FR3004052A1 - METHOD AND DEVICE FOR TRANSMITTING VIDEO CONTENT, METHOD AND DEVICE FOR RESTITUTING VIDEO CONTENT, VIDEO FLOW AND CORRESPONDING COMPUTER PROGRAM. - Google Patents

Abstract

The invention relates to a method for transmitting video content comprising a video component and at least one audio or subtitling component. According to the invention, such a method comprises the following steps: identification (31) in the audio or subtitling component or components of at least two tracks each associated with a distinct audio source, the tracks being intended to be reproduced simultaneously to the video component on a terminal of a user, assigning (32) to at least one of the tracks of at least one identification data, transmission (33) of the video content comprising the tracks and the data or data of identification in a video stream.

Description

Method and device for transmitting video content, method and device for rendering video content, video stream and corresponding computer program. FIELD OF THE DISCLOSURE The field of the invention is that of the transmission or broadcasting of video content, also called audiovisual content, and the personalized rendering of such content. Such video contents are for example broadcast in a live video stream (ie "live" or in real time, with respect to the creation of video content) or offline (ie once the creation has been made), or stored on a video stream. recording medium. More specifically, the invention relates to the management of audio components and / or subtitling of such video contents. 2. Prior art Current video contents are generally composed of a video component and one or more audio components. The audio components correspond to the different languages and to the different accessible broadcast formats, and are conventionally transmitted synchronously with the video component. By language, reference is made in the rest of the document to any spoken language (English, French, Chinese, etc.). By broadcast format, also called "sound quality", is meant for example a format belonging to the group comprising the following formats: mono (for monophonic), stereo (for stereophonic), multichannel 5.1, multichannel 6.1, multichannel 7.1, DTS (of English "Digital Theater System"), Dolby, SDDS (Sony Dynamic Digital Sound), etc. Each audio component therefore includes all the sound reproduction: background noise, music, voice of all the characters, etc. For example, as illustrated in FIG. 1, a video content 11 comprises a video component 111, a first audio component 112 corresponding to the entire sound reproduction in French (also called "French sound component"), and a second audio component 113 corresponding to the entire sound record in English (also called "English sound component"). Each audio component is for example composed of a component identifier, a description, and audio data. The transmission of such video content 11 implements a simultaneous and synchronized transmission of the video component 111, the French sound component 112 and the English sound component 113. Such video contents also include, optionally, one or more components subtitling. By subtitling, we mean the written transcription of dialogues or audio comments of video content that appears in the image, a subtitle that may appear for example at the top, bottom of the image, or even directly in the image. Each subtitling component corresponds to a specific language. If the video content includes several audio or subtitling components, the end user may choose, when rendering the content, the language in which he wishes to view the video content and / or the broadcast format, by selecting the component corresponding audio and / or subtitling. Following the choice of the user on the language and / or the quality, the decoder decodes only the selected components. On the other hand, its choice is limited to these audio or subtitling components, and it is not possible for the user to build a more personalized rendering of the video content.

For example, as illustrated in FIG. 2, a decoder 21 receives the video content 11 comprising the video component 111, the French sound component 112, and the English sound component 113. The user can then select (211) the language in which he wants to view the video content, and the decoder will only decode the corresponding audio component. Thus, the decoder will restore the decoded components, for example the video component 111 and the French sound component 112. There is therefore a need for new techniques for transmitting and rendering video contents, making it possible to improve these techniques of the art. prior. 3. DISCLOSURE OF THE INVENTION The invention proposes a new solution in the form of a transmission method (and / or storage) of a video content comprising a video component and at least one audio or subtitling component, comprising the steps of: identifying in the audio or subtitle component or components at least two tracks each associated with a distinct audio source, the tracks being intended to be simultaneously rendered to the video component on a user's terminal; , assigning at least one of the tracks of at least one identification data, transmission of the video content including the tracks and identification data or data in a video stream. The invention thus proposes to improve the existing techniques for transmitting video content comprising at least one audio component and / or at least one subtitling component, by transmitting at least one identification data item enabling identification, or accessing a sub-component of an audio component or a subtitling component, called a track, which track is directly associated with an audio source. By audio source, we mean the audio source "producing" or "generating" a track (audio or subtitling), such as for example a character, a narrator, a group of characters, a music generator, a sound generator etc. The track associated with such a source corresponds for example to the voice of such a character or narrator (or cry for an animal), to a background noise (crowd noise, herd noise, ambient noise, etc.), a music (film soundtrack, game show credits, etc.), or the corresponding written transcription (or more precisely the corresponding data).

