METHOD AND AUDIOVISUAL DECODER FOR RECOMMENDING A LIST OF AT LEAST ONE AUDIOVISUAL PROGRAM

Abstract

METHOD AND AUDIOVISUAL DECODER FOR RECOMMENDING A LIST OF AT LEAST ONE AUDIOVISUAL PROGRAM The present invention relates to a method and an audiovisual decoder (10) for recommending a list of at least one audiovisual program to at least one user, the program(s) being categorized by genre and / or subgenre. According to the invention, the method (100) comprises the steps performed by said audiovisual decoder (10) of: - associating (E20) said genres / subgenres with emotional states; - detecting (E40) a current emotional state of said at least one user; - sorting (E60) said programs in said catalog, taking into account at least said detected emotional state and said genres and / or subgenres associated with said detected emotional state; - display (E80) of a list of at least one audiovisual program taking into account said sorting of said programs of said catalogue.In this way, the invention makes it possible to take into account the user's current emotional state in order to improve the relevance of the programs recommended to them. Fig. 3.

Description

Title of the invention: METHOD AND AUDIOVISUAL DECODER FOR RECOMMENDING A LIST OF AT LEAST AN AUDIOVISUAL PROGRAM Technical field

[0001] The technical field of the invention is that of the distribution of audiovisual content or programs, in particular. The present invention relates more particularly to a method for recommending a program or a list of audiovisual programs taking into account the user's emotions. STATE OF PRIOR ART

[0002] Program distribution platforms (particularly for audiovisual content) have increasingly extensive program catalogs. It is therefore sometimes difficult for users to search for and select a program that suits them. To facilitate this selection, platforms generally implement program recommendation processes that take into account, in particular, the program category and the number of times the programs have been viewed, in order to offer users the most popular programs in each category.

[0003] However, this technique is unsatisfactory because it does not take into account the user's current emotional state, and the programs offered to them are therefore not necessarily in line with the user's expectations at that moment. Indeed, depending on the user's emotional state, their preferences may differ and not correspond to the recommendations typically made by distribution platforms.

[0004] There is therefore a need to provide a method for recommending a program, particularly an audiovisual program, to a user, taking into account their emotional state in order to improve the relevance of the programs offered to them. Description of the invention

[0005] To this end, according to a first aspect, the invention proposes a method of recommending a list of at least one audiovisual program from a catalogue of programs to at least one user directed towards a decoder, in particular an audiovisual decoder, and display means, said programs of said catalogue being categorized into genres and / or subgenres.

[0006] In this description, the term catalogue refers to a database of programs available for playback by a user, here via an audiovisual decoder.

[0007] It should be noted that the present invention is described in particular with regard to the playback of audiovisual programs. It is readily understood that the present solution could be implemented for the recommendation of audio and / or visual type programs, such as podcasts, audiobooks, music, etc. or photos, in particular.

[0008] Thus, reading is understood to mean the fact that a user maintains the reproduction, by the audiovisual decoder 10, of a program for a predetermined period of time, regardless of the type of program.

[0009] The invention also relates to an audiovisual decoder capable of implementing such a recommendation process.

[0010] Thus, the present invention makes it possible to take into account the current emotional state detected of at least one user in order to improve the relevance of the list of at least one program that is recommended to him.

[0011] According to a particular aspect of the invention, the method comprises the steps performed by said audiovisual decoder of:

[0012] - association of said genres and / or subgenres with emotional states;

[0013] - detection of a current emotional state of said at least one user;

[0014] - carrying out a sorting of said programs of said program catalogue relating to less taking into account the said detected emotional state of the said genres and / or subgenres associated with the said detected emotional state;

[0015] - displaying a list of at least one audiovisual program taking into account said sorting of the said programs of the said program catalogue.

[0016] According to a particular embodiment of the invention, said step of associating said genres and / or subgenres with said emotional states includes a substep of using a priority table in which, for each of said emotional states, said genres and / or subgenres are ranked according to an order of priority, said ranking taking into account at least a counter of readings of each genre and / or subgenre associated with each emotional state, said priority table further comprising a default genre and / or subgenre assigned last according to the order of priority for each of said emotional states.

[0017] According to a particular aspect of this particular method, said step of sorting said programs from said catalogue, taking into account at least said detected emotional state and said genres and / or subgenres associated with said detected emotional state, comprises a substep of:

[0018] - scoring, in said priority table, of said genus and / or said subgenre classified in first according to said order of priority for said detected emotional state, and

[0019] - detection of a selection by the user and if the user does not select a program corresponding to the first genre and / or subgenre indicated, indicating the next genre and / or subgenre in the priority table until the user selects a program of said genre and / or subgenre indicated or until said genre and / or subgenre pointed to is said genre and / or subgenre by default.

[0020] According to a particular aspect of this particular mode, said step of displaying said list of at least one audiovisual program taking into account said sorting of said programs of said catalogue comprises the substeps of:

[0021] - scheduling of said at least one program from said list of at least one program to be displayed, said scheduling taking at least in part into account a priority criterion for said at least one program; and

[0022] - display of said list of at least one program according to said order funding.

[0023] According to another particular embodiment of the invention, said step of associating said genres and / or subgenres with said emotional states includes a substep of using a counting table comprising a counter of readings by the user for each genre and / or subgenre associated with an emotional state.

[0024] According to a particular aspect of this other particular mode, said step of performing a sorting (E60) of said programs of said catalogue, taking into account at least said detected emotional state and said genres and / or subgenres associated with said detected emotional state, comprises the substeps of:

[0025] - calculation of a weighting coefficient for each of said genres and / or a sub- gender associated with audit emotional state detected;

[0026] - for each program included in a list of programs offered in whereby each programme presents a predetermined score and a genre and / or subgenre, calculation of a weighted score by multiplying said predetermined score of said programme by said weighting coefficient corresponding to said genre and / or said subgenre;

[0027] - ranking of said programs from said list of programs proposed by descending order of weighted score.

[0028] According to a particular aspect of this other particular mode, the substep of calculating a weighting coefficient for each of said genres and / or subgenres associated with said detected emotional state comprises:

[0029] - for said detected emotional state, the determination of a maximum value among said counters associated with each of said genres;

[0030] - for each gender associated with said detected emotional state, the division of said value said corresponding counter by said maximum determined value.

[0031] According to another particular aspect of this other particular mode, said step of performing a sorting (E60) of said programs of said catalogue taking into account at least said detected emotional state and said genres and / or subgenres associated with said detected emotional state comprises a substep of:

[0032] - ranking of programs from a list of proposed programs, in which Each program presents a predetermined score, in descending order of said counter associated with said genre and / or subgenre and said detected emotional state; then

[0033] - ranking of said programs in said list of programs having the same genre and / or subgenre in descending order of predetermined score.

