preferred language and routes this translated language version of the live announcement to the network address of the VDU and / or PED of the seat providing service to that passenger.

[0092] Other definitions and embodiments:

[0093] In the above description of various embodiments of this disclosure, it is understood that the terminology used herein is solely for the purpose of describing particular embodiments and is not intended to limit the invention. Unless otherwise defined, all terms (including technical and scientific terms) used in this document have the same meaning as that commonly understood by a person in the field to which this disclosure relates. It is also understood that terms, such as those defined in dictionaries in common use, should be interpreted in a manner consistent with their meaning in the context of this specification and the relevant art and should not be interpreted in an idealized or overly formal sense unless expressly defined as such in this document.

[0094] When an element is said to be "connected," "coupled," "responsive," or variants of these terms, to another element, it may be directly connected, coupled, or reactive to the other element, or intermediate elements may be present. Conversely, when an element is designated as being "directly connected," "directly coupled," "directly reactive," or variants of these terms, to another element, there are no intermediate elements present. Similar numbers refer to similar elements throughout this document. Furthermore, the terms "coupled," "connected," "responsive," or variants of these terms, as used herein, may include the terms wirelessly coupled, connected, or reactive. Well-known functions or constructions may not be described in detail for reasons of brevity and / or clarity.The term "and / or" includes any combination and all combinations of one or more of the listed associated elements.

[0095] In this document, the terms "include," "comprising," "comprises," "include," "includes," "have," "has," "having," or variants thereof are open-ended and include one or more features, one or more integers, elements, steps, components, or functions, but do not exclude the presence or addition of one or more other features, one or more other integers, elements, steps, components, functions, or groups thereof. Furthermore, in this document, the common abbreviation "e.g.", meaning "for example," may be used to introduce or specify one or more general examples of a previously mentioned element and is not intended to be restrictive of that element. The common abbreviation "i.e.", meaning "that is," may be used to specify a particular element of a more general enumeration.

[0096] Examples of embodiments are described here with reference to functional diagrams and / or illustrations of flowcharts of computer-implemented processes, devices (systems and / or apparatus) and / or program outputs computer systems. It is understood that a block of the functional diagrams and / or flowchart illustrations, and combinations of blocks in the functional diagrams and / or flowchart illustrations, can be implemented by computer program instructions executed by one or more computer circuits.These computer program instructions can be provided to a processor circuit of a general-purpose computer circuit, a special-purpose computer circuit, and / or other programmable data processing circuit to produce a machine, such that the instructions, which are executed through the computer processor and / or other programmable data processing device, transform and control transistors, values ​​stored in memory locations, and other hardware components within these circuits to implement the functions / actions specified in the functional diagrams and / or the block(s) of the flowchart, and thus create means (a functionality) and / or a structure for implementing the functions / actions specified in the functional diagrams and / or the block(s) of the flowchart.

[0097] These computer program instructions can also be stored in a non-transient computer-readable medium that can command a computer or other programmable data processing device to operate in a particular way, so that the instructions stored in the computer-readable medium produce a manufactured article comprising instructions that implement the functions / actions specified in the functional diagrams and / or the flowchart block(s).

[0098] A non-transient, computer-readable medium may include an electronic, magnetic, optical, electromagnetic, or semiconductor data storage system, apparatus, or device. More specific examples of computer-readable media include: a portable computer diskette, a random access memory (RAM) circuit, a read-only memory (ROM) circuit, an erasable programmable read-only memory (EPROM or Flash memory) circuit, etc.

[0099] Computer program instructions can also be loaded onto a computer and / or other programmable data processing device to cause a series of operational steps to be performed on the computer and / or other programmable device in order to produce a computer-implemented process such that the instructions executing on the computer or other programmable device provide steps to implement the functions / actions specified in the functional diagrams and / or the flowchart block(s). Accordingly, embodiments of this disclosure can be implemented in hardware and / or software (including firmware, resident software, microcode, etc.) that runs on a processor such as a digital signal processor, which can be collectively referred to as "circuit", "module" or variants thereof.