Thus, if it is considered that the audio or subtitling component or components each correspond to a language distinct from a video content, a first audio track may be associated with a first actor, and correspond to the voice of the first actor. , and a second audio track can be associated with a second actor, and correspond to the voice of the second actor. In the same way, if we consider a subtitling component, a first subtitle track can be associated with a first actor, and correspond to a written transcription of the lyrics of the first actor, and a second subtitle track. -titling can be associated with a second actor, and correspond to a written transcription of the lyrics of the second actor. It is then possible to identify these different tracks and transmit them "separately" in the video stream. Restitution side, it is then possible to restore only the voice of the first actor, or the voice of a first actor and an external voice replacing the voice of the second actor, or the voice of the first actor in French and the voice of the second actor in English, etc. In the same way, it is possible to restore a subtitle for the first actor only, or a subtitle in French for the first actor and a subtitle in English for the second actor, etc. It is thus possible for the user to build a more personalized rendering of the video content, through the transmission of audio components and / or subtitling of video content with a level of segmentation more "fine" or "small" (segmentation at the "track" level) than the standard segmentation level (segmentation at "component" level). Indeed, the audio elements forming an audio component (or a component of subtitling) are not currently identifiable and accessible individually in restitution. The identification and transmission of "tracks" according to the invention makes it possible, at the level of the terminal of a user, to access these tracks rather than the global component, and thus offers the user the possibility of constructing a more personalized rendering of the video content. In other words, thanks to the invention, it is possible to independently transmit the track associated with an audio source and the identification data of this track, which allows a user to easily identify the contribution an audio source within the video content. A separation of the sources in transmission is maintained, whereas according to the prior art, all the sources are mixed in the same "channel". In particular, at least one of the audio sources is visible in at least one image of the video content.

Thus, unlike the audio description where the image translator does not physically participate in the image taking, and is therefore not represented by an image in this video content, the invention provides access to the image. at least two tracks each associated with a distinct audio source, including at least one source that "physically" intervenes at least once in the video content, ie that is displayed at least once in the video content (actor, presenter of a game show, sports commentator, ...). If we consider the case where the audio source corresponds to a group of characters, this group of characters appears for example at least once in the video content. For example, this is the crowd present at a football match or at a concert broadcast live by a television channel.

For example, according to the invention, the audio or subtitling component is composed of at least three tracks, on the one hand two tracks each associated with a distinct audio source, the two distinct audio sources being each represented by at least two tracks. an image of the video content, and secondly another track associated with a distinct audio source not represented by an image in the video content. This example corresponds for example to two actors appearing in images of the video content combined with music, or the contribution of an image translator or any other track associated with an audio source not represented by an image in the video content. In the example above, it is possible to assign identification data to one of the two tracks associated with a separate audio source represented by at least one image of the video content, and other identification data to the track. associated with the audio source not represented by an image in the video content, for example the music, or the audio component of "description" of the audio description. Thus, in the case where the audio component is formed of at least three tracks, separate identification data is assigned to at least two tracks associated with two separate audio sources, however these two separate audio sources are not necessarily all two represented by at least one image in the video content. According to another example, it is also possible according to the invention to assign as much identification data as there are separate audio sources to intervene in the video content. According to a particular characteristic of the invention, the identification data identifies at least one audio source. We can therefore have only one identification data per audio source. In this case, the same identification data can be assigned to an audio track of an audio component and a closed caption track of a subtitling component, if they are associated with the same audio source (for example to the same actor). In this case, additional data can be inserted into the stream to indicate the type of data that composes the track (audio data, subtitle data). One can also have an identification data per set or group of audio sources. For example, we can have the same identification data for all actors, or a first identification data for all female actresses and a second identification data for all male actors, etc. According to another particular characteristic, at least one of the tracks is defined for at least two of the audio or subtitling components. In other words, at least one track is common to at least two distinct languages or at least two distinct formats. Indeed, it is for example possible according to the invention to have the same track corresponding to background noise or music for a set of languages. A track corresponding to the sound of the sea, the soundtrack of a film, or the credits of a game for example can be restored regardless of the language. This aspect thus makes it possible to avoid transmitting as much sea noise or music as there are accessible languages for the restitution of the video content. This aspect therefore has the advantage of optimizing the amount of data to be transmitted by avoiding transmitting twice the same "content" (background noise or music for example). According to a specific embodiment, the transmission method is implemented in real time.

By "real-time implementation", it is understood implementation synchronized in real time with respect to the generation of video content. This is for example the case for the live broadcast of a television program, for example a television newscast. In another embodiment, the invention relates to a device for transmitting video content comprising a video component and at least one audio or subtitling component. According to the invention, such a device comprises: an identification module, in the audio or subtitling component or components, of at least two tracks each associated with a separate audio source, the tracks being intended to be simultaneously reproduced; to the video component on a terminal of a user, a module for assigning to at least one of the tracks of at least one identification data, a module for transmitting the video content comprising the tracks and the data or data of identification in a video stream.