[0034] According to a particular aspect of the invention, the method includes a step of updating said priority table or said counting table, when said at least one user reads one of said recommended programs for at least a predetermined duration, by incrementing a value of said counter associated with said program read by said at least one user.

[0035] According to another particular aspect of the invention, each genre comprises at least one subgenre, and said priority table and said counting table comprise a reading counter for each genre and / or a reading counter for each subgenre. The step of updating said priority table and said counting table comprises the substeps of, when said at least one user reads one of said recommended programs:

[0036] - incrementing said counters of the genre and / or subgenre of the program se selected by said at least one user corresponding to said emotional state detected of a first value;

[0037] - incrementing the counters of the other subgenres associated with the genre of program selected by said at least one user corresponding to said emotional state detected of a second value.

[0038] According to one particular aspect, when said audiovisual decoder detects at least two users, the method comprises:

[0039] - a step of determining the primary user based on at least one predetermined criterion, said substep of using a priority table or a counting table being carried out from said priority table or counting table associated with said determined primary user; or

[0040] - a step of combining said priority tables or counting each of said detected users, said substep of using a priority table or a counting table being carried out from the priority table or the counting table resulting from this combination.

[0041] According to a particular aspect of the invention, the process includes a step for triggering said process:

[0042] - upon starting up said audiovisual decoder; or

[0043] - upon detection of at least one user by said audiovisual decoder; or

[0044] - to the selection / search of a program by said at least one user.

[0045] The invention also provides an audiovisual decoder for implementing the A method for recommending a program or a list of audiovisual programs from a program catalog to at least one user as described above, said programs from said catalog being categorized into genres and / or subgenres. Said audiovisual decoder being intended to be connected to a remote database, a display means, and a control device usable by a user to operate said audiovisual decoder, said decoder comprising electronic circuitry adapted and configured for:

[0046] - to associate said genres and / or said subgenres with emotional states;

[0047] - detect an emotional state of said at least one user;

[0048] - to carry out a sorting of said programs of said catalogue taking at least into account said detected emotional state and of the genres and / or subgenres associated with said detected emotional state

[0049] - said display means being intended to display a list of at least one audiovisual program taking into account said sorting of said programs of said catalogue.

[0050] Also proposed herein is a computer program that can be stored on a medium and / or downloaded from a communication network for reading by a processor. This computer program includes instructions for implementing the method mentioned above in any of its embodiments when said program is executed by the processor. The invention also relates to an information storage medium for storing such a computer program. Brief description of the drawings

[0051] The features of the invention mentioned above, as well as others, will become clearer upon reading the following description of an exemplary embodiment, said description being made in relation to the accompanying drawings, among which:

[0052] [Fig-1] schematically illustrates a telecommunications system in which the the present invention is implemented;

[0053] [Fig.2] schematically illustrates an example of the hardware architecture of an audiovisual decoder of the invention;

[0054] [Fig.3] schematically illustrates a recommendation method according to the invention;

[0055] [Fig.4] schematically illustrates a step of associating genres (and possibly subgenres) with emotional states of the process of [Fig.3], according to a first mode of realization;

[0056] [Fig.5] represents an example of a priority table implemented in the process of [Fig.3], the priority table being in a first state;

[0057] [Fig. 6] schematically illustrates a step in sorting programs from a catalog, taking into account at least one detected emotional state and the genres and / or subgenres associated with the detected emotional state of the process in [Fig. 3], according to the first embodiment;

[0058] [Fig.7] represents an example of a priority table during implementation in the process of [Fig.3], the priority table being in a second state;

[0059] [Fig.8] represents an example of a priority table during implementation in the process of [Fig.3], the priority table being in an initial state;

[0060] [Fig.9] schematically illustrates a step of displaying a list of at least one audiovisual program taking into account the sorting of programs from the catalogue of the process of [Fig.3];

[0061] [Fig. 10] schematically illustrates a method of filling the priority table by learning, according to the first embodiment;

[0062] [Fig. 11] schematically illustrates a step of associating genres (and possibly subgenres) with emotional states of the process of [Fig.3], according to a second embodiment;

[0063] [Fig. 12] represents an example of a counting table implemented in the process of [Fig.3], the counting table being in an initial state;

[0064] [Fig. 13] schematically illustrates a step in carrying out a sorting of programs from a catalogue taking into account at least one detected emotional state and the genres and / or sub-genres associated with the detected emotional state of the process of [Fig.3], according to the second embodiment;

[0065] [Fig. 14] represents an example of a list of programs proposed by the operator, each program including a predetermined score;

[0066] [Fig. 15] represents an example of a counting table implemented in the process of [Fig.3], in which the counters have been incremented following the implementation of the process of [Fig.3];

[0067] [Fig. 16] illustrates an example of a table for calculating a weighted score;

[0068] [Fig. 17] illustrates an example of a weighting table for a list of programs offered by the operator;

[0069] [Fig. 18] illustrates an example of a list of programs hierarchically ordered by the implementation of the process of [Fig.3], according to the second embodiment;

[0070] [Fig. 19] schematically illustrates a step of updating the counting table, according to the second embodiment;

[0071] [Fig.20] schematically illustrates a variant of a step in carrying out a sorting of programs from a catalogue taking into account at least one detected emotional state and the genres and / or sub-genres associated with the detected emotional state of the process of [Fig.3], according to the second embodiment;

[0072] [Fig.21] schematically illustrates a variant of a step for updating the priority table or the counting table, compatible with the first and second embodiments; and

[0073] [Fig.22] schematically illustrates a step in detecting the number of users of the process of [Fig.3].

[0074] DETAILED DESCRIPTION OF IMPROVEMENTS

[0075] Fig. 1 schematically illustrates a telecommunication system 1 in which the present invention is implemented.

[0076] In [Fig. 1], an external data source 14 comprising a server 141 and a remote database 142 is connected via a wide area network 20 to local area networks 40. The wide area network 20 is, for example, an internet-type network. The remote database 142 is, for example, configured to store information relating to audiovisual programs.

[0077] For example, local area networks 40 are wired and / or wireless home networks. Only four home networks 40a - 40d are shown in [Fig. 1] for the sake of simplicity. In one particular example, local area networks 40a - 40d are Wi-Fi type wireless networks.

[0078] In the example of [Fig. 1], each local area network 40a - 40d includes an audiovisual decoder 10 enabling the implementation of a method for recommending at least one program list. The decoder 10 is connected to at least one display device 13 (via an HDMI (High Definition Multimedia Interface), SCART (Syndicat des Constructeurs d'Appareils Radioreceveurs et Télévisions), or any other suitable type of link) and to a control device 121 (preferably via a wireless connection) used by at least one user 12 to control the audiovisual decoder 10. According to one approach, the user can interact with the audiovisual decoder 10 via a microphone integrated into the audiovisual decoder 10. In the illustrated example, only one user and one display device are shown.

