EP2874336A1 - Method for measuring the audience of content on terminals, by extracting digital watermarks and/or digital fingerprints from the audio component of the content - Google Patents

Abstract

There is provided a method of measuring an audience of content broadcast in streaming mode or by downloading over a computer network, the method being implemented with an audience measurement server and a plurality of terminals. A given terminal performs the following steps: a) it receives a tracing block previously associated with the content, or associates a tracing block with the content; b) during the consumption of the content with a multimedia player included in the given terminal, the tracking block performs the following steps: * obtain from the multimedia player an audio component of the content; extracting, from the audio component obtained, at least one digital tattoo and / or at least one digital fingerprint comprising at least one absolute time index; * transmit to the audience measurement server tracing messages each comprising a unique tracing identifier, stored by the given terminal, and said at least one digital tattoo and / or said at least one digital imprint.

Description

1. DOMAIN OF THE INVENTION

The field of the invention is that of audience measurement techniques.

More specifically, the invention relates to a technique for measuring the audience of a content broadcast on a computer network (for example the Internet), in continuous mode (also called "streaming content" or "streaming content") or in download (eg pre-roll, mid-roll and post-roll downloaded in file mode). This technique is implemented with an audience measurement server and a plurality of terminals.

The invention applies in particular, but not exclusively, to the measurement of the user-centric audience of a streamed content (TV program for example) transmitted via the Internet network. In this case, the technique is implemented with a group of panellists distributed by households and having at least a part of the plurality of terminals. Each household has either a single panelist or several panellists. In the case of a household with a single panelist, it is not excluded that other people share the same household and terminals; these people will not be known to the audience measurement company who will be able to treat them as guests in the reporting interfaces.

• tout type de réseau informatique ;
• tout type de contenu streamé ou téléchargé (contenu audio et/ou vidéo, contenus en direct ou en différé, TV de rattrapage (« catch up TV »), Vidéo à la demande (VOD), Webradios, ...), que ce soit en ligne (contenus disponibles sur les sites Internet, c'est-à-dire référencés dans les pages web consultables avec un navigateur exécuté par le terminal) ou sur applications tierces (contenus disponibles avec les applications tierces (aussi appelées « applications mobiles » ou « applications éditeur ») exécutées par le terminal) ;
• tout type de terminal (fixe ou mobile, personnel ou partagé) : ordinateur, smartphones, tablettes, TV connectées, etc.

More generally, the proposed technique can be applied to:

• any type of computer network;
• any type of streamed or downloaded content (audio and / or video content, live or deferred content, catch-up TV, video-on-demand (VOD), Webradios, etc.) either online (content available on Internet sites, ie referenced in web pages that can be viewed with a browser run by the terminal) or on third-party applications (content available with third-party applications (also known as "mobile applications") or "publisher applications") executed by the terminal);
• any type of terminal (fixed or mobile, personal or shared): computers, smartphones, tablets, connected TVs, etc.

2. TECHNOLOGICAL BACKGROUND

• la mesure de la fréquentation, basée sur des outils et technologies appelés « site-centric » en anglais, ou « côté sites » en français ; et
• la mesure de l'audience (au sens strict), basée sur des outils et technologies appelés « user-centric » en anglais, ou « côté utilisateurs » en français.

Two major measures (also called "Web Analytics" tools) coexist today to measure the audience (in a broad sense) of a website, on the Internet:

• the measurement of attendance, based on tools and technologies called "site-centric" in English, or "side sites" in French; and
• the measurement of the audience (in the strict sense), based on tools and technologies called "user-centric" in English, or "user side" in French.

2.1 The "site-centric" measure

It is a measurement made on request of the website which itself marks the web pages (HTML pages) that it wishes to measure. The main objective of the site-centric measure is the measurement of traffic volumes. It allows a fine and precise follow-up of the user's behavior on the site, the most consulted contents, the most used services, the Internet origin of the visits. The "site-centric" measure also applies to measuring the effectiveness of online marketing actions: clicks, conversion rate, turnover, return on investment, etc.

The operation of the "site-centric" measurement relies on the marking of each page of a site with a marker (also called "tag"), for example a Javascript code. As soon as a page is called by a web browser via a terminal connected to the Internet (computer, smartphones, tablets, TV connected, etc.), the marker placed on the page sends a request ("hit") to the server of measured. The traffic on the pages is thus counted. The "site-centric" measure is used to analyze the number of terminals (identified by their Internet browsers) that visited a site during a given period. Browsers are identified from cookies, or from the IP address / User Agent pair, when cookie measurement is not possible. For a given browser and terminal, a cookie is deposited on the terminal by the server of the audience measurement company (Web analytics system), the first time a marker is downloaded from the server of the measurement company hearing. This cookie is common to all sites marked by the audience measurement company, since it is a "third party cookie" managed by this company.

The technique of marking web pages (originally designed for measuring website traffic as explained above) was then extended to third-party applications (also called "mobile applications" or "publisher applications"). It has been proposed to insert a marker ("tag") in the mobile application. For this purpose, there are tagging libraries for the different mobile application platforms. Mobile apps are tools that make it easy to find and view all types of content for users. A mobile application, when connected, behaves globally as a browser, since at each operation or data access, information is read on the Internet. Unlike the browser, the mobile application can generally take into account the functionality of the terminal on which it is run, thus offering the user increased efficiency.

The technique of marking web pages and applications can be summarized as follows. When a publisher tags his website to allow an Internet audience measurement company to measure the corresponding audience, it includes a marker in its HTML pages (the marker then consists of a call to a JavaScript routine on the server of the audience measurement company) or in its third-party applications (the marker then consists of a library written in native code and integrated with the third-party application). When loading each page or the third-party application, a tagging identifier (unique and persistent) is generated (if this has not already been done). In the case of a page (of a publisher's site) that can be consulted with the web browser of the terminal, the marking identifier is deposited in the form of a cookie on the terminal (it is generally common to all the marked sites by the audience measurement company). In the case of a third-party application proposed by an editor on a third-party application store, the marking identifier is stored in a storage area of the terminal, specific to each application, if it is not possible to use the unique identifier of the terminal. It is usually different from one application to another. During the presentation of a content (including launching), the marker sends periodic requests (also called "hits") on the Internet audience measurement server, to transmit the operating status of the reading device (player). These hits convey the identity of the content, the position in the content (stream), the state of the reading device (pause / read), the tagging identifier (allowing to de-duplicate the sessions of the same terminal at the same time). same day), as well as a temporary identifier of session renewed with each new launch of the reading device. In nominal mode, hits are sent every 60 seconds. The technique of marking web pages and applications makes it possible to measure the consumption (number and duration) of contents.

A disadvantage of the marking technique is that it is limited to the "site-centric" measure, independently of any notion of panel (this notion being specific to the "user-centric" measure). The "site-centric" measure does not make it possible to identify in a distinct way the Internet users who connect to the site (we do not know who is behind the terminal connected to the Internet). It does not allow for a qualified audience. This is why the "user-centric" measure (detailed below) complements the "site-centric" measure.