Such a transmission device (and / or storage) is particularly suitable for implementing the transmission method described above. It is for example integrated into a transmitter or headend of a broadcast system. This device may of course include the various characteristics relating to the transmission method according to the invention, which can be combined or taken separately. Thus, the features and advantages of this device are the same as those of the transmission method. Therefore, they are not detailed further. The invention further relates to a video stream comprising a video component and at least one audio or subtitle component, wherein at least one of the audio or subtitle components comprises at least two tracks each associated with a separate audio source. , the tracks being intended to be rendered simultaneously to the video component on a terminal of a user. According to the invention, such a video stream also comprises at least one identification datum assigned to at least one of the tracks. Such a video stream may be transmitted using the transmission method described above. It can also be stored on a data carrier, for a delayed reading. This video stream may of course include the various characteristics relating to the transmission method according to the invention. The invention also relates to a method for rendering a video content from a video stream comprising a video component, at least one audio or subtitling component, in which at least one of the audio or sub-audio components titration comprises at least two tracks each associated with a separate audio source, the tracks being intended to be simultaneously rendered to the video component on a terminal of a user. Such a stream also includes at least one identification data assigned to at least one of the tracks.

According to the invention, such a method comprises the following steps: identification of at least one of the tracks from the identification data or data, delivering at least one identified track, restitution of the video content implementing a simultaneous restitution: the video component and at least one of the tracks identified or at least one audio content generated from an external source replacing the identified track or tracks. Such a method is particularly adapted to receive and process a video stream transmitted according to the transmission method described above. It allows to record, download, display on a screen, etc., video content.

In particular, the invention makes it possible to restore, at the level of a terminal, a subset or "portion" of an audio component or a subtitling component, rather than the entirety of such a component. as proposed in the prior art. This subset corresponds to one or more audio type or subtitle tracks each associated with an audio source among all the tracks associated with the audio or subtitling component. As a result, it is possible to build a more personalized rendering of video content. All the characteristics described previously with regard to the transmission method also apply to the rendering process. Thus, with regard to these aspects previously described, the rendering method according to the invention makes it possible in particular to identify the voices of the different actors or the sub-titles associated with these actors. As a result, it is subsequently possible to select the voice and / or subtitle associated with a first actor, then to render only the voice and / or subtitle associated with this first actor or a default combination of the voice and / or subtitle associated with this first actor with the audio track and / or subtitle "ambient noise", and the audio track "music" for example the soundtrack of a film. According to this example, the voice of a second actor may not be restored, and a user can perform an oral dubbing of the second actor (with or without microphone, possibly by coupling his phone ("smartphone") to the playback device). It is therefore possible to replace one or more tracks with audio generated from an external source. According to the invention, the term audio content generated from an external source, a content that has not been transmitted in the video stream and which intervenes only during the personalized rendering of the video content, for example in replacement of a track transmitted in the video stream but not decoded.

According to a particular characteristic, the restitution step implements a step of mixing the audio content or generated from one or more external sources with at least one track of a separate audio or subtitling component. of the identified track (s) to be replaced. In other words, in order to maintain synchronization between the video and audio and / or subtitling components, the audio content generated from an external source used to replace one or more identified tracks is mixed in real time with other tracks of the audio or subtitle component not replaced. In particular, the rendering method comprises a prior step of receiving the audio content generated from an external source, when the external source is located at a remote location of the terminal.

According to another particular characteristic, the restitution step implements a step of mixing at least one track corresponding to a predetermined language with at least one track corresponding to a distinct language. Indeed, the transmission of the audio and / or subtitling components of a video content with a level of segmentation more "fine" or "small" (segmentation at the "track" level) than the level of classical segmentation (segmentation at "component" level) allows mixing when restoring tracks associated with different languages: for example reproduction of voice and / or subtitling in English for the first actor and voice and / or subtitling in French for the second actor.

According to one particular characteristic, at least one new audio content generated from an external source is added during the rendering of the video content. It is thus possible to supplement the initial video content, with or without selection of tracks, by adding new information to it. For example, it is possible to add an additional voice to the video content, to comment on a sporting event.

According to a specific embodiment, the rendering method is implemented in real time. In other words, the selection of tracks to be rendered, or the reproduction of audio content generated by an external source instead of a track, is performed as the video stream is processed.

The video stream is therefore not modified when identifying tracks for selection / replacement. It is only during the restitution of the video content that these selections / replacements are taken into account. In another embodiment, the invention relates to a device for rendering a video content from a video stream comprising a video component, at least one audio or subtitling component, according to which at least one of the components audio or subtitling includes at least two tracks each associated with a separate audio source, the tracks being intended to be simultaneously rendered to the video component on a terminal of a user. Such a video stream also comprises at least one identification data assigned to at least one of the tracks.