[0079] Figure 2 schematically illustrates an example of a hardware architecture enabling the implementation, in the form of electronic circuitry, of the audiovisual decoder according to the invention.

[0080] According to the hardware architecture example shown in [Fig. 2], the decoder 10 comprises, connected by a communication bus 200: a processor or CPU (Central Processing Unit) 201; a RAM (Random Access Memory) 202; a ROM (Read Only Memory) 203, for example of the ROM (Read Only Memory) or EEPROM (Electrically-Erasable Programmable ROM) type, such as Flash memory; a storage unit 204, such as a HDD (Hard Disk Drive) or a storage media reader, such as an SD (Secure Digital) card reader; at least one I / F interface manager 205. The Ff interface manager 205 enables the audiovisual decoder 10 to communicate via the wide area network 20 and to interact with the control device 121.

[0081] The processor 201 is capable of executing instructions loaded into RAM 202 from ROM 203, external memory (not shown), storage media (such as an SD card), or a communication network. When the audio-visual decoder 10 is powered on, the processor 201 is capable of reading instructions from RAM 202 and executing them. These instructions form a computer program causing the processor 201 to implement all or part of the process described herein.

[0082] The storage unit 204, or the RAM 202, is suitable for storing a local database 207 storing the priority and / or counting tables described in more detail later in this description.

[0083] The decoder 10 further includes means 206 for detecting a user's emotion. To this end, the detection means 206 include means for capturing video and / or audio (for example, at least one camera and / or microphone) of the user, as well as an association module configured to associate the captured images and / or audio with a current emotional state of the user. Optionally, the detection means 206 may implement a connected device of the IoT (Internet of Things) type based on biomarkers.

[0084] All or part of the method described herein can thus be implemented in software form by the execution of a set of instructions by a programmable machine, for example a DSP (Digital Signal Processor) or a microcontroller, or be implemented in hardware form by a dedicated machine or electronic component (chip) or a dedicated set of electronic components (chipset), for example an FPGA (Field Programmable Gate Array) or ASIC (Application-Specific Integrated Circuit). Generally speaking, the audiovisual decoder 10 includes electronic circuitry adapted and configured to implement the method described herein.

[0085] The method described below with reference to Figs. 3 to 22 can be implemented in software form by executing a set of instructions by a programmable machine, for example a DSP (Digital Signal Processor) or a microcontroller, or can be implemented in hardware form by a dedicated machine or component, for example a FPGA (Field-Programmable Gate Array) or an ASIC (Application-Specific Integrated Circuit). In general, the audiovisual decoder 10 includes electronic circuitry configured to implement the method described with reference to Figs. 3 to 22.

[0086] It should be noted here that [Fig. 2] represents a hardware architecture of a single audiovisual decoder 10. Some of the various constituent elements of the audiovisual decoder 10 can be distributed among different computer devices included in a computer cloud.

[0087] Figure 3 schematically illustrates a method 100 for recommending a list of at least one audiovisual program from a catalog of audiovisual programs according to the present invention. The recommendation method 100 is intended for at least one user 12 directed to the audiovisual decoder 10 and to the display means 13. It should be noted that the programs in the program catalog are categorized by genre and, according to another approach, by genre and subgenre. Information relating to the category of each program can be retrieved by the audiovisual decoder 10 using known methods. For example, the broadcasting of a program on satellite, cable, or terrestrial television networks uses the DVB (Digital Video Broadcasting) standard, which provides for the reporting of metadata related to a program via the EIT (Event Information Table).An EIT table can notably indicate the genre of the program via the content description, as detailed in document ETSI EN 300 468. The audiovisual decoder 10 of the invention is therefore capable of obtaining information relating to the genre of programs by a suitable means (such as IP (Internet Protocol), DSL (Digital Subscriber Line), cable, satellite, ...), from the external data source 14, which can be a Satellite, Cable, DTT (Digital Terrestrial Television), IP or any other audiovisual stream source input.

[0088] According to another approach, program information can be retrieved by the audiovisual decoder 10 via a description text file (for example, JSON (JavaScript Object Notation)) downloaded from an Internet server. Generally, the possible program types used according to the invention are those provided by the content provider (DVB, OTT (Over-The-Top), etc.).

[0089] For example, programmes can be categorized according to the following genres: Film / Drama, News, Show / Game Show, Sports, Children's / Youth Programme, Music / Ballet / Dance, Arts / Culture (without music), Social / Political / Economic Issues, Education / Science / Factical Topics, Leisure, etc.

[0090] Optionally, programmes can also be categorized into sub-genres, such as for the genre Film / drama: General, Detective / thriller, Adventure / western / war, Science fiction / fantasy / horror, Comedy, Serial / melodrama / folkloric, Romance, Serious / classical / religious / historical film / drama, Adult film / drama.

[0091] It is understood that the invention may use all or part of the genera listed above. The invention may also use other genera not listed above. Furthermore, the invention may use the main genera alone or in combination with their sub-genera.

[0092] Furthermore, the audiovisual decoder 10 has a database 207 containing a number of emotional states that have already been detected or that it is capable of detecting (as described below) by the detection means 206. These emotional states include, for example: anger, disgust, fear, joy, sadness, surprise, neutrality, etc. It is clear that the invention can use all or part of the emotional states listed above. The invention can also use other emotional states not listed above.

[0093] According to the invention, the recommendation method 100 comprises the steps, implemented by the audiovisual decoder 10, of: - association E20 of genres and / or subgenres with emotional states; - E40 detection of a current emotional state of the user; - implementation of an E60 sorting of said programs of said catalogue containing at least taking into account the said detected emotional state and the said genres and / or subgenres associated with the said detected emotional state; - display E80 of a list of at least one audiovisual program taking into account said sorting of said programs of said catalogue.

[0094] To facilitate understanding of the method according to the first embodiment, the latter is described considering only the genres of programs (and not subgenres). It is readily understood that the method of the invention could take into account, in addition to or instead of genres, the subgenres of programs without departing from the operating principle of this embodiment. An example of an embodiment taking subgenres into account is described later in the description.

[0095] According to a first embodiment, the step E20 of associating genres (and possibly subgenres) with emotional states includes, as illustrated in [Fig. 4], a substep E21 of using a priority table. More specifically, the priority table comprises a list of common emotional states to which program genre suggestions are associated. The priority table allows each genre suggested for each emotional state to be ranked E211 according to a priority order. This ranking E211 of genres associated with each emotional state takes into account at least a counter of the number of times the user reads each genre associated with each emotional state. In other words, for each emotional state, genres are ranked in order of priority; that is, for a given emotional state, genres are ranked in order of the genres most frequently read by the user when the user is in that emotional state.Reading counters locally count the number of times a user reads something, for example, within the home.