[0100] It should also be noted that, in some alternative embodiments, the functions / actions indicated in the blocks may occur in a different order than that shown in the flowcharts. For example, two blocks illustrated successively may in fact be executed substantially at the same time, or the blocks may sometimes be executed in reverse order, depending on the functionality / actions concerned. Furthermore, the functionality of a given block in the flowcharts and / or functional diagrams may be separated into several blocks, and / or the functionality of two or more blocks in the flowcharts and / or functional diagrams may be at least partially integrated. Finally, other blocks may be added / inserted between the illustrated blocks.Furthermore, although some of the diagrams include arrows on the communication channels to indicate a primary direction of communication, it should be understood that communication can occur in the opposite direction to the arrows shown.

[0101] Numerous different embodiments have been disclosed in this document, in connection with the above description and drawings. It is understood that it would be unduly repetitive and confusing to describe and illustrate literally all the combinations and subcombinations of these embodiments. Accordingly, this specification, including the drawings, is to be interpreted as constituting a complete written description of various examples of combinations and subcombinations of embodiments, as well as the manner and process of manufacturing and using them, and is to support claims relating to any such combination or subcombination.

[0102] Numerous variations and modifications can be made to the embodiments without deviating substantially from the principles of the present invention. All such variations and modifications are deemed to fall within the scope of the present invention.

Claims

Demands

1. An in-flight entertainment computer server on an aircraft, referred to as the IFE computer server (150), comprising: at least one network interface (260); at least one processor (270) connected to communicate via the at least one network interface (260) with video display units, each referred to as VDUs, and / or personal electronic devices, each referred to as PEDs (122); and at least one memory (280) comprising: a language conversion module (210) configured to translate an announcement comprising speech and / or digitized text from a first language into an available set of translated languages; a passenger preference language module (200) configured to identify a set of passenger preference languages ​​based on information obtained for passengers on board an aircraft and managed by the IFE computer server (150);and an announcement broadcasting module (230) configured to: receive the announcement in the first language generated by a crew member or a control device; process the announcement in the first language via the language conversion module (210) to produce a set of translated announcements, each corresponding to a different language from the set of passenger preferred languages; and for each passenger with a preferred language different from the first language of the announcement, identify a VDU and / or PED (122) associated with a passenger and route through an aircraft cabin network one of the announcements translated into the passenger's preferred language to the VDU and / or PED associated with the passenger.

2. Server (150) according to claim 1, wherein the language conversion module (210) comprises a machine learning model (510) trained to associate input digitized speech and / or text in order to generate digitized speech and / or text in a plurality of languages ​​from the available set of translated languages.

3. Server (150) according to claim 2, wherein the machine learning model (510) is trained using a set of preconfigured passenger announcements to translate one any of the pre-configured passenger announcements from a first language in the translated language set to any of the other translated languages ​​in the translated language set.

4. Server (150) according to claim 3, wherein the machine learning model (510) is further trained using a probability value that indicates a probability that a preconfigured passenger announcement from the set of preconfigured passenger announcements will be received during one or more defined aircraft operating states in a set of defined aircraft operating states.

5. Server (150) according to claim 1, wherein the language conversion module (210) is configured to access a repository storing a set of preconfigured passenger announcements, each of the preconfigured passenger announcements in the set is stored in a plurality of languages ​​corresponding to the set of translated languages, and each of the preconfigured passenger announcements in the set is stored with a probability value indicating the probability that the preconfigured passenger announcement will be received during one or more defined aircraft operating states in a set of defined aircraft operating states.

6. Server (150) according to claim 5, wherein the announcement broadcasting module (230) is further configured to: obtain aircraft state data from the aircraft cabin network indicating that the aircraft is currently in a state of a set of defined aircraft operating states when the announcement is received, and match the received announcement to one of the preconfigured passenger announcements in the set by correlating a content pattern of the received announcement to a content pattern of one of the preconfigured passenger announcements in the set and correlating the probability value indicating the probability that the preconfigured passenger announcement will occur in the indicated current aircraft operating state.

7. Server (150) according to claim 5, wherein the set of defined operational states of the aircraft comprises a plurality of elements including aircraft door closure, taxiing of the aircraft, takeoff and climb to cruising altitude, turbulence, in-flight emergencies, cruising altitude, approach descent for landing, ground emergencies and opening of aircraft doors.

8. Server (150) according to claim 1, wherein the information obtained for passengers on board the aircraft and managed by the IFE computer server (150) includes at least one of the following: information indicating the language settings that a passenger has selected for a user interface of the VDU and / or PED (122) associated with the passenger, information indicating a country of origin, and information indicating the passenger's nationality.