According to the invention, such a device comprises: a module for identifying at least one of the tracks from the identification data or data, delivering at least one identified track, a module for restoring the video content implementing simultaneous rendering: of the video component and of at least one of the identified tracks or at least one audio content generated from an external source replacing the identified track or tracks. Such a rendering device (and / or recording) is particularly suitable for implementing the restitution method described above. It is for example integrated in a terminal of a user (TV, computer, mobile phone, tablet, etc.), possibly combined with a set-top box. This device may of course include the various features relating to the restitution method described above, which can be combined or taken separately. Thus, the features and advantages of this device are the same as those of the rendering process. Therefore, they are not detailed further. In yet another embodiment, the invention relates to one or more computer programs comprising instructions for implementing the transmission method and / or instructions for implementing the restitution method as described hereinabove. above, when this or these programs are executed by a processor.

The methods according to the invention can therefore be implemented in various ways, in particular in hard-wired form or in software form. The invention also relates to a computer readable information medium, and comprising instructions of a computer program as mentioned above. 4. List of Figures Other features and advantages of the invention will appear more clearly on reading the following description of a particular embodiment, given as a simple illustrative and non-limiting example, and the accompanying drawings, among which: - Figure 1, presented in connection with the prior art, illustrates the structure of a video content according to the prior art; FIG. 2, also presented in relation with the prior art, illustrates the restitution of a video content according to the prior art; FIG. 3 presents the main steps implemented by a method for transmitting video content according to one embodiment of the invention; FIG. 4 shows the main steps implemented by a method of rendering a video content according to one embodiment of the invention; FIG. 5 illustrates the structure of a video content obtained by implementing the steps of FIG. 3; FIG. 6 illustrates the rendering of a video content obtained by implementing the steps of FIG. 4; FIGS. 7A and 7B show the fields composing a video content according to two examples of implementation of the invention; Figures 8 and 9 respectively illustrate the simplified structure of a transmission device and the simplified structure of a rendering device according to a particular embodiment of the invention. 5. Description of an embodiment of the invention 5.1 General principle The general principle of the invention is based on the identification of tracks associated with different audio sources within one or more audio and / or audio components. one or more components of subtitling of video content, and on the transmission of tracks rather than the transmission of an audio component or global subtitling, so as to allow a user to build a more personalized rendition video content. In particular, it is noted that such tracks can be identified within audio components or subtitling previously formed, or be identified during the formation of audio components or subtitling of video content. The tracks thus identified can be transmitted by maintaining this separation in tracks in the video stream, instead of being grouped in a global component "indivisible" as proposed in the prior art. The main steps implemented by a method for transmitting video content C according to a particular embodiment of the invention are presented in relation with FIG. Such video content is considered to comprise a video component, one or more audio components, and possibly one or more subtitling components. It is noted here that the term "video component" refers to all the data contributing to the restitution of the images of the video content, the "audio component" being the set of data contributing to a complete sound reproduction of the video content, and " subtitling component "all the data contributing to a complete rendering of subtitles of the video content. The invention is therefore not limited to the use of "components" as defined in the MPEG 2 standard for example.

More precisely, during a first step 31, at least two tracks each associated with a distinct audio source are identified in the audio or subtitling component or components. Note that these tracks are intended to be simultaneously rendered to the video component on a terminal of a user, especially if they are selected by a user.

During a second step 32, at least one of the tracks is assigned at least one identification data D_id. In a particular embodiment, an identification data is assigned to each track. Note that the first or the first two steps can be implemented in the studio, when creating content. For example, different tracks are identified at the sound level, corresponding, for example, to ambient noise and the voices of the various actors. Then we assign, either in the studio or during the construction of the video stream, identification data to the different tracks. We can therefore consider that we assign a "channel" to the ambient noise, and a "channel" to the voices of the actors, and that these two channels are maintained in the broadcast channel (while the ambient noise and the voices of the various actors were mixed in a single "channel" according to the prior art, thus forming an indivisible whole). These first two steps can also be implemented by decomposing one or more audio and / or subtitling components previously formed. During a third step 33, the video content in which the audio or subtitle components consist of separate tracks and the identification data or data are transmitted in a video stream F, or stored on a support for a second time. subsequent use. The main steps implemented on the rendering side are illustrated in FIG. 4. More specifically, a method of restoring a video content from a video stream F constructed as described with reference to FIG. 3 implements a first step 41. identifying at least one of the tracks from the identification data or data, delivering at least one identified track. During a second step 42, the content is restored, by simultaneously reproducing either the video component and at least one of the tracks identified to be rendered, or the video component and at least one audio content generated from an external source replacing the identified track (s) to be replaced. It is recalled that the invention can be applied to video streams transmitted or broadcast live (in English "live") or downloaded streams. 5.2 Description of a particular embodiment A particular embodiment of the invention is described below, according to which the video content that it is desired to transmit comprises a video component and two audio components, one corresponding to the French sound component and the other to the English sound component. By repeating the steps of FIG. 3, during the first step 31, the sound components are identified in its French and English video content at least two tracks each associated with a distinct audio source.