[0096] For example, and as illustrated in [Fig. 5], which represents a priority table according to the first embodiment, for the emotional state corresponding to the When experiencing sadness, the most watched genres are comedy, followed by entertainment, and then psychology. This means that when a user is in a sad emotional state, they are more likely to have watched comedy programs, followed by entertainment programs, and then psychology programs.

[0097] As mentioned above, the ranking of genres by emotional state takes into account, in particular, a counter of readings associated with each genre for each emotional state. However, other criteria can be considered when ranking the genres associated with each emotional state. Furthermore, it is also possible to predefine the order of preference for the genres to be offered for each emotional state. This predefined order could, for example, be configured by the user when setting up the audiovisual decoder 10 or by the operator providing the audiovisual decoder 10.

[0098] It should be noted that in the priority table shown in [Fig. 5], only three genders are detailed, but n genders can be associated with each emotional state. It is clear that a number higher or lower than three genders could be associated with each emotional state. The number of genders associated with each emotional state could also be limited, for example to 20 genders per emotional state, in order to limit the size of the priority table and optimize storage space, among other things.

[0099] Optionally, the number of genres taken into account in the priority table (i.e., the number of priorities) could correspond to the total number of genres proposed by the audiovisual content broadcasting standard. Optionally, the number of genres taken into account in the priority table could be extended, for example, by the broadcaster or by the audiovisual decoder 10 when the addition of a new genre is detected. This addition could, for example, be made at the end of the table by adding a new column to the priority table.

[0100] In addition, for each emotional state, a default genre (and possibly a default sub-genre) is provided which is offered to the user when the priority table is initialized or when the user does not select any of the genres offered, as described below.

[0101] The priority table can be embedded in the audiovisual decoder 10 or downloaded from an external network. For example, a possible implementation of the priority table could be in SQL (Structured Query Language) or it could be a JSON file.

[0102] Step E40 of detecting a current emotional state of user 12 may, for example, implement a known detection technique which consists of using trained neural networks to detect an individual's emotions from of images. This technique can, for example, implement algorithms such as Gabor filters, histograms of oriented gradients (HOGs), or local binary patterns (LBPs) to extract facial features from an image. For classification, this technique can, for example, implement a support vector machine (SVM), a random forest (RF), or a nearest neighbors algorithm (kNN).

[0103] Other techniques make it possible to detect an individual's emotional state from their voice, as described in documents US11341986 or TW201606760A.

[0104] According to the first embodiment, the step of performing an E60 sort of the catalog programs, taking into account at least the detected emotional state and the genres (and / or subgenres when the process considers subgenres) associated with the detected emotional state, makes it possible to extract a program or a list of programs to suggest / recommend to the user based on their current detected emotional state. To do this, the audiovisual decoder 10 refers to the priority table in [Fig. 5]. The audiovisual decoder 10 implements the step of performing an E60 sort, which includes the substeps (illustrated in [Fig. 6]): - point E61, in the aforementioned priority table, of the type ranked first according to the priority order for the detected emotional state, and - detection E62 of a selection by the user and, if the user does not select a program corresponding to said first pointed kind, pointing E63 to the next kind in the priority table until the user selects a program of said pointed kind or until said pointed kind is said default kind E64.

[0105] More specifically, and referring to the priority table illustrated in [Fig. 5], when an emotional state is detected, the pointer is placed on the first genre according to the priority order of the table (as illustrated in [Fig. 5]). Continuing with the example of the emotional state corresponding to sadness, the first genre according to the priority order is the comic genre. Therefore, it is the comic genre that is pointed to (see the arrow in [Fig. 5]) at the time of substep E61.

[0106] A substep E611 for extracting a list of at least one program corresponding to the selected genre from among the programs in the catalog is therefore implemented by the audiovisual decoder 10. Subsequently, a substep E67 for displaying the extracted list containing at least one audiovisual program corresponding to the selected genre, i.e., the genre indicated in the table, is implemented. The display substep E67 corresponds to the display step E80 as described below.

[0107] At substep E62, the audiovisual decoder 10 detects whether the user selects a program among the programs in the extracted list corresponding to the genre pointed to in the priority table.

[0108] At substep E68, the audiovisual decoder 10 checks if the user selects a program from this extracted list of programs corresponding to the pointed-out genre.

[0109] If the user selects a program from this extracted list of programs corresponding to the genre pointed to in the priority table, the decoder proceeds to substep E65.

[0110] If the user does not select a program from this extracted list of programs corresponding to the genre pointed to in the priority table, and if there are still genres to be offered in the priority table, the audiovisual decoder 10 checks, in substep E69, whether there are still genres to be offered in the priority table. If so, the audiovisual decoder 10 proceeds to substep E63 in which the pointer is shifted one column in the priority table, and a substep E611 is implemented to extract a new list of at least one program corresponding to the pointed-to genre, followed by a substep E67 to display a list of at least one audiovisual program corresponding to the selected genre, i.e., the genre pointed to in the table.If there are no genres left to offer in the priority table, the audiovisual decoder 10 points to E64 as the default genre and a sub-step E611' of extracting a new list of at least one program corresponding to the default genre, then a sub-step E67' of displaying a list of at least one audiovisual program corresponding to the default genre are implemented.

[0111] In the present example relating to the emotional state "sad", if the user does not select a program corresponding to the genre "comedy", then the pointer is shifted to the genre "entertainment", as illustrated in [Fig. 7]. A list of at least one audiovisual program corresponding to the entertainment genre is then extracted and displayed. If the user still does not select a program from the extracted list of programs corresponding to the entertainment genre, then the audiovisual decoder 10 proceeds to substep E69 and if there are still genres to be offered in the priority table, the audiovisual decoder 10 proceeds to substep E63 so that the pointer is shifted to the next genre in the priority table (psychology genre), and so on as long as there are still genres to be offered in the priority table. Thus, the audiovisual decoder 10 checks at substep E69 that there are still genres to be offered in the priority table.If yes, the decoder proceeds to step E63, while if no, the audiovisual decoder 10 proceeds to step E64.