Another disadvantage of the marking technique is that the marker sends hits to the audience measurement server with time indexes relating to the start of the player's launch in the HTML page or the third-party application. In case of pause in the reading of the content (flow), the hits indicate the change of state and the index does not increment anymore. For delayed content (for example, catch-up TV content), this does not create a measurement error since the player receives the content from the beginning. On the other hand, for a content (live stream), there is no absolute time reference with respect to the content itself (ie no absolute time index), its transmission being able to undergo a shift more or less variable due to the Internet distribution networks. Therefore, it is difficult to know precisely which part of a content (for example a broadcast or an advertising tunnel) has actually been viewed by a user since the programming grid (pige) which serves as a reference is that of the broadcast radio.

Yet another disadvantage of the tagging technique is that it is radically different from that currently used for the daily TV benchmark (see the first known user-centric measurement technique described hereinafter). It would therefore be necessary to build a hybrid infrastructure at the back-office level (that is to say at the level of the audience measurement server), making it possible to manage both television sets and new types of terminals (computers, smartphones , tablets, etc.).

2.2 The user-centric measure

It is based on the monitoring of surfing on the Internet of a panel (sample of Internet users) representative of a population of Internet users (for example French Internet users). Panel navigation is measured and a scoping survey is done in parallel to extrapolate their navigation to the Internet population studied. By precisely identifying the panellist, the "user-centric" measure is used to analyze the number of unique visitors who visited a site during a given period. A single panel visitor is a qualified individual identified. In extrapolated value, a panelist corresponds to thousands of unique visitors. The "user-centric" measure provides agencies, publishers and marketing professionals with accurate and reliable reports on the audience of Internet sites and the use of Internet users, which information can be used for strategic planning, competitive studies or media planning. For example, the results are updated every month.

A first known technique of user-centric audience measurement is based on the audio watermarking of the audio channel (also called "audio component") of the broadcasted content (TV channels for example) , or the comparison of their audio fingerprinting with reference fingerprints.

This first known technique is the basis of the meters used for daily TV reference measurement. In the case of the television set, the audimeter (typically a mini-PC) captures the audio channel of the displayed channel (electrically or acoustically), in order to perform the signal processing and the transmission of the audience data. to the server of the audience measurement company. A disadvantage of this first known technique is that, in the case of laptops, smartphones or tablets, it is difficult to envisage permanently connecting an audimeter (that is to say an external device) to these equipment by mobile nature. Another disadvantage of this first known technique is that on some terminals (typically Apple terminals (registered trademark)), it is very difficult to capture the audio channel in digital format, directly at the audio circuits.

A second known technique of user-centric audience measurement is based on the analysis of the network traffic between the terminal and the Internet network. This second Known technique is to exploit network traces left at a router installed in the home panelists or at a http proxy through which the terminals of panelists must pass. The main disadvantage of this second known technique is that the network traffic can be completely decorrelated content actually restored to the terminal. For example, when the user puts the playback device (player) in pause, the audiovisual stream continues to be stored in buffer memory, with no guarantee that it will be decoded and viewed later. There is therefore an inaccuracy in the measurement of the streamed contents that can be buffered in the terminals and never consumed. Another disadvantage is that this solution is limited to the home where the router is located.

A third known "user-centric" audience measurement technique relies on a measurement software ("meter software") embedded in the terminal in order to transmit and analyze the connection data (surf) in real time. The main disadvantage of this third known technique is that it requires a considerable R & D effort to maintain a measurement software on all operating systems (OS) and all Internet browsers terminals, taking into account the evolution of major players of the Web, and that as well on computer (PC), tablet, smartphone, etc. On the other hand, when an audiovisual content (stream) is protected by a technical protection measure (or DRM, for "Digital Rights Management" in English), the measurement software can not usually measure the consumption.

3. OBJECTIVES OF THE INVENTION

The invention, in at least one embodiment, is intended in particular to overcome these various disadvantages of the state of the art.

More specifically, in at least one embodiment of the invention, an objective is to provide a technique for measuring the audience of a content broadcast on a computer network (for example the Internet), in a continuous mode (streamed content). or in download mode, without imposing an audimeter (individual device) to the panelists, thus increasing the acceptance of the measure and its reliability.

Another objective of at least one embodiment of the invention is to provide such a technique, making it possible to perform a "site-centric" or "user-centric" measurement.

Another objective of at least one embodiment of the invention is to provide such a technique for combining a "site-centric" measurement with a "user-centric" measurement.

Another objective of at least one embodiment of the invention is to provide such a technique, operating both at home users (for example panelists) and mobility (at work, on vacation, in transport, etc.).

Another objective of at least one embodiment of the invention is to provide such a technique, which is simple to implement and inexpensive.

Another objective of at least one embodiment of the invention is to provide such a technique, which requires only a small modification of the sites and applications for accessing streamed or downloaded content, and therefore a very difficult effort. ecosystem actors (including the publishers of these sites and applications).

Another objective of at least one embodiment of the invention is to provide such a technique allowing, for a content broadcast live (live) on a computer network (for example the Internet), to perform an audience measurement. by becoming independent of potentially variable transmission delays in this network.

4. PRESENTATION OF THE INVENTION

1. a) il reçoit un bloc de traçage préalablement associé au contenu, ou il associe un bloc de traçage au contenu ;
2. b) pendant la consommation du contenu avec un lecteur multimédia compris dans le terminal donné, le bloc de traçage effectue les étapes suivantes :
   • * obtenir auprès du lecteur multimédia une composante audio du contenu ;
   • * extraire, de la composante audio obtenue, au moins un tatouage numérique et/ou au moins une empreinte numérique comprenant au moins un index temporel absolu ;
   • * transmettre au serveur de mesure d'audience des messages de traçage comprenant chacun un identifiant unique de traçage, stocké par le terminal donné, et ledit au moins un tatouage numérique et/ou ladite au moins une empreinte numérique.

In a particular embodiment of the invention, there is provided a method for measuring an audience of a content broadcast in continuous mode or by downloading on a computer network, the method being implemented with a data measurement server. audience and a plurality of terminals. A given terminal performs the following steps:

1. a) it receives a tracing block previously associated with the content, or associates a tracing block with the content;
2. b) during the consumption of the content with a multimedia player included in the given terminal, the tracing block performs the following steps:
   • * obtain from the media player an audio component of the content;
   • extracting, from the audio component obtained, at least one digital tattoo and / or at least one digital fingerprint comprising at least one absolute time index;
   • * transmit to the audience measurement server tracing messages each comprising a unique tracing identifier, stored by the given terminal, and said at least one digital tattoo and / or said at least one digital imprint.

Thus, the proposed solution is based on a completely new and inventive approach, based on the use of audio digital watermarking technology (audio watermarking) or digital audio fingerprinting (audio fingerprinting). The innovation is based solely on a software implementation on the terminal: the tracing block (executed by the terminal) obtains from the multimedia player the audio component of the content, then treats it in the same way as an audimeter (extraction of digital tattoos or fingerprints, and transmission of trace messages to the audience measurement server).