For example, with reference to FIG. 5, which represents a video content 51, there is identified a first track 521 corresponding to the voice of the actor 1 in its French, a second track 522 corresponding to the voice of the actor 2 in its French, a third track 531 corresponding to the voice of the actor 1 in his English, a fourth track 532 corresponding to the voice of the actor 2 in his English, and a fifth track 541 corresponding to the background noise. Note that this fifth track 541 corresponding to the background noise is common to the sound components French and English sound, if we consider that the background noise is invariant depending on the language. It is thus transmitted only once and identified as common to the French and English audio components.

It is thus possible to identify N audio tracks in an audio component, with N an integer greater than or equal to 2. For example, the sound component French sound is composed of three tracks: the first track 521 corresponding to the voice of the actor 1, the second track 522 corresponding to the voice of the actor 2, and the fifth track 541 corresponding to the background noise. The video content 51 also includes a video component 511.

By repeating the steps of FIG. 3, during a second step 32, at least one of the tracks is assigned at least one identification data item D_id. For example, an identification data is assigned to each track: voice1FR for the first track 521, voice2FR for the second track 522, voice1GB for the third track 531, voice2GB for the fourth track 532 and BF for the fifth track 541.

The video content including the tracks and the at least one identification data can then be transmitted in a video stream, or stored on a medium for later use. According to a first example illustrated in FIG. 7A, the audio portion of the video content is transmitted in the form of "components" as defined in the MPEG 2 standard, in which tracks have been identified. Each audio component is for example composed of a component identifier (IdC1, IdC2, ...), a description of the component (DescrC1, ...), and for each track of the audio component, an identification data of the track (IdP11, IdP12, ...), a description of the track (DescrP11, DescrP12, ...), and audio data of the track (D_P11, D_P12, ... ) according to this particular embodiment. These different fields can of course be multiplexed during transmission.

According to a second example illustrated in FIG. 7B, the audio part of the video content is directly transmitted in the form of tracks each composed of a piece of track identification data (IdP1, IdP2, ...), a description of the track (DecrP1, DescrP2, ...), and audio data of the track (D_P1, D_P2, ...). It is thus possible to transmit the voice of the actor 1 independently of the voice of the actor 2, in the same language. Thus, the "component" level as defined in the MPEG 2 standard is dispensed with.

On the restitution side illustrated in FIG. 6, a decoder 61 receives the video content comprising the different tracks and the identification data or data and decodes at least some tracks, intended to be rendered simultaneously to the video component on a terminal 62 of the user. .

In particular, if the video stream comprises the video component 511 and each audio component, where each audio component is composed of a component identifier, a description of the component, and, for each track of the audio component, an identification data of the track, a description of the track, and audio data of the track, the user can select (611) one or more audio tracks to be rendered and the decoder 61 can decode only these tracks. If the video stream comprises the video component 511 and the different tracks, each composed of a track identification data, a description of the track, and audio data of the track, the user can directly select ( 611) the one or more tracks to be rendered that interest it and the decoder 61 can decode only these tracks.

For example, the user can select the first track 521 corresponding to the voice of the actor 1 in his French from the identifier voice1FR, the fourth track 532 corresponding to the voice of the actor 2 in his English from the voice2GB identifier, and the fifth track 541 corresponding to the background noise from the identifier BF. It is thus possible to reproduce audio content with the original voices of actors, such as the voice of Sophie Marceau in French and the voice of Pierce Brosnan in English in the film "The World is not enough". According to another example, called "karaoke", the user can select the first track 521 corresponding to the voice of the actor 1 in his French from the identifier voice1FR and the fifth track 541 corresponding to the background noise at from the identifier BF, and choosing to replace the second track 522 corresponds to the voice of the actor 2 in French by external audio content, such as his own voice or that of a friend. To do this, it is possible for example to connect a microphone, or any equipment equipped with a microphone (computer, smart phone, tablet, etc.) to the decoder 61 (for example a set-top box or "set top box") and "Parody" the text of the actor 2, or "play" the role of the actor 2. The audio content thus generated can then be mixed with the other selected tracks (first track 521 corresponding to the voice of the actor 1 in its French and fifth track 541 corresponding to background noise) and the video component 511 to build a more personalized rendering of video content. Such an operation can be performed by a mixer 612 able to receive data from different sources and to mix them, and makes it possible to maintain synchronization between the external audio content, the different tracks, and the video component. We therefore seek to render video content from a video stream received by a decoder (live or delayed), and an external voice transmitted to the same decoder, by selecting a coherent subset of a component audio as the voice of a character (corresponding to a track) and replacing it with a sound like an external voice, in order to render a karaoke type.