[0112] In this way, the pointer is shifted until it reaches the default genre, which is ranked last in the priority table for each emotional state (here the default genre associated with the emotional state sadness is the comedy genre). The default genre is defined beforehand. For example, it can be defined

[0113]

[0114]

[0115]

[0116]

[0117]

[0118] automatically by the audiovisual decoder 10 or manually by the user during the configuration of the audiovisual decoder 10. In other words, the pointer traverses the priority table to successively suggest genres ranked in the priority table that correspond to the detected emotional state. If no program matching these genres is selected by the user, a default genre is chosen, and a list of at least one program associated with that default genre is extracted and displayed. If the user selects a program from the extracted list of programs corresponding to a genre indicated in the priority table, then the audiovisual decoder 10 updates the priority table (UPD substep E65) by incrementing the value of the play counter for the genre corresponding to the selected program for the detected emotional state, and the different genres belonging to the detected emotional state are reordered so that: - Priority 1 contains the genre with the most readers, - priority 2 contains the second genre with the most readings following the genre with priority 1; etc. Step E80, which displays a list of at least one program taking into account said sorting of said programs in said catalog, includes, in this example, the sub-steps of (illustrated in [Fig.9]): - scheduling E81 of said list of at least one program to be displayed, the scheduling taking into account at least in part a priority criterion of at least one program; and - display E82 of the list of at least one program according to the scheduling. More specifically, when the step of performing an E60 sort of the catalog programs, taking into account at least the detected emotional state and the genres and / or subgenres associated with that emotional state, is completed, a list of at least one program matching the selected genre is extracted. It is this extracted list that must be displayed to the user (E80). The ordering of the programs in the extracted list takes into account at least one priority criterion associated with each program. For example, the priority criterion might consider the program's popularity with other users, its release or availability date, or its commercial / marketing contribution (e.g., advertisements shown in certain programs sell for more than in others, a program's producer might pay for increased visibility, etc.).A combination of this data can be used to determine the priority criterion for each program. Other data can obviously be used in determining this priority criterion. Preferably, the priority criterion is received beforehand by the audiovisual decoder 10. for example, when receiving information relating to audiovisual programs

[0119] In this example, the programs in this extracted list are ordered according to their popularity with other users. Here, the programs are therefore ordered from the most popular to the least popular among other users. For example, a program's popularity can be determined by considering the number of times it has been viewed, its rating by other users, its rating by platforms, its rating by media outlets, or any other criterion. A combination of these criteria can also be used to determine the program's popularity.

[0120] When the programs in the list are ordered, the audiovisual decoder 10 displays, via the display means 13, the ordered list to the user to facilitate selection. An example of an ordered program list might consist of a presentation of the Top 10 most-watched comedy films (for the emotional state "sad" with priority 1) or the Top 10 highest-rated entertainment programs (for the emotional state "sad" with priority 2), etc.

[0121] Fig. 8 represents an empty priority table, as it may be when the audiovisual decoder 10 is initialized. When the audiovisual decoder 10 is initialized, the default column proposing a default genre for each emotional state may, however, be automatically populated.

[0122] The priority table can be populated at initialization based on a user panel and by copying the content of the priority table to that of the user panel.

[0123] Another solution may be to fill the priority table by learning, that is to say as the user selects a program.

[0124] Figure 10 illustrates an example of a method for filling the priority table by learning. This filling method 90 includes the steps of: - measurement by the audiovisual decoder 10 at step E91 of the viewing time of a program by the user; - verification by the audiovisual decoder 10 at step E97 whether the measured viewing time is greater than a predetermined minimum duration (for example, ten minutes). If not, the audiovisual decoder 10 interrupts this algorithm. If so, the audiovisual decoder 10 proceeds to step E92 of detecting the user's current emotional state; - At step E98, the audiovisual decoder 10 checks if the detected emotional state includes the genre of the program viewed by the user in the priority table. If so, the audiovisual decoder 10 proceeds to step E93. If not, the audiovisual decoder 10 proceeds to step E95. At the step E93, the audiovisual decoder 10 adds the genre of the program viewed by the user for the emotional state detected in the priority table and initializes at step E94 the reading counter corresponding to 1; - if the detected emotional state already contains the genre of the program viewed by the user in the priority table, the audiovisual decoder 10 increments E95 by one unit of the value of the reading counter of the genre corresponding to the detected emotional state; - Audiovisual decoder 10 updates at step E96 the ranking of genres in order of priority for the detected emotional state.

[0125] This method can be implemented in parallel with the recommendation 100 process described above.

[0126] According to a second embodiment, the recommendation process 100 involves the user's current emotional state being detected to weight a predetermined score associated with a program so as to make a program recommendation to the user that takes into account their current emotional state.

[0127] To facilitate understanding of the method according to the second embodiment, the latter is described considering only the genres of programs (and not subgenres). It is readily understood that the method of the invention could take into account, in addition to or instead of genres, the subgenres of programs without departing from the operating principle of this embodiment. An example of an embodiment taking subgenres into account is described later in the description.

[0128] To this end, in the association step E20 of genres (and possibly subgenres) with emotional states, the audiovisual decoder 10 implements, as illustrated in [Fig. 11], a substep E31 of a counting table comprising a counter of user readings for each genre associated with an emotional state. More specifically, the counting table includes a list of common emotional states to which program genre suggestions are associated. The counting table allows the audiovisual decoder 10 to count E311, by incrementing by one value a reading counter associated with each genre for each emotional state, the number of readings of a program of each genre associated with each emotional state when the user selects a program suggested by the method of the invention.In other words, each time a user selects and plays a program offered by the invention's method, the audiovisual decoder 10 increments the counter corresponding to the genre associated with the detected emotional state by one value. In this way, the counting table makes it possible to count and store the number of readings of each genre associated with each emotional state.

[0129] It is also possible to count readings that are performed outside the process of the invention. For example, it is possible to count readings performed by a spontaneous selection by the user or when the user switches from one program to another, in particular. In order to optimize the relevance of the recommendations, the counting table can also be populated by learning, in a manner substantially similar to the technique described above in relation to [Fig. 10].

[0130] Figure 12 shows a counting table during the initialization of the audiovisual decoder 10. Each counter is initialized with a value of 1. It is understood that the counters could be initialized with any other value, preferably non-zero. The genres and emotional states entered in the counting table can be fixed in a predetermined manner, or retrieved from an external server, for example.

[0131] As before, the association step E20 of genres (and possibly subgenres) with emotional states is followed by a step E40 of detecting a current emotional state of the user 12. This step E40 of detecting a current emotional state of the user 12 can for example implement the known detection technique described in relation to the first embodiment and which consists of detecting the emotions of an individual from images or from the voice of the user 12.

[0132] The step of performing an E60 sort of the catalogue programs, taking into account at least the detected emotional state and the genres and / or subgenres associated with the detected emotional state, comprises the substeps (illustrated in [Fig. 13]) of:

[0133] - calculation E71, by the audiovisual decoder 10, of a weighting coefficient for for each of the genres associated with the detected emotional state, the calculation of the weighting coefficient including:

[0134] o for the detected emotional state, the determination E710, by the audiovisual decoder 10, of a maximum value among the counters associated with each of the genres;

[0135] o for each genre associated with the detected emotional state, the division E712, by the audiovisual decoder 10, of the value of the corresponding counter by the maximum value determined;

[0136] - for each program included in a list of programs offered in where each programme presents a predetermined score and a genre, calculation E73, by the audiovisual decoder 10, of a score weighted by the multiplication of the predetermined score of the programme by the weighting coefficient corresponding to said genre;

[0137] - E75 prioritization, by the audiovisual decoder 10, of the programs in the list of programs offered in descending order of weighted score.