• elle ne nécessite pas d'appareil externe (de type audimètre) pour capturer le son ;
• elle permet de réaliser une mesure en « user-centric » (si au moins certains des terminaux sont utilisés dans des foyers panélistes et enregistrés comme tels auprès du serveur de mesure d'audience - voir détail plus bas) ou en « site-centric » (sinon) ;
• elle fonctionne aussi bien au domicile des utilisateurs (par exemple des panélistes) qu'en mobilité : dès qu'un terminal est connecté à un débit suffisant pour pouvoir recevoir un contenu (flux) streamé ou téléchargé, le bloc de traçage (exécuté par le terminal) transmet des messages de traçage au serveur de mesure d'audience ;
• du fait qu'elle s'appuie sur une technologie de tatouage numérique audio
  (« audio watermarking ») ou d'empreinte numérique audio (« audio fingerprinting ») déjà utilisée dans la mesure de référence TV quotidienne, elle permet de mesurer l'audience sur les nouveaux types de terminaux (ordinateurs, smartphones, tablettes, etc., autres que les postes de télévision), elle permet de réutiliser l'infrastructure serveur (qu'il n'est donc pas utile d'adapter pour ces nouveaux types de terminaux) ;
• du fait que le serveur de mesure d'audience reçoit des tatouages numériques (watermarks) ou des empreintes numériques (fingerprints) contenant des index temporels absolus extraits du contenu, la solution proposée permet au serveur de mesure d'audience de récupérer une référence absolue et donc d'effectuer une mesure d'audience indépendante des délais de transmission éventuellement variables dans le réseau (par exemple Internet), même pour un contenu diffusé en direct (live) sur le réseau.

The proposed solution has several advantages:

• it does not require an external device (audimeter type) to capture the sound;
• it allows a "user-centric" measurement (if at least some of the terminals are used in panellist households and recorded as such with the audience measurement server - see detail below) or in "site-centric" (if not) ;
• it works equally well in the homes of users (for example panelists) and mobile: as soon as a terminal is connected at a rate sufficient to be able to receive streamed or downloaded content (stream), the trace block (executed by the terminal) transmits trace messages to the audience measurement server;
• because it relies on digital audio tattoo technology
  (Audio watermarking) or audio fingerprinting already used in the daily TV benchmark measurement, it can measure the audience on new types of terminals (computers, smartphones, tablets, etc.). , other than television sets), it allows to reuse the server infrastructure (which is therefore not useful to adapt for these new types of terminals);
• because the audience metering server receives watermarks or fingerprints containing absolute time indices extracted from the content, the proposed solution allows the audience measurement server to retrieve an absolute reference and therefore to make an independent audience measurement potentially variable transmission times in the network (eg Internet), even for content broadcast live on the network.

In a first implementation, in step a), the given terminal receives at least one web page in which the tracking block is integrated, said at least one web page referencing at least part of the content and being executable with an included browser. in the given terminal. In step b), the tracking block, when executed by the browser, communicates with the media player to obtain the audio component of the content, the media player being integrated with or accessed by the browser.

In this first implementation, content editors just have to integrate the tracking block into the web pages of their websites, including the contents.

In a second implementation, in step a): the given terminal receives at least one web page referencing at least part of the content and executable with a browser included in the given terminal; and an extension module, included in the given terminal and completing the browser, integrates the tracking block with said at least one web page, before the execution of said at least one web page by the browser. In step b), the tracking block, when executed by the browser, communicates with the media player to obtain the audio component of the content, the media player being integrated with or accessed by the browser.

In this second implementation, content publishers just have to distribute the plug-in, and therefore do not have to modify their websites.

In a third implementation, in step a), the given terminal receives a third-party application for accessing the content when it is executed by the given terminal, the third-party application being written in native code and encapsulating the written trace block in executable code with a browser included in the given terminal. In step b), the tracking block, when executed by the browser, communicates with the media player to obtain the audio component of the content, the media player being integrated with or accessed by the browser.

In this third implementation, content editors just have to integrate into their third-party applications a tracing block executable by the terminal browser.

In a fourth implementation, in step a), the given terminal receives a third-party application allowing access to the content when it is executed by the given terminal, the third-party application being written in native code and integrating the tracing block as well. written in native code. In step b), the trace block, when executed by the terminal, communicates with the media player to obtain the audio component of the content, the media player being integrated with or accessed by the third-party application.

In this fourth implementation, the content editors just have to integrate into their third-party applications a tracing block executable by the terminal (directly, without the help of the browser).

According to a particular characteristic, said unique tracing identifier is a unique identifier of the given terminal, stored by the terminal.

Thus, it is possible to reuse an already existing terminal identifier (for example IDFA ("Identifier For Advertiser") under Apple's iOS operating system, or Android_ID under Google's Android operating system), which avoids that the terminal or the audience measurement server has to generate it.

According to one variant, said unique tracing identifier is an identifier generated by the tracing block and then stored by the terminal.

The generated identifier is for example stored in a cookie.

1. i) il s'enregistre comme terminal panéliste auprès du serveur de mesure d'audience;
2. ii) au moins une fois, il transmet au serveur de mesure d'audience un message de déclaration comprenant :
   • * l'identifiant unique de traçage ;
   • * un identifiant du terminal donné, ou une information relative au type du terminal donné ; et
   • * une liste de panéliste(s) déclaré(s).

In a particular embodiment, the method is implemented with a group of panellists distributed by households and having a first set of terminals included in said plurality of terminals, each home comprising at least one panelist. The given terminal, if it is part of said first set of terminals, also performs the following steps when it is used by at least one panelist of a given household:

1. i) it is registered as a panellist terminal with the audience measurement server;
2. ii) at least once, it transmits to the audience measurement server a declaration message comprising:
   • * the unique tracing identifier;
   • an identifier of the given terminal, or information relating to the type of the given terminal; and
   • * a list of panelist (s) declared.

Thus, it is possible to perform a user-centric audience measurement.

According to a particular characteristic, in step i): the given terminal transmits to the audience measurement server the identifier and the type of the given terminal; and the audience measurement server associates the identifier and the type of the given terminal with a unique identifier linked to the given focus and a list of possible panelists related to the given focus.

According to a variant, in step i): the given terminal transmits to the audience measurement server the type of the given terminal; and the audience measurement server generates an identifier of the given terminal, and associates the identifier and the type of the given terminal with a unique identifier linked to the given focus and a list of possible panelists related to the given focus.

According to a particular characteristic, the given terminal executes a declaration application to perform steps i) and ii). The tracing block forwards the unique tracing identifier to the declaration application.

According to a first variant, the given terminal executes a declaration application to perform step i). The tracing block performs step ii). The unique tracing identifier is a unique identifier of the terminal, which is stored by the given terminal and accessible by the declaration application and the tracing block.

According to a second variant, the given terminal executes a declaration application to perform step i). The tracing block performs step ii). The unique tracing identifier is a unique identifier of the terminal. In step i), the declaration application launches the browser on a URL containing said unique identifier of the terminal and that when it is executed deposits on the terminal a cookie containing said unique identifier of the terminal, said cookie being accessible by the tracing block when it is executed by the browser.

According to a third variant, the browser performs step i), during which is deposited on the terminal a cookie containing said unique identifier of the terminal, said cookie being accessible by the tracking block when it is executed by the browser, in the unique tracing identifier is said unique identifier of the terminal, and in that the tracing block performs step ii).

• ledit groupe de panélistes répartis par foyers et disposant dudit premier ensemble de terminaux compris dans ladite pluralité de terminaux ; et
• des utilisateurs non-panélistes disposant d'un second ensemble de terminaux
  compris dans ladite pluralité de terminaux ;

et le serveur de mesure d'audience effectue une étape de redressement :

• de premières données de mesure d'audience, obtenues à partir d'une corrélation entre les messages de traçage et les messages de déclaration transmis par les terminaux du premier ensemble,
• en fonction de secondes données de mesure d'audience, obtenues à partir des messages de traçage transmis par les terminaux du second ensemble.