In this example, the closed caption track associated with the same audio source as the replaced audio track (ie the closed caption track transcribing Actor 2 lyrics into French) can be automatically selected, so that the user playing the role of the actor 2 can read the text of the actor 2. According to another example, the user can choose to replace the first track 521 corresponding to the voice of the actor 1 in his French by a first content external audio, generated by a first friend, replace the second track 522 corresponds to the voice of the actor 2 in French by a second external audio content generated by a second friend, and select only the fifth track 541 corresponding to the background noise ( or choose to replace this fifth track 541 with a third external audio content generated by a third friend). For example, three friends spend the evening together and look for a distraction. They turn on the television and fall on the film "Tontons flingueurs", broadcast live and transmitted according to the method of the invention. Everyone can then choose the track associated with an actor, and replace the voice of this actor by his own voice. In particular, if one of the friends has chosen a female role (respectively male) while it is male (respectively female), it is possible to choose a "woman" option (respectively "man") at the level of the selection interface and its voice is modified in female voice (respectively male) by the mixer 612. To enable this option, it is considered that the identification data associated with the tracks include information specifying the "nature" of the track, as an acute, serious voice, etc.

It is thus possible to decode only certain tracks and to mix them with one or more audio contents coming from one or several external sources. In particular, these sources may be "local", i.e. located at the rendering terminal and directly connected to the rendering terminal, or "remote", i.e. located at a location remote from the terminal. It is also possible to combine audio content from local and remote sources. For example, our three previous friends are in their respective apartments, and wish to watch the same video content. Everyone can choose the track (i.e. the voice) he wants to interpret. The audio content generated by each of the three friends can be retrieved by the respective decoder of each of the friends. Each decoder then mixes the audio contents of the three friends possibly with other selected tracks (such as background noise for example) and renders it with the video component on the terminal of each user. The audio contents generated by the three friends can also be retrieved by a remote intermediate application, implemented in a third party equipment of the network, which mixes them possibly with other selected tracks (like the background noise for example) and the rebroadcast with the video component to all of the friends. According to yet another example, an operator can broadcast a game / contest channel which consists in hiding the voice of certain characters in the broadcasted videos. A user can then identify the track or tracks corresponding to the voice of the character (s) to be masked, and replace this track with his / her own voice. The video content thus generated can be recorded and then transferred to the contest site, and the best interpretation is declared a winner. In summary, the choice of different tracks can be made by displaying, at the terminal of at least one user, a selection interface, for selecting one or more tracks, associated with one or more audio components, to be restored simultaneously with the video component or replaced by external audio content. It is also possible to display at this interface external audio content that you simply want to add in restitution to the original video content. Note that it is not possible to select two tracks associated with the same audio source but belonging to different audio components: for example, it is not possible to simultaneously select the first track 521 corresponding to the voice of the audio. actor 1 in his French and the third track 531 corresponding to the voice of the actor 1 in his English. In addition, some tracks can be selected by default, such as background noise. 5.3 Variants Different variants are possible. In particular, various examples above relating to the identification of a set of audio tracks independent of one or more audio components have been described. Of course, it is also possible to identify a set of closed caption tracks independent of one or more subtitling components, for example to display the text of the actor 1 in French and the text of the subtitles. actor 2 in English, or to display only the text of the actor 1 if one wishes to replace the audio track associated with the actor 1 in French by an external audio content.

A subtitling track specific to a single character is therefore possible according to the invention. It is also possible to restore the voice of the actor 1 in French and the subtitling associated with the actor 1 year English, and to restore the voice of the actor 2 in English and the subtitling of the actor 2 in French. It is thus possible to render tracks belonging to different initial components, or to mix different languages at audio level and / or subtitling.