[0138] In this second embodiment, the audiovisual decoder 10 receives a list of programs in which a predetermined score is associated with each program in the list (in addition to the bibliographic information and the indication of genre (and possibly subgenre) typically received). For example, the predetermined score includes a rating between 0 and 5 corresponding to the program's popularity among other users. The predetermined score can be determined by taking into account the number of times the program has been viewed (total or over a given period), the program's age, its average rating given by other users, its rating by platforms, its rating by the media, or any other criteria. A combination of these criteria can also be used to determine the program's popularity. The predetermined score is independent of the user and is set by rules external to the method of the invention, for example.

[0139] Figure 14 illustrates an example of a program list in which a predetermined score is associated with each program. To facilitate understanding of the step of performing an E60 sort of said programs in the catalog, taking into account at least the detected emotional state and the genres and / or subgenres associated with the detected emotional state, Figure 15 illustrates an example of a tally table after several days of use. Note that the counters for certain genres associated with certain emotional states have been incremented.

[0140] In order to calculate the weighting coefficient p(G) in step E71, the audiovisual decoder 10 first determines in step E711 the maximum value among the counters associated with each genre. In other words, the audiovisual decoder 10 will scan the counting table and find which counter, for the detected current emotional state, is the highest. If we take as an example the emotional state corresponding to anger in the counting table of [Fig. 15], the highest counter value corresponds to the genre "Music". The maximum value retained is therefore that of the counter associated with the genre "Music" for the emotional state of anger, namely a counter value of 50.

[0141] In one variant, the weighting coefficient could be calculated, or determined, differently, for example by dividing the value of the corresponding counter by the sum of the values ​​of the counters associated with the detected emotional state, or by directly taking the value of the corresponding counter as the weighting coefficient.

[0142] Next, for each genre associated with the current detected emotional state, the audiovisual decoder 10 will divide E712 the value of each counter by the maximum value determined previously.

[0143] In this example, and as illustrated in [Fig. 16], the values ​​of the counters for each genre associated with the emotional state of anger are divided by the determined maximum value, namely divided by 50. Thus, the genre film (“Movie” in English) for the emotional state of anger has a weighting coefficient P(G) of 4 / 50, corresponding to the value of the film genre counter for the emotional state of anger (i.e. a value of 4) divided by the maximum value determined (i.e. 50).

[0144] The weighting coefficient P(G) is in this example calculated for each gender corresponding to the detected emotional state.

[0145] When the weighting coefficients P(G) of each genre corresponding to the detected emotional state are calculated, the audiovisual decoder 10 calculates E73 the predetermined weighted score of each program in the list of proposed programs retrieved by the audiovisual decoder 10.

[0146] To do this, the audiovisual decoder 10 multiplies the predetermined score of each program by the weighting coefficient corresponding to the genre of the program.

[0147] Referring to the example described here, the table illustrated in [Fig. 17] shows the weighted score of each program in the program list of [Fig. 14]. Program 1 has a weighted score of 0.24, corresponding to its predetermined score (namely 3) multiplied by the weighting coefficient for the film genre (namely 4 / 50).

[0148] The weighted score calculation is carried out for each program in the program list offered to the audiovisual decoder 10. The weighted score therefore makes it possible to weight the predetermined score of the program by taking into account the current emotional state of the user.

[0149] Next, the programs in the program list offered to the audiovisual decoder 10 are ranked E75 in descending order of weighted score so as to offer the user a program list which takes into account his current emotional state.

[0150] Figure 18 illustrates the list of programs ranked in descending order of weighted score. It should be noted that, again for the emotional state of anger, the audiovisual decoder 10 will offer the user a music program, then a culture program, then a film program, and so on.

[0151] In this embodiment, the audiovisual decoder 10 sorts the proposed programs based on a weighting coefficient that takes into account the detected emotional state of the user. The list of programs displayed to the user can therefore include programs of different genres and / or subgenres.

[0152] During step E80 of displaying a list of at least one audiovisual program taking into account the sorting of said programs from the catalogue, the audiovisual decoder 10 will display the hierarchical program list described above.

[0153] The process 100 further comprises, in this second embodiment, a step E74 executed by the decoder 10 for updating the counting table. To do this, and as illustrated in [Fig. 19], following step E80 of displaying the list of at least one audiovisual program taking into account the sorting of programs in the catalog, the Audiovisual decoder 10 will detect a user selection at step E741.

[0154] At step E745 the audiovisual decoder 10 checks if the user has selected one of the programs from the hierarchical program list.

[0155] If no, the audiovisual decoder 10 interrupts the present algorithm. If yes, the audiovisual decoder 10 updates the counting table by incrementing, at step E742, the counter corresponding to the genre of the program selected for the detected current emotional state.

[0156] In a variant of this second embodiment, the audiovisual decoder 10 does not calculate a weighted score at the E60 selection step of at least one genre based on at least the detected emotional state. According to this variant, and as illustrated in [Fig. 20], the E60 sorting step of programs in the catalog, taking into account at least the detected emotional state and the genres and / or subgenres associated with the detected emotional state, comprises the following sub-steps: - E611 ranking, by the audiovisual decoder 10, of the programs in the list of proposed programs, in which each program has a predetermined score, in descending order of the counter associated with the genre and the detected emotional state; then - E612 ranking, by the audiovisual decoder 10, of the programs in the list of programs having the same genre in descending order of the predetermined score.

[0157] In another embodiment, compatible with both of the invention described above and their variants, each genre comprises at least one subgenre. The priority and counting tables therefore include a counter for each subgenre, in addition to or instead of the counters associated with each genre. Thus, and as illustrated in [Fig. 21], the priority table update step E65 and the counting table update step E74 each include the substeps of, when at least one user views one of the recommended programs: - incrementing E651, E743 of the counters of the genre and / or sub-genre of the program selected by at least one user corresponding to the detected emotional state of a first value; - incrementing E652, E744 of the counters of the other subgenres associated with the genre of the program selected by at least one user corresponding to the detected emotional state of a second value.

[0158] In this way, it is possible to take into account the genre and / or sub-genre of a program in order to make an optimal recommendation to the user based on their current detected emotional state.