9. Server (150) according to claim 1, wherein the ad delivery module (230) is further configured to route one of the ads translated into the passenger's preferred language to the VDU and / or PED (122) associated with the passenger, concurrently with a video via at least one network interface (260) to the same VDU and / or PED (122) for simultaneous display.

10. A method carried out by an in-flight entertainment computer server on board an aircraft, referred to as the IFE computer server (150), the method comprising: receiving an announcement including digitized speech and / or text in a first language generated by a crew member or by a control device; processing the announcement, by means of a language conversion module (210), in the first language in order to produce a set of translated announcements, each corresponding to a different language from among the set of passengers' preferred languages; translating, by means of the language conversion module (210), the announcement from one language into an available set of translated languages; identifying a set of passengers' preferred languages ​​based on information obtained for passengers on board an aircraft and managed by the IFE computer server (150);and for each passenger having a preferred language different from the first language of the announcement, the identification of a video display unit, referred to as VDU, and / or a personal electronic device, referred to as PED, (122) associated with a passenger and the routing through a; aircraft cabin network of one of the announcements translated into a passenger's preferred language to the VDU and / or PED (122) associated with the passenger.

11. A method according to claim 10, further comprising: training a machine learning model (510) to associate input digitized speech and / or text in order to generate digitized speech and / or text in a plurality of languages ​​from the set of available translated languages.

12. A method according to claim 11, wherein the machine learning model (510) is trained using a set of pre-configured passenger announcements to translate any one of the pre-configured passenger announcements from a first language in the set of translated languages ​​into any of the other translated languages ​​in the set of translated languages.

13. A method according to claim 12, wherein the machine learning model (510) is further trained using a probability value that indicates a probability that a preconfigured passenger announcement from the set of preconfigured passenger announcements will occur in a current operational state of the aircraft.

14. A method according to claim 10, further comprising: access to a repository storing a set of preconfigured passenger announcements, each of the preconfigured passenger announcements in the set stored in a plurality of languages ​​corresponding to the set of translated languages, and each of the preconfigured passenger announcements in the set is stored with a probability value indicating the probability that the preconfigured passenger announcement will be received during one or more defined aircraft operating states in a set of defined aircraft operating states.

15. A method according to claim 14, further comprising: obtaining aircraft state data from the aircraft cabin network indicating that the aircraft is currently in one of a set of defined operational states of the aircraft when the announcement is received, and matching the received announcement with one of the preconfigured passenger announcements from the set by correlating a content model of the received announcement to a content model of one of the pre-configured passenger announcements in the set and by correlating the probability value indicating the probability that the pre-configured passenger announcement will occur in the indicated current operational state of the aircraft.

16. A method according to claim 14, wherein the set of defined operational states of the aircraft comprises a plurality of elements from the closing of the aircraft doors, the taxiing of the aircraft, takeoff and climb to cruising altitude, turbulence, in-flight emergencies, cruising altitude, the descent approach for landing, ground emergencies and the opening of the aircraft doors.

17. A method according to claim 10, wherein the information obtained for passengers on board the aircraft and managed by the IFE computer server (150) includes at least one of the pieces of information indicating the language settings that a passenger has selected for a user interface of the VDU and / or the PED (122) associated with the passenger, information indicating a country of origin, and information indicating the nationality of the passenger.

18. A method according to claim 10, further comprising: routing one of the advertisements translated into the passenger's preferred language to the VDU and / or PED (122) associated with the passenger, in parallel with a video to the same VDU and / or PED (122) for simultaneous display.

19. A non-transient, computer-readable storage medium on which is stored a computer program comprising instructions which, when executed on at least one processor (270), cause the at least one processor (270) to perform operations comprising: receiving an announcement comprising digitized speech and / or text in a first language generated by a crew member or by a control device; processing the announcement, by means of a language conversion module (210), in the first language in order to produce a set of translated announcements, each corresponding to a different language from among the set of passengers' preferred languages; translating, by means of the language conversion module (210), the announcement from one language into an available set of translated languages; the identification of a set of passengers' preferred languages ​​based on information obtained for passengers on board an aircraft and managed by an onboard in-flight entertainment computer server, known as an IFE computer server (150); and for each passenger having a preferred language different from the first language of the announcement, the identification of a video display unit, called VDU, and / or a personal electronic device, called PED, (122) associated with a passenger and the routing through a network of the aircraft cabin of one of the announcements translated into a preferred language of the passenger to the VDU and / or the PED (122) associated with the passenger.

Images