It is also possible to add, at the time of the return of the video content, content generated by an external source in addition to this rendered video content, and not only in substitution. This is particularly relevant when the main video content is created from a broadcast base (such as the national edition of a newscast), and then supplemented with external content (such as the regional newscast edition). This option may also be of interest when the video content presents a sports match, which friends wish to comment on. It is also possible to add external content to other actions than playing the video stream, such as recording. Indeed, conventionally, when recording at a user's terminal of a stream broadcast live, all components (video, audio, and possibly subtitling) are recorded, out of decoding. According to the invention, it is possible to record new content generated by adding content generated by an external source to the original content, or to record new content obtained by replacing certain tracks of the initial content with content generated by a source external, see to delete / invalidate some tracks present in the stream broadcast. On the other hand, in the examples described above, it has been considered that an identification datum per track was affected. We can therefore have only one identification data per audio source. In this case, the same identification data can be assigned to an audio track of an audio component and a closed caption track of a subtitling component, if they are associated with the same audio source (for example to the same actor). In this case, additional data can be inserted into the stream to indicate the type of data that composes the track (audio data, subtitle data). One can also have an identification data per set of audio source. For example, we can have the same identification data for all actors, or a first identification data for all female actresses and a second identification data for all male actors, etc. In particular, identification data may be assigned to a group of at least two tracks, the group being formed according to a predetermined criterion of grouping characteristic of the audio sources associated with these tracks. In other words, the same identification data is in this case assigned to two tracks associated with two different audio sources having the same characteristic corresponding to the predetermined criterion of grouping. This aspect of the invention makes it possible, for example, to distinguish: a group of female voices or a group of male voices according to a predetermined criterion corresponding to the sex of the audio sources associated with said group voice sub-components, a group of children's voices or unlike mature people according to a predetermined criterion corresponding to the age of the audio sources associated with said group voice subcomponents, a group of cowboy voices or, conversely, Indians according to a predetermined criterion corresponding to the belonging to a tribe of the audio sources associated with said group voice subcomponents, - a group of nocturnal or diurnal sounds according to a predetermined criterion corresponding to the brightness and / or the presence of the day in the images of the group. video content, a group of sounds indoors or outdoors according to a predetermined criterion of location of audio sources, etc. Thus, according to this aspect of the invention, there is less identification data than tracks of the audio component or subtitling. Indeed, there are for example as many tracks indian, indian2, indian3 ... indian N that there are actors (in the number of N) playing the role of indians in the video content of film type , whereas only "Indian" identification data is assigned to these "Indian" N tracks. Such a "partial" assignment makes it possible to reduce the number of identification data to be transmitted. According to another variant, it is also possible to implement a "double assignment step", according to which identification data is assigned to each track associated with this distinct audio source, and identification data is assigned to a group tracks whose corresponding audio sources are characterized by the same predetermined grouping criterion. Thus, the invention makes it possible, inter alia, for: - during the generation and the transmission of flows, an identification of N tracks in an audio or subtitling component (during the formation of the component or by decomposition of a component existing), consisting for example of background noise and voices of each character of the video content, or a textual transcription of the background noise and voices of each character; a selective local decoding of tracks chosen by the user, among all the components / tracks proposed. The human-machine interface of the decoder can allow a user to choose the transmitted track or tracks to be restored or replaced by one or more external sounds associated for example with microphones, in addition to (or instead of) choosing the language in which content is played (French, English, ...) or the quality (stereo, 5.1, ...) a real-time mixing in the decoder between the decoded tracks, from the video content or from external sources, for the restitution of the same global content. The advantages of the proposed solution include: an extension of the audio choices or subtitling by the user; a personalization of the sound by external sounds (from other sources); a possibility of composing global content from remote sources; a possibility to mix different languages according to the actors (mix of tracks belonging to different initial components). The solution therefore allows an operator to offer a new interactive and fun use of these broadcast content, for example by offering a karaoke option (possibly paid). The option "karaoke" allows to add value to the content broadcast, and is of interest both in the form of a game but also in the form of learning in training in diction and intonations. transmitted stream according to the invention can also be restored in a "conventional" manner. The invention thus provides an additional service, in addition to the conventional rendering of video content. 5.4 Description of the transmission and reproduction devices according to the invention Finally, with reference to FIGS. 8 and 9 respectively, the simplified structure of a transmission device implementing a technique for transmitting video content and the structure of a rendering device implementing a technique for rendering a video content according to a particular embodiment of the invention. As illustrated in FIG. 8, such a transmission device comprises a memory 81 comprising a buffer memory, a processing unit 82, equipped for example with a microprocessor, uP and driven by the computer program 83, implementing the method transmission according to the invention.