[0159] In an unillustrated example, the audiovisual decoder 10 only increments the sub-genre counters. Thus, the audiovisual decoder 10 increments by a first value equal to two the value of the counter corresponding to the pair consisting of the main genre and the sub-genre of the program selected by the user, while the audiovisual decoder 10 increments by a second value equal to one the value of the counters corresponding to the main genre but not to the sub-genre of the program chosen.

[0160] In this example, if we consider that the user selects a comedy film offered by the process, the comedy film is a program belonging to the main genre "film" and the subgenre "film / comedy". Then, the counter of the subgenre "film / comedy" is incremented by a first value equal to 2, while the counters of the other subgenres of the genre "film" (such as the subgenres "film / drama", "film / science fiction", "film / adventure", etc.) are only incremented by a second value equal to 1.

[0161] In the example described above, only the subgenre counters are considered. It is easy to understand that the number of readings of the genre can be obtained by incrementing the corresponding counter or by determining its value from the sum of the counters of the subgenres associated with it, for example.

[0162] It should be noted that several genres may have similar subgenres; for example, there may be a genre of "theater" with subgenres similar to those of the genre of "film" in the preceding example, such as "theater / comedy," "theater / drama," etc. In a preferred embodiment of the invention, subgenres of distinct genres are themselves considered distinct. Thus, for example, the subgenre of "film / comedy" and the subgenre of "theater / comedy" each have a distinct counter. According to an alternative embodiment, similar subgenres use the same counter, so that there is a single common counter for the subgenres of "film / comedy," "theater / comedy," "series / comedy," etc.

[0163] Thus, the method according to the invention makes it possible to improve the targeting of programs recommended to the user.

[0164] It should be noted that the recommendation process 100 can be triggered in different ways. To this end, process 100 may include a step E101 for launching / triggering said process 110: - when the audiovisual decoder 10 is switched on; or - upon detection by the decoder of the presence of at least one user by the audiovisual decoder 10; or - upon detection by the decoder of a selection / search for a program by at least one user.

[0165] It should be noted that, in the case where there are several users in the same household, each user can have their own priority and / or counting table. In In this case, the recommendation method 100 may include a step of user identification by the audiovisual decoder 10. For example, user identification can be performed manually via the control device 121 or automatically via the emotional state detection means 206. Indeed, the detection means 206 can be used not only for detecting the user's current emotional state but also to perform user identification, for example, through visual recognition or audio recognition of the user's voice. Prior art techniques exist for performing such identification and are therefore not described in detail here.

[0166] It should be noted that, optionally, the recommendation method 100 may include a detection step E50, by the audiovisual decoder 10, of the number of users directed to the detection means 206. In this case, the recommendation method 100 may either take into account only one of the users, or take into account two or more users to make the recommendation.

[0167] To this end, and as illustrated in [Fig. 22], in the event of E50 detection of at least two users, the recommendation method 100 further includes: - a substep of determination E51, by the audiovisual decoder 10, of the main user on the basis of at least one predetermined criterion, the substep of use E21 of a priority table or the substep of use E31 of a counting table being carried out from said priority table or the said counting table associated with said main user determined; or - a substep of combination E52, by the audiovisual decoder 10, of the priority or counting tables of each of the detected users, the substep of use E21 of a priority table or E31 of a counting table being carried out from the priority table or the counting table resulting from this combination.

[0168] Thus, according to the first case, the audiovisual decoder 10 determines, at step E51, a primary user from among the detected users so as to only take into account the priority or counting table of this primary user. This substep E51 for determining the primary user may, for example, be based on one or more of the following criteria: the seniority of the priority or counting table, the highest number of program viewing hours, a manual declaration by the users, etc.

[0169] In this first case, only the priority or counting table of the user identified as the primary user is taken into account for sorting the programs in the catalog. It is also possible to consider only the emotional state of the user identified as the primary user when making the recommendation.

[0170] According to the second case, the audiovisual decoder 10 combines E51 the priority or counting tables of each of the detected users. To do this, the audiovisual decoder 10 can, for example, add up the counters of the different users relating to each genre associated with each emotional state to obtain only one priority or counting table, called the combined table. Thus, this combined priority or counting table will only have one counter per given genre associated with a given emotional state, the value of this counter representing the sum of the counters of each user for the given genre associated with the given emotional state.

[0171] According to another approach (compatible with the first embodiment only), the audiovisual decoder 10 can, for example, combine the priority tables of the different users by taking into account the counters of each user in the same priority table (and reordering the counters according to the order of priority). Thus, the preferences of each user can be taken into account, without discriminating between secondary users and a primary user.

[0172] In both cases, it is possible to update the priority or counting tables of each user, either individually (for example when the table used is that of the main user) or collectively (that is to say that the priority or counting table of each user can be updated according to the selected program).

[0173] It is also possible, when it is desired to take into account the emotional state of each user, to mix the priority tables of the different users and to present one or more lists of programs which take into account the emotional state of the different users.

[0174] According to another approach (compatible with the second embodiment only), the audiovisual decoder 10 detects the current emotional state of each user present and calculates a weighting coefficient for each user based on the detected current emotional state and each user's rating table. The audiovisual decoder 10 then calculates a weighted score for each program by multiplying the predetermined score by the weighting coefficients of all users present corresponding to the genre (and / or subgenre, if applicable) of the program. Thus, the recommendation also takes into account the emotional state of each user.

[0175] It should be noted that, during the initialization of the audiovisual decoder 10, the creation of priority and / or counting tables by the audiovisual decoder 10 may take into account user preferences regarding program recommendations. For example, it is possible to present the user with a questionnaire so as to take their preferences into account until a predetermined number of playbacks is reached. so that the recommendation process proposes particularly relevant programs based on the counters of each type associated with each emotional state.

[0176] For example, the questionnaire provided by the audiovisual decoder 10 may include questions such as: - "What kinds of films do you prefer to watch when you are happy?", so that the user is prompted to select their first choice, then their second, and so on. - "Would you like to watch an entertainment program when you are "Sad?", so that the user is prompted to answer yes or no.

[0177] According to another approach, during the initialization of the audiovisual decoder 10, the creation of priority and / or counting tables by the audiovisual decoder 10 can take into account information from an external source, such as information provided by the operator via the network 20.

[0178] Thus, instead of starting with an empty priority table or a counting table in which all counters are initialized with the value 1, the priority and counting tables can be pre-populated, for example, from an average of all the number of readings from a subscriber base. Then, the learning process (i.e., updating the tables as the user makes selections) can continue as described above in order to adapt the tables to each user.

[0179] The priority and / or counting tables at the initialization of the audiovisual decoder 10 can be transmitted via the TCP / IP network or via a private table or descriptor extending the DVB standard in an RF transmission.

[0180] Optionally, the priority and / or counting tables at the initialization of the audiovisual decoder 10 can be linked to a general profile of the kind, for example; "Film profile", "Sport profile" which would be chosen by the user when initializing the decoder.