According to a particular characteristic, the method is implemented with:

• said group of panellists distributed by households and having said first set of terminals included in said plurality of terminals; and
• non-panellist users with a second set of terminals
  included in said plurality of terminals;

and the audience measurement server performs a recovery step:

• first audience measurement data, obtained from a correlation between the trace messages and the declaration messages transmitted by the terminals of the first set,
• based on second audience measurement data, obtained from the trace messages transmitted by the terminals of the second set.

Thus, we combine a measure in "site-centric" with a measure in "user-centric". The data of one ("site-centric") allowing to straighten the data of the other ("user-centric").

In another embodiment of the invention there is provided a computer program product which comprises program code instructions for implementing the aforesaid method (in any one of its various embodiments), when said program is run on a computer.

In another embodiment of the invention, there is provided a computer-readable and non-transitory storage medium, storing a computer program comprising a set of instructions executable by a computer for implementing the aforementioned method (in any of its different embodiments).

1. a) des moyens de réception d'un bloc de traçage préalablement associé au contenu, ou des moyens d'association d'un bloc de traçage au contenu ;
2. b) les moyens suivants, compris dans le bloc de traçage et activés pendant la consommation du contenu avec un lecteur multimédia compris dans le terminal :
   • * des moyens d'obtention auprès du lecteur multimédia d'une composante audio du contenu ;
   • * des moyens d'extraction, de la composante audio obtenue, d'au moins un tatouage numérique et/ou au moins une empreinte numérique comprenant au moins un index temporel absolu ;
   • * des moyens de transmission, au serveur de mesure d'audience, de messages de traçage comprenant chacun un identifiant unique de traçage, stocké par le terminal donné, et ledit au moins un tatouage numérique et/ou ladite au moins une empreinte numérique.

In another embodiment of the invention, there is provided a terminal included in a system for measuring an audience of a content broadcast in continuous or download on a computer network, the system comprising an audience measurement server and a plurality of terminals to which said terminal belongs. Said terminal comprises:

1. a) means for receiving a tracing block previously associated with the content, or means for associating a tracing block with the content;
2. b) the following means, included in the trace block and activated during the consumption of the content with a multimedia player included in the terminal:
   • means for obtaining from the multimedia player an audio component of the content;
   • means for extracting, from the audio component obtained, at least one digital tattoo and / or at least one digital fingerprint comprising at least one absolute time index;
   • means for transmitting, to the audience measurement server, tracing messages each comprising a unique tracing identifier, stored by the given terminal, and said at least one digital tattoo and / or said at least one digital imprint.

Advantageously, the terminal comprises means for implementing the steps it performs in the audience measurement method as described above, in any of its various embodiments.

5. LIST OF FIGURES

• la figure 1 présente l'architecture logicielle d'un terminal dans un cas idéal ;
• la figure 2A présente l'architecture logicielle d'un terminal dans une première implémentation particulière de l'invention, avec un bloc de traçage compris dans une page HTML ;
• la figure 2B présente l'architecture logicielle d'un terminal dans une deuxième implémentation particulière de l'invention, avec un bloc de traçage ajouté dans une page HTML par un module d'extension (plug-in) ;
• la figure 3 présente l'architecture logicielle d'un terminal dans une troisième implémentation particulière de l'invention, avec un bloc de traçage compris dans une application éditeur ;
• la figure 4 présente un diagramme de séquence illustrant le fonctionnement d'un procédé de mesure d'audience selon un premier mode de réalisation de l'invention, avec une application de déclaration « forte » et un bloc de traçage « léger » dans une application éditeur ou une page HTML ;
• la figure 5 présente un diagramme de séquence illustrant le fonctionnement d'un procédé de mesure d'audience selon un deuxième mode de réalisation de l'invention, avec une application de déclaration « légère » et un bloc de traçage « fort » dans une application éditeur ;
• la figure 6 présente un diagramme de séquence illustrant le fonctionnement d'un procédé de mesure d'audience selon un troisième mode de réalisation de l'invention, avec une application de déclaration « légère » et un bloc de traçage « fort » dans une page HTML ;
• la figure 7 présente un diagramme de séquence illustrant le fonctionnement d'un procédé de mesure d'audience selon un quatrième mode de réalisation de l'invention, avec une application de déclaration « légère » et un bloc de traçage « fort » dans une page HTML ;
• la figure 8 présente un diagramme de séquence illustrant le fonctionnement d'un procédé de mesure d'audience selon un cinquième mode de réalisation de l'invention, sans application de déclaration et avec un bloc de traçage « fort » dans une page HTML ; et
• la figure 9 présente la structure d'un terminal selon un mode de réalisation particulier de l'invention.

Other features and advantages of the invention will appear on reading the following description, given by way of indicative and nonlimiting example, and the appended drawings, in which:

• the figure 1 presents the software architecture of a terminal in an ideal case;
• the Figure 2A presents the software architecture of a terminal in a first particular implementation of the invention, with a tracing block included in an HTML page;
• the Figure 2B presents the software architecture of a terminal in a second particular implementation of the invention, with a tracing block added in an HTML page by an extension module (plug-in);
• the figure 3 presents the software architecture of a terminal in a third particular implementation of the invention, with a tracing block included in an application editor;
• the figure 4 shows a sequence diagram illustrating the operation of an audience measurement method according to a first embodiment of the invention, with a "strong" declaration application and a "light" trace block in an editor application or a HTML page;
• the figure 5 presents a sequence diagram illustrating the operation of an audience measurement method according to a second embodiment of the invention, with a "light" declaration application and a "strong" trace block in a publisher application;
• the figure 6 shows a sequence diagram illustrating the operation of an audience measurement method according to a third embodiment of the invention, with a "light" declaration application and a "strong" trace block in an HTML page;
• the figure 7 shows a sequence diagram illustrating the operation of an audience measurement method according to a fourth embodiment of the invention, with a "light" declaration application and a "strong" trace block in an HTML page;
• the figure 8 shows a sequence diagram illustrating the operation of an audience measurement method according to a fifth embodiment of the invention, without declaration application and with a "strong" trace block in an HTML page; and
• the figure 9 presents the structure of a terminal according to a particular embodiment of the invention.

6. DETAILED DESCRIPTION

In all the figures of this document, the elements and identical steps are designated by the same numerical reference.

The proposed technique makes it possible to measure the audience, on terminals (computers, smartphones, tablets, etc.), of content broadcast in continuous mode (streamed content) or download on a computer network (eg Internet).

6.1 Ideal case

As illustrated on the figure 1 Ideally, we would like to have an application audience measurement 6 that the panelist would install on their device. This would turn in the background, without discontinuity, on the operating system 9 of the terminal. It would include a trace block 7 having the task of capturing (through the operating system 9) the audio signal 8 restored on the terminal, to extract digital tattoos (watermarks) or fingerprints, and this independently of the client applications 4, 5 accessing the audio circuits (each via an instance 10, 11 of the media player (Media Player)).

Unfortunately, some devices such as those operated under Apple's iOS operating system do not allow such an implementation. On the one hand, running in the background is a privilege that Apple only grants to a certain type of applications (listening to music, GPS navigation ...). On the other hand, the processing of the audio signal captured at the level of the media player (Media Player) is carried out within one application to the exclusion of all others (see the following link: https://developer.apple .com / library / ios / documentation / MusicAudio / Conceptual / CoreAu dioOverview / WhatisCoreAudio / WhatisCoreAudio.html, which reads: "Because you must statically link your audio code to your application, audio units that you can not use other applications in iOS ").

The object of the present invention is to circumvent this difficulty for the following two scenarios: access to the content (flow) from an HTML page and access to the content (flow) from a publisher application. These two cases are discussed in detail below.

6.2 Tracing block included in a publisher application 6.2.1 Software architecture of the terminal

The figure 3 presents the software architecture of a terminal in a particular implementation of the invention, with a tracing block 1 included in a publisher application 30. The latter is for example proposed by the editor and downloadable from an online store of third-party applications (AppStore, PlayStore, ...).

It is common that this type of editor application is only encapsulation in native code of an HTML application. In other words, in this case, the terminal receives an editor application 30 for accessing the content when it is executed by the given terminal. This editor application is written in native code and encapsulates a trace block 1 written in executable code with a browser included in the terminal. In this case, the second solution that follows is applicable (cf § 6.3 and Figure 2A the tracking block 1, when it is executed by the browser 23, communicates with the media player 21 to obtain the audio component 24 of the content, the multimedia player 21 being integrated into or accessible by the browser).

On the other hand, in the case where the editor application 30 is entirely written in native code, it integrates the trace block 1 also written in native code (it is for example an audio signal processing library). For the restitution of the Audio / Video content, the editor application 30 uses a media player 33 (which executes itself on the operating system 32 of the terminal). The editor application 30 accesses the audio signal, in the heart of the media player, and passes the audio frames 34 to the tracing block 1 (also called "watermarks / fingerprints extraction module") integrated into the application editor.

• « https://developer.apple.com/library/ios/documentation/MusicAudio/Conceptual/Audi oUnitHostingGuide_iOS/AudioUnitHostingFundamentals/AudioUnitHostingFundame ntals.html ».

For example, under the iOS operating system, it is possible to access the audio signal at the "Audio Unit" layer 31, as indicated in the following link:

• "Https://developer.apple.com/library/ios/documentation/MusicAudio/Conceptual/Audi oUnitHostingGuide_iOS / AudioUnitHostingFundamentals / AudioUnitHostingFundame ntals.html".

• « http://developer.android.com/reference/android/media/AudioRecord.html ».

Under the Android operating system, the "AudioRecord" API provides this same functionality through the "Read" function, as shown in the following link:

• "Http://developer.android.com/reference/android/media/AudioRecord.html".

• il obtient auprès du lecteur multimédia une composante audio du contenu (trames audio 34) ;
• il extrait, de la composante audio obtenue, des tatouages numériques (watermarks) et/ou des empreintes numériques (fingerprints) comprenant chacun au moins un index temporel absolu ;
• il transmet au serveur de mesure d'audience des messages de traçage comprenant chacun un identifiant unique de traçage, stocké par le terminal donné, et un des tatouages numériques et/ou une des empreintes numériques.

During the consumption of the content with the media player 33, the tracing block 1 performs the following steps:

• it obtains from the media player an audio component of the content (audio frames 34);
• it extracts, from the audio component obtained, watermark tattoos and / or digital fingerprints each comprising at least one absolute time index;
• it transmits to the audience measurement server tracing messages each comprising a unique tracing identifier, stored by the given terminal, and one of the digital tattoos and / or one of the digital fingerprints.

In other words, the tracing block 1 associates a unique tracing identifier with each watermark or fingerprint transmitted to the audience measurement server. This unique tracing identifier can be generated by the tracing block and stored for example in a zone accessible by the editor application 30. It is also possible to use as a unique tracing identifier a unique terminal identifier (for example IDFA under iOS, or Android_ID on Android, ...) which has the advantage of associating the same identifier for all the application publishers using this technology.

6.2.2 Embodiments of the audience measurement method

The figure 4 presents a sequence diagram illustrating the operation of an audience measurement method according to a first embodiment of the invention, with a "strong" declaration application 2 and a "light" trace block 1 in a publisher application (cf. figure 3 ) or an HTML page (cf. Figures 2A and 2B ).

In this first embodiment, the system comprises an audience measurement server 3 and a group of panellists distributed by households and having a plurality of terminals (for the sake of simplification, a single terminal 90 is represented on the figure 4 ). The terminals are for example computers, smartphones, tablets, connected TVs, etc.

In addition to the software architecture elements already described in relation to the Figures 2A, 2B and 3 , including the tracking block 1, the terminal includes a declaration application 2. This declaration application 2 (developed by the audience measurement company) is for example downloaded from an application store or from the Internet portal of the audience measurement company. Each focus of the panel is committed to install this application of declaration on each of its terminals (PC, smartphone, tablet). Each household also receives a unique identifier and a password (the identifier can be for example its email address and the password a panellist household number).

The general principle of the first embodiment is as follows: trace block 1 sends to the audience measurement server 3 successive trace messages (each containing a watermark or a fingerprint comprising an absolute time index) and, in parallel, the declaration application 2 sends to the audience measurement server 3 successive declaration messages (each containing a list of panelists declared). The audience measurement server 3 performs a correlation between these two types of message, taking as association key a unique tracing identifier contained both in the tracing messages and in the declaration messages. Thus, the audience measurement server 3 is able to associate each watermark or fingerprint (contained in one of the consumption messages) with one of the lists of panelists declared (s) (contained in the one of the declaration messages). By performing this operation for all the terminals of each of the households of panelists, the audience measurement server 3 has, for each content that the tracking block 1 can measure, information on the consumption of this content by panelists, and can therefore perform audience measurement for this content.

• une phase d'enregistrement, référencée 4-1 et comprenant les étapes 41 à 44 ; et
• une phase de suivi de consommation d'un contenu donné, référencée 4-II et comprenant les étapes 45 à 415.

We now present the detailed operation of the system of the figure 4 (first particular embodiment). There are two phases:

• a recording phase, referenced 4-1 and comprising the steps 41 to 44; and
• a consumption monitoring phase of a given content, referenced 4-II and comprising steps 45 to 415.

In step 41, the declaration application 2 transmits to the audience measurement server 3 a registration request, parameterized with the login / password pair (previously assigned to the home and entered by a user of the home). terminal), as well as with an identifier of the terminal (terminal ID), the type of the terminal, information indicating whether the terminal is shared or not, and an identifier of the user (in the case of an un-shared terminal) .

In step 42, after verifying the identifier / password pair provided by the declaration application 2, the audience measurement server 3 proceeds to register the terminal 90 as a panellist terminal. It associates the identifier (terminal ID) and the type of the terminal with a unique identifier of the focus (focus ID).

In step 43, the audience measurement server 3 transmits to the terminal 90 the unique identifier of the home (home ID).

In step 44, the terminal 90 stores the unique identifier of the home (focus ID), in a storage area managed by the declaration application 2.

In a variant of step 41, the declaration application 2 does not transmit to the audience measurement server 3 the identifier of the terminal (terminal ID). In this case, it is the audience measurement server 3 which generates it (in step 42) and transmits it to the terminal 90 (in step 43), so that the latter stores it (in the step 44).

In step 45, at the start of the reading of a content, the tracing block 1 requests the opening of the declaration application 2 (for example via the technique of the "URL scheme") and passes it a unique tracing identifier (tracing ID).

This implies that the terminal 90 has previously obtained and stored this tracing identifier, for example loading the HTML page or the third-party application (depending on whether the tracing block 1 is integrated with one or the other) . In the case of an HTML page viewable with the web browser of the terminal 90, the tracking identifier is for example deposited in the form of a cookie on the terminal ("third party cookie" managed by the measurement company of hearing). In the case of a third-party application, the tracing identifier is for example stored in a storage area of the terminal, specific to this third-party application.

In step 46, the declaration application 2 sends to the audience measurement server 3 a request parameterized with the unique identifier of the home (home ID), in order to obtain a list of possible panelists for this home.

In response, in step 47, the audience measurement server 3 transmits to the declaration application 2 the list of possible panelists.

In step 48, the declaration application 2 displays an individual declaration window (also called a "declaration interface"), displaying the list of possible panelists (obtained in step 47). Via the individual declaration window, a user selects one or more panelists from the list of possible panelists, to form a list of panelist (s) declared.

The list of possible panellists and the list of panelist (s) declared can be implemented in a variety of ways. In a first implementation, each list is a list of unique identifiers of panelists (possibly each associated with a first name, for ease of understanding), which the audience measurement company has previously recorded in a database (during recruitment home panelist) and to go back to the unique identifier of the home (ID foyer). In a second implementation, each list includes non-unique identifiers of panelists (for example a list of first names that the audience measurement company has previously recorded in a database, during the recruitment of the panellist home), and is transmitted with the unique identifier of the household (focus ID).

In step 49, the declaration application 2 transmits to the audience measurement server 3 a declaration message comprising the unique tracing identifier (tracing ID), the terminal identifier (terminal ID) and / or the type of terminal, and the list of panelist (s) declared (list obtained in step 48).

In step 410, the audience measurement server 3 associates the unique tracing identifier (tracing ID) with the identifier of the terminal (terminal ID), and therefore with the unique identifier of the home (home ID) and the type of the terminal.

In step 411, the declaration application 2 gives back to the tracing block 1 (for example via the "URL scheme" technique).

Then, in step 412, the tracing block 1 actually launches the playback of the content.

During (including at launch) the consumption of the content, the tracking block 1 sends to the audience measurement server 3 trace messages each containing a watermark or a fingerprint (itself including an absolute time index) and the unique tracing identifier (tracing ID). On the figure 4 for the sake of simplicity, only two trace message transmission steps, referenced 413 and 414, are shown.

In the first embodiment illustrated on the figure 4 , the idea is to separate the measure from the consumed content (with the sending of trace messages) and the declaration of panelists (with the sending of declaration messages). Both are correlated (on the fly or a posteriori) by the audience measurement server 3, in step 415, using the unique tracing identifier (tracing ID) and time as association keys.

• * il détermine, parmi les listes de panéliste(s) déclaré(s) contenues dans les messages de déclaration reçus successivement, la liste de panéliste(s) déclaré(s) qui était la liste courante à l'instant donné ;
• * il associe le watermark ou le fingerprint donné à un ensemble de paramètres comprenant l'identifiant unique de traçage (ID traçage), l'identifiant unique de foyer (ID foyer), l'identifiant du terminal (ID terminal), le type de terminal et la liste de panéliste(s) déclaré(s) qui était la liste courante à l'instant donné.

In step 415, for a given watermark or fingerprint, received at a given moment, the audience measurement server 3 performs the following steps for example:

• * it determines, among the lists of panelist (s) declared in the declaration messages received successively, the list of panelist (s) declared (s) which was the current list at the given moment;
• * it associates the given watermark or fingerprint with a set of parameters including the unique identifier of tracing (ID tracing), the unique identifier of focus (ID foyer), the identifier of the terminal (ID terminal), the type of terminal and list of panelist (s) declared which was the current list at the given time.

In the step of determining the list of panelist (s) declared (s) which was the current list at the given time, the audience measurement server 3 for example performs the treatment on the fly: it stores the last list of panelist (s) declared, contained in the last declaration message he / she has received; and when he receives (at the given moment) the given watermark or fingerprint, he takes as current list at the given moment the last list of panelist (s) declared.

• il stocke, en association avec une première information d'horodatage, chaque liste de panéliste(s) déclaré(s) contenue dans l'un des messages de déclaration qu'il a reçus successivement ;
• il stocke, en association avec une deuxième information d'horodatage, le watermark ou le fingerprint donné ;
• il détermine la liste courante à l'instant donné, en fonction de la deuxième information d'horodatage et des premières informations d'horodatage stockées avec les listes de panéliste(s) déclaré(s) qu'il a reçues successivement.

In a variant, in the step of determining the list of panelist (s) declared (s) which was the list current at the given time, the audience measurement server 3 performs the following post-processing:

• it stores, in association with a first timestamping information, each list of panelist (s) declared (s) contained in one of the declaration messages that he received successively;
• it stores, in association with a second time stamping information, the given watermark or fingerprint;
• it determines the current list at the given time, according to the second timestamp information and the first timestamp information stored with the lists of panelist (s) declared (s) he received successively.

In this variant, the timestamp of each list of panelist (s) declared (that is, its association with a first timestamp information) and the timestamp of each watermark or the fingerprint (c ') ie its association with a second time stamping information) can be performed either by the terminal or by the audience measurement server.

It is possible that a content is consumed while the terminal (for example a mobile terminal) momentarily loses its connection with the audience measurement server 3. In this case, the trace messages of the trace block 1 are lost, which generates a measurement error. In order to overcome this measurement error, it is possible to implement an interpolation correction strategy.

It is also possible that a change of panelist (s) (ie a change in the list of panelist (s) declared) occurs during a loss of connection between the terminal and the measurement server 3. In this case, this change must be notified to the audience measurement server 3, in order to assign subsequent trace messages to the good panelist (s). One solution is, for example, to timestamp the messages of declarations at the terminal and to provide a buffering (buffer storage) of the declaration data, in the declaration application (or elsewhere in the terminal), in order to shift in time the sending of the declaration data.

The figure 5 presents a sequence diagram illustrating the operation of an audience measurement method according to a second embodiment of the invention, with a "light" declaration application 2 and a "strong" trace block 1 in a publisher application (cf. figure 3 ).

To overcome the possible difficulty of running a declaration application 2 in the background on certain platforms (operating systems), as well as the difficulty of transmitting parameters between two applications without deteriorating the user experience, the second mode embodiment of the invention consists in using the declaration application 2 only to record a unique identifier of the terminal (for example IDFA, Android_ID, ...) on the audience measurement server 3, as unique tracing identifier (tracing ID). This unique identifier of the terminal must be accessible both by the application of declaration 2 that by the tracing block 1 (the latter being included in the editor application). By associating this identifier of the terminal with the household identifier (focus ID) corresponding to the login / password entered in the declaration application 2, it is thus possible to declare a panelist terminal.

• une phase d'enregistrement, référencée 5-1 et comprenant les étapes 51 et 52 ; et
• une phase de suivi de consommation d'un contenu donné, référencée 5-II et comprenant les étapes 53 à 515.

We now present the detailed operation of the system of the figure 5 (second particular embodiment). There are two phases:

• a recording phase, referenced 5-1 and comprising steps 51 and 52; and
• a consumption monitoring phase of a given content, referenced 5-II and comprising steps 53 to 515.

In step 51, the declaration application 2 transmits to the audience measurement server 3 a registration request, parameterized with the login / password pair (previously assigned to the home and entered by a user of the home). terminal), the type of the terminal, information as to whether the terminal is shared or not, and an identifier of the user (in the case of a non-terminal). sharing).

In step 52, after verifying the identifier / password pair provided by the declaration application 2, the audience measurement server 3 proceeds to register the terminal 90 as a panelist terminal. It associates the identifier of the terminal (for example IDFA) and the type of the terminal with a unique identifier of the focus (focus ID).

In step 53, at the start of the reading of a content (restitution of the content by the editor application), the tracing block 1 transmits a request to the audience measurement server 3, with as parameter the identifier of the content. terminal (eg IDFA), to know if the terminal is a panelist terminal.

In step 54, the audience measurement server 3 determines whether the terminal is a panelist terminal, thanks to the identifier of the terminal (for example IDFA). Then in step 55, it sends the corresponding response to the tracing block 1.

In step 56, if the terminal is a panelist terminal, the tracing block 1 transmits a request to the audience measurement server 3 to know if a new panelist statement (s) is needed.

In step 57, the audience measurement server 3 determines whether a new panelist declaration (s) is necessary (for example by analyzing a possible inactivity of the terminal). Then in step 58, it sends the corresponding response to trace block 1, including a statement box (including a list of possible panelists for that focus) if a new statement is needed.

• soit la boîte de déclaration (aussi appelée « fenêtre de déclaration » ou « interface de déclaration »), affichant la liste de panélistes possibles, si une nouvelle déclaration est nécessaire. Via cette boîte de déclaration individuelle, un utilisateur sélectionne un ou plusieurs panélistes de la liste de panélistes possibles, pour former une liste de panéliste(s) déclaré(s) ;
• soit un pictogramme discret que l'utilisateur peut à tout moment activer pour lui-même provoquer une nouvelle déclaration.

In step 59, the tracing block 1 shows on the terminal screen, superimposed (without interrupting the flow):

• either the declaration box (also called the "declaration window" or "reporting interface"), displaying the list of possible panelists, if a new declaration is needed. Via this individual declaration box, a user selects one or more panellists from the list of possible panelists, to form a list of panelist (s) declared;
• either a discreet pictogram that the user can at any time activate for himself cause a new declaration.

As already described above, the list of possible panellists and the list of panelist (s) declared can be implemented in various ways.

In step 510, the tracing block 1 transmits to the audience measurement server 3 a declaration message comprising the identifier of the terminal (for example IDFA, acting as a unique tracing identifier) (ID tracing) and the list of panelist (s) declared.

Then, in step 511, the tracing block 1 actually starts the playback of the content.

During (including at launch) the consumption of the content, the tracking block 1 sends to the audience measurement server 3 trace messages each containing a watermark or a fingerprint (itself including an absolute time index) and the identifier of the terminal (eg IDFA, playing the role of unique identifier of tracing (ID tracing)). On the figure 5 for the sake of simplicity, only two trace message transmission steps, referenced 512 and 513, are shown.

When a condition is verified (for example every x minutes), the mechanism of steps 56 to 510 is reiterated, in order to allow trace block 1 to send a new declaration to the audience measurement server 3, if necessary. .

As in the first embodiment, the idea of this second embodiment is to dissociate the measurement of the consumed content (with the sending of trace messages) and the declaration of the panelists (with the sending of declaration messages). . Both are correlated (on the fly or a posteriori) by the audience measurement server 3, in step 515, using the identifier of the terminal (for example IDFA, acting as the unique tracing identifier ( Tracing ID)) and time as association keys (see above description of step 415 of the figure 4 ).

6.3 Tracing block included in an HTML page 6.3.1 Software architecture of the terminal

The Figure 2A presents the software architecture of a terminal in another particular implementation of the invention, with a tracing block 1 included in an HTML page 20. In this case, it is considered that the content (stream) is accessed from this HTML page, downloadable from a website of the publisher. For the restitution of the Audio / Video content, the browser 23 (which itself runs on the operating system 22 of the terminal) uses a media player 21. The media player 21 accesses the audio signal and passes the audio frames 24 to trace block 1 (also called "watermarks / fingerprints extraction module") integrated in the HTML page 20. In other words, the trace block 1 (included in the HTML page 20), when it is executed by the browser 23, communicates with the media player 21 to obtain the audio component 24 of the content, the media player 21 is integrated in or accessible by the browser.

• « http://chromium.googlecode.com/svn/trunk/samples/audio/visualizer-gl.html ».

To fully understand the proposed technique, you should know that the W3C has defined under the impetus of Google a programming interface (API) called "Web Audio API" to access audio media by a Javascript script to perform various treatments in relation to the sound. This API is naturally supported first and foremost on Google's Chrome browser. The following example is illustrative of what the API can do:

• "Http://chromium.googlecode.com/svn/trunk/samples/audio/visualizer-gl.html".

It can be seen in this example that it is possible to calculate in real time the Fourier transform of the audio file listened to by a Javascript code integrated into the HTML page and to display its frequency spectrum.

There is also a W3C initiative, inspired by the Web Audio API, which aims to access audio / video streams for processing (see https://dvcs.w3.org/hg/audio/raw-file /tip/streams/StreamProcessing.html). In the first place, this API targets applications such as browser-based video telephony, which means that you can access the microphone and camera from the HTML code.

In the particular implementation of the Figure 2A , the proposed technique consists in defining, standardizing and exploiting a new API allowing access to the audio track of an audiovisual stream streamed from a Javascript script integrated into an HTML page. The script (tracing block) then processes the audio signal in the same way as an audimeter: extraction of watermarks or fingerprints, transmission of data to the audience measurement server.

Note that it is only in the case where streamed content (stream) received is watermarked that watermarks will be transmitted to the audience measurement server. On the other hand, when using audio channel fingerprinting technology, the fingerprints extracted by the trace block will be transmitted continuously to the audience measurement server for comparison with the reference fingerprints.

The trace block (script) associates a unique tracing identifier with each watermark or fingerprint transmitted to the audience measurement server. This unique tracing identifier can be generated by the script and stored for example in a cookie. It is also possible to use as a unique tracing identifier a unique terminal identifier (for example IDFA on iOS, or Android_ID on Android, ...) which has the advantage of associating the same identifier for all Internet sites. using this technology.

Like a marking technology, it is therefore enough for an audiovisual content editor to integrate this tracking block (new API for example in the form of Javascript code) to each of the HTML pages of its website to enable the measure of the audience of the contents (video streams for example) referenced in these HTML pages on the terminals equipped with Internet browsers supporting this new API.

6.3.2 Embodiments of the audience measurement method

In the case where the tracing block 1 is included in an HTML page, the first embodiment of the invention (described above in relation to the figure 4 ) can be implemented, in the same way. He is not described again.

On the other hand, the second embodiment of the invention (described above in relation to the figure 5 ) can not be implemented identically. Indeed, unlike an editor application, the browser when it executes the tracing block 1 (for example a script in Javascript code) can not access the unique identifier of the terminal (for example IDFA, Android_ID, ...) . It is therefore necessary to modify the recording phase 5-1 of the second embodiment. The Figures 6 to 8 present three solutions in this direction.

• une phase d'enregistrement, référencée 6-1 et comprenant les étapes 61 à 63 ; et
• une phase de suivi de consommation d'un contenu donné, référencée 6-II et identique à la phase référencée 5-II sur la figure 5. Cette phase n'est pas décrite à nouveau.

The figure 6 presents a sequence diagram illustrating the operation of an audience measurement method according to a third embodiment of the invention, with a "light" declaration application and a "strong" trace block in an HTML page (cf. Figure 2A ). There are two phases:

• a recording phase, referenced 6-1 and comprising steps 61 to 63; and
• a consumption monitoring phase of a given content, referenced 6-II and identical to the phase referenced 5-II on the figure 5 . This phase is not described again.

Steps 61 and 62 are identical to steps 51 and 52 respectively.

In step 63, the declaration application 2 launches the browser on a URL containing the unique identifier of the terminal (for example IDFA, playing the role of unique identifier of tracing (ID tracing)) in parameter and which, when it is executed, deposits on the terminal a cookie containing this unique identifier of the terminal. This cookie is accessible by trace block 1 when it is executed by the browser, in phase 6-11, to know the unique identifier of the terminal.

• une phase d'enregistrement, référencée 7-1 et comprenant les étapes 71 à 73 ; et
• une phase de suivi de consommation d'un contenu donné, référencée 7-II et identique à la phase référencée 5-II sur la figure 5. Cette phase n'est pas décrite à nouveau.

The figure 7 presents a sequence diagram illustrating the operation of an audience measurement method according to a fourth embodiment of the invention, with a "light" declaration application and a "strong" tracing block in an HTML page (cf. Figure 2A ). There are two phases:

• a recording phase, referenced 7-1 and comprising steps 71 to 73; and
• a consumption monitoring phase of a given content, referenced 7-II and identical to the phase referenced 5-II on the figure 5 . This phase is not described again.

Step 71 is different from step 51 of the figure 5 in that the declaration application 2 does not transmit to the audience measurement server 3 the identifier of the terminal. In this case, in step 72, it is the audience measurement server 3 that generates it (terminal ID) and associates it with the terminal type and the unique identifier of the home (home ID).

In step 73, the declaration application 2 launches the browser on a URL containing the unique identifier of the terminal (terminal ID, playing the role of unique identifier of tracing (ID tracing)) and which, when it is executed, deposits on the terminal a cookie containing this unique identifier of the terminal. This cookie is accessible by trace block 1 when it is executed by the browser, in phase 7-11, to know the unique identifier of the terminal.

• une phase d'enregistrement, référencée 8-1 et comprenant les étapes 81 à 83 ; et
• une phase de suivi de consommation d'un contenu donné, référencée 8-II et identique à la phase référencée 5-II sur la figure 5. Cette phase n'est pas décrite à nouveau.

The figure 8 presents a sequence diagram illustrating the operation of an audience measurement method according to a fifth embodiment of the invention, without declaration application and with a "strong" trace block in an HTML page (cf. Figure 2A ). There are two phases:

• a recording phase, referenced 8-1 and comprising steps 81 to 83; and
• a consumption monitoring phase of a given content, referenced 8-II and identical to the phase referenced 5-II on the figure 5 . This phase is not described again.

Steps 81 to 83 are distinguished from steps 71 to 73 of the figure 7 in that they are executed by the browser (instead of the declaration application).

In other words, to go even further, this variant completely disregards a declaration application, by registering the panelist terminal from the browser 23. To do this, in step 81, the panelist is invited to connect, with the browser of his terminal, to the audience measurement server 3 where he is asked to identify himself by username / password so that a unique identifier of the terminal (Terminal ID) is generated and saved locally in a cookie for example (the browser can not access the terminal identifier of the terminal). type IDFA or Android_id, so this type of identifier will not be used).

6.4 Tracing block added in an HTML page by an extension module 6.4.1 Software architecture of the terminal

The Figure 2B presents the software architecture of a terminal in another particular implementation of the invention, with a tracing block 1 added to an HTML page 20 by a plug-in 25 completing the browser 23.

This extension module, which only the panelists will install in the case of a "user-centric" measurement, integrates the tracing block 1 (script) with the loaded HTML page 20 (as symbolized by the arrow referenced 26). This prevents publishers from modifying their website. After the tracing block 1 has been integrated in the HTML page 20, the operation is identical to the case of the Figure 2A : for the restitution of the Audio / Video content of the HTML page 20, the browser 23 (which runs itself on the operating system 22 of the terminal) uses a media player 21. The media player 21 accesses the audio signal and passes the audio frames 24 to the tracing block 1 that the extension module 25 has integrated into the HTML page 20.

6.4.2 Modes of performing the audience measurement process

Since the tracing block 1 is integrated in the HTML page 20 by the plug-in module, it is possible to implement the same embodiments as in the case where the tracing block 1 is already integrated in the received HTML page, namely the first, third, fourth and fifth embodiments of the invention (described above in relation to the figures 4 , 6 , 7 and 8 respectively). They are not described again.

6.5 User-centric data recovery with site-centric data

• de premières données de mesure d'audience, obtenues à partir d'une corrélation entre les messages de traçage et les messages de déclaration transmis par les terminaux du premier ensemble (comme décrit plus haut en relation avec les figures 4 à 8),
• en fonction de secondes données de mesure d'audience, obtenues à partir des messages de traçage transmis par les terminaux du second ensemble (ces messages n'étant corrélés à aucun message de déclaration).

In a particular embodiment, the audience measurement method is implemented with a group of panelists, distributed by households and having a first set of terminals, and a group of non-panelists having a second set terminals. The audience measurement server performs a recovery step:

• first audience measurement data obtained from a correlation between the trace messages and the declaration messages transmitted by the terminals of the first set (as described above in relation to the Figures 4 to 8 )
• based on second audience measurement data obtained from the tracing messages transmitted by the terminals of the second set (these messages not being correlated to any declaration message).

6.6 Example of the structure of the terminal and the audience measurement server

The figure 9 presents the simplified structure of a terminal 90 according to a particular embodiment of the invention, implementing any one of the particular embodiments described above in relation to the Figures 2A, 2B and 3 to 7 . This terminal comprises a random access memory 93 (for example a RAM memory), a processing unit 92, equipped for example with a processor, and controlled by a computer program stored in a read-only memory 91 (for example a ROM or a hard disc). At initialization, the code instructions of the computer program are for example loaded into the random access memory 93 before being executed by the processor of the processing unit 92.

• sur une machine de calcul reprogrammable (un ordinateur PC, un processeur DSP ou un microcontrôleur) exécutant un programme comprenant une séquence d'instructions, ou
• sur une machine de calcul dédiée (par exemple un ensemble de portes logiques comme un FPGA ou un ASIC, ou tout autre module matériel).

This figure 9 illustrates only one particular way, among several possible, to realize on the terminal side the various algorithms detailed above. Indeed, the technique of the invention is carried out indifferently, on the terminal side:

• on a reprogrammable calculating machine (a PC computer, a DSP processor or a microcontroller) executing a program comprising a sequence of instructions, or
• on a dedicated computing machine (for example a set of logical gates such as an FPGA or an ASIC, or any other hardware module).

In the case where the technique of the invention is carried out, on the terminal side, on a reprogrammable calculation machine, the corresponding program (that is to say the sequence of instructions) can be stored in a removable storage medium ( such as for example a floppy disk, a CD-ROM or a DVD-ROM) or not, this storage medium being readable partially or totally by a computer or a processor.

The audience measurement server structure (referenced 3 on the Figures 4 to 7 ) is for example identical to that presented on the figure 9 for the terminal 