At initialization, the code instructions of the computer program 83 are for example loaded into a RAM memory before being executed by the processor of the processing unit 82. The processing unit 82 receives as input a content video C to broadcast. The microprocessor of the processing unit 82 implements the steps of the transmission method described above, according to the instructions of the computer program 83, to identify the audio and / or subtitle tracks composing the video content and assign them identifying data, and transmitting a video stream F comprising the identification data and the video content. For this, the transmission device further comprises: a module M_Id 84 for identifying at least two tracks each associated with a distinct audio source, said tracks being intended to be simultaneously returned to said video component on a terminal of a user, a module M_D_id 85 assigning to at least one of the tracks of at least one identification data, and a module M_Tx 86 for transmitting the video content comprising the tracks and the identification data or data in the stream These modules are controlled by the microprocessor of the processing unit 82. As illustrated in FIG. 9, such a device for rendering a video content comprises a memory 91 comprising a buffer memory, a processing unit 92, equipped for example with a microprocessor, uP, and driven by the computer program 93, implementing the method of rendering a video content according to the invention. At initialization, the code instructions of the computer program 93 are for example loaded into a RAM memory before being executed by the processor of the processing unit 92. The processing unit 92 receives as input the stream F video, as well as a selection of tracks to render / replace. The microprocessor of the processing unit 92 implements the steps of the rendering method described above, according to the instructions of the computer program 93, to identify the different tracks and build a more personalized rendering of the video content. For this, the rendering device further comprises: a module M_D_id 94 identifying at least one track from the identification data or data, delivering at least one identified track, and a module M_rest 95 for restitution of the video content implementing simultaneous rendering: of the video component and at least one of the identified tracks or at least one audio content generated from an external source replacing the identified track or tracks. These modules are driven by the microprocessor of the processing unit 92.

Claims (13)

REVENDICATIONS1. A method for transmitting video content comprising a video component and at least one audio or subtitling component, characterized in that it comprises the following steps: identification (31) in said at least one audio or sub-sub component. titling at least two tracks each associated with a distinct audio source, said tracks being intended to be simultaneously rendered to said video component on a terminal of a user, assignment (32) to at least one of said tracks of at least one identification data, transmission (33) of said video content comprising said at least two tracks and said at least one identification data in a video stream. 2. Method for transmitting video content according to claim 1, characterized in that at least one of said audio sources is visible in at least one image of said video content. 3. A method of transmitting video content according to any one of claims 1 and 2, characterized in that said at least one identification data identifies at least one audio source. 4. A method of transmitting video content according to any one of claims 1 to 3, characterized in that the one or more audio or subtitling components each correspond to a language and / or a distinct sound quality. 5. A method of transmitting video content according to any one of claims 1 to 4, characterized in that at least one of said tracks is defined for at least two of said audio or subtitling components. 6. Video stream comprising a video component and at least one audio or subtitling component, at least one of said audio or subtitle components comprising at least two tracks each associated with a separate audio source, said tracks being intended to be simultaneously rendered to said video component on a terminal of a user, characterized in that said video stream also comprises at least one identification data assigned to at least one of said tracks. 7. Method for rendering a video content from a video stream comprising a video component, at least one audio or subtitling component, at least one of said audio or subtitling components comprising at least two associated tracks each to a separate audio source, said tracks being intended to be simultaneously rendered to said video component on a user's terminal, and said stream also comprising at least one identification data assigned to at least one of said tracks, characterized in that said method comprises the following steps: identification (41) of at least one of said tracks from said at least one identification data, delivering at least one identified track, restitution (42) of said video content implementing a restitution simultaneous: - of said video component and - of at least one of said identified tracks or of at least one audio content generated from a source this external replacing said at least one track identified. 8. Process for rendering a video content according to claim 7, characterized in that said restitution step implements a step of mixing said at least one audio content generated from an external source with at least one audio track. one of said audio or subtitle components, separate from said at least one identified track. 9. A method of restoring a video content according to any one of claims 7 and 8, characterized in that said restitution step implements a step of mixing at least one track corresponding to a predetermined language with at least a track corresponding to a separate language. 10. Method for rendering a video content according to any one of claims 7 to 9, characterized in that it comprises a prior step of receiving said audio content generated from an external source, when said external source is located at a location remote from said terminal. 11. Device for transmitting a video content comprising a video component and at least one audio or subtitling component, characterized in that it comprises: an identification module (84), in said at least one audio component or subtitling, at least two tracks each associated with a separate audio source, said tracks being intended to be simultaneously rendered to said video component on a terminal of a user, an assignment module (85) to at least one of said tracks of at least one identification data, a transmission module (86) of said video content comprising said at least two tracks and said at least one identification data in a video stream. 12. Device for rendering a video content from a video stream comprising a video component, at least one audio or subtitling component, at least one of said audio or subtitling components comprising at least two associated tracks each to a separate audio source, said tracks being intended to be simultaneously rendered to said video component on a user's terminal, and said stream also comprising at least one identification data assigned to at least one of said tracks, characterized in that said device comprises: an identification module (94) of at least one of said tracks from said at least one identification data, delivering at least one identified track, a rendering module (95) of said video content setting simultaneous rendering of: said video component and at least one of said identified tracks or at least one audio content generated from a external source replacing said at least one identified track. Computer program comprising instructions for carrying out a method according to claim 1 or claim 7 when said program is executed by a processor.