[0181] Optionally, the priority and / or counting tables at the initialization of the audiovisual decoder 10 may be linked to a user profile linked to their age and / or gender, or any other criterion.

Claims

Claims

1. Method (100) for recommending a list of at least one audiovisual program from a program catalog to at least one user (12) directed towards an audiovisual decoder (10) and display means (13), said programs of said catalog being categorized into genres and / or subgenres, characterized in that the method (100) comprises the steps executed by said audiovisual decoder (10) of: - associating (E20) said genres and / or said subgenres with emotional states; - detecting (E40) a current emotional state of said at least one user; - performing a sorting (E60) of said programs of said program catalog taking at least into account said detected emotional state and said genres and / or subgenres associated with said detected emotional state; - display (E80) of a list of at least one program on the audiovisual system taking into account said sorting of said programs from said program catalog.

2. Method (100) according to claim 1, wherein said step of associating (E20) said genres and / or said sub-genres with said emotional states comprises a sub-step of using (E21) a priority table in which, for each of said emotional states, said genres and / or said sub-genres are classified according to an order of priority, said classification taking into account at least a reading counter of each genre and / or sub-genre associated with each emotional state, said priority table further comprising a default genre and / or sub-genre assigned last according to the order of priority for each of said emotional states.

3. Method (100) according to claim 2, in which said step of carrying out a sorting (E60) of said programs of said catalog taking into account at least said detected emotional state and said genres and / or sub-genres associated with said detected emotional state comprises a sub-step of: - pointing (E61), in said priority table, said genre and / or said sub-genre classified first according to said order of priority for said detected emotional state, and - detection (E62) of a selection by the user and if the user does not select a program corresponding to said first genre and / or sub-genre pointed out, pointing (E63) of the next genre and / or sub-genre in the priority table until the user selects a program of said genre and / or sub-genre pointed out or until said genre and / or sub-genre pointed out is said default genre and / or sub-genre (E64).

4. Method (100) according to claim 1, wherein said step of displaying (E80) said list of at least one audiovisual program taking into account said sorting of said programs of said catalog comprises the sub-steps of: - ordering (E81) said at least one program of said list of at least one program to be displayed, said ordering taking at least in part into account a priority criterion of said at least one program; and - displaying (E82) said list of at least one program according to said ordering.

5. Method (100) according to claim 1, wherein said step of associating (E20) said genres and / or said sub-genres with said emotional states comprises a sub-step of using (E31) a counting table comprising a counter of readings by the user for each genre and / or sub-genre associated with an emotional state.

6. Method (100) according to claim 5, wherein said step of performing a sorting (E60) of said programs of said catalog taking into account at least said detected emotional state and said genres and / or sub-genres associated with said detected emotional state comprises the sub-steps of: - calculating (E71) a weighting coefficient (P(G)) for each of said genres and / or said sub-genres associated with said detected emotional state; - for each program included in a list of proposed programs in which each program presents a predetermined score and a genre and / or sub-genre, calculation (E73) of a weighted score by multiplying said predetermined score of said program by said weighting coefficient (P(G)) corresponding to said genre and / or said sub-genre; - ranking (E75) of said programs from said list of programs proposed in descending order of weighted score.

7. Method (100) according to claim 6, wherein the sub-step of calculating (E71) a weighting coefficient (P(G)) for each of said genres and / or said sub-genres associated with said detected emotional state comprises: • for said detected emotional state, the determination (E711) of a maximum value among said counters associated with each of said genres and / or said sub-genres; • for each genre and / or sub-genre associated with said detected emotional state, the division (E712) of said value of said corresponding counter by said determined maximum value.

8. Method (100) according to claim 5, wherein said step of carrying out a sorting (E60) of said programs of said catalog taking into account at least said detected emotional state and said genres and / or sub-genres associated with said detected emotional state comprises a sub-step of: - prioritization of programs from a list of proposed programs, in which each program has a predetermined score, in descending order of said counter associated with said genre and / or said sub-genre and said detected emotional state; then - ranking of said programs from said list of programs having the same genre and / or sub-genre in descending order of the predetermined score. Method (100) according to claim 2 or 5, in which it comprises a step of updating (E65, E74) said priority table or said counting table, when said at least one user reads one of said recommended programs for at least a predetermined duration, by incrementing a value of said counter associated with said program read by said at least one user.

10. Method (100) according to claim 9, wherein each genre comprises at least one sub-genre and wherein said priority table and said counting table comprise a counter of readings of each genre and / or a counter of readings of each sub-genre, the step of updating (E65, E74) said priority table and said counting table comprising the sub-steps of, when said at least one user plays one of said recommended programs: - incrementing (E651, E743) said counters of the genre and / or sub-genre of the program selected by said at least one user corresponding to said detected emotional state by a first value; - incrementing (E652, E744) the counters of the other sub-genres associated with the genre of the program selected by said at least one user corresponding to said detected emotional state by a second value.

11. Method (100) according to claim 2 or 5, wherein, when said audiovisual decoder (10) detects (E50) at least two users, the method (100) comprises: - a step of determining (E51) the main user on the basis of at least one predetermined criterion, said sub-step of using (E21) a priority table or (E31) a counting table being carried out from said priority table or said counting table associated with said determined main user; or - a step of combining (E52) said priority or counting tables of each of said detected users, said sub-step of using (E21) a priority table or (E31) a counting table being carried out from the priority table or the counting table resulting from this combination. characterized in that it comprises a step (E101) of triggering said method (100): - upon starting said audiovisual decoder (10); or - upon detection of at least one user by said audiovisual decoder (10); or - upon selection / search of a program by said at least one user.

13. Audiovisual decoder (10) for implementing the method (100) for recommending a list of at least one audiovisual program from a catalog of programs to at least one user according to any one of claims 1 to 12, said programs of said catalog being categorized into genres, said audiovisual decoder (10) being intended to be connected to a remote database, a display means (13) and a control device (121) usable by a user (12) to control said audiovisual decoder (10), said decoder (10) comprising electronic circuitry adapted and configured to: - associate said genres and / or said sub-genres with emotional states; - detect a current emotional state of said at least one user; - sort said programs of said catalog taking at least into account said detected emotional state and said genres and / or sub-genres associated with said detected emotional state;said display means (13) being intended to display a list of at least one audiovisual program taking into account said sorting of said programs of said catalog.;

14. A computer program product characterized in that it comprises instructions for implementing, by an audiovisual decoder, the method according to any one of claims 1 to 12, when said program is executed by a processor of the audiovisual decoder.

15. A storage medium characterized in that it stores a computer program comprising instructions for implementing, by an audiovisual decoder, the method according to any one of the claims- indications 1 to 12, when said program is executed by a processor of the audiovisual decoder.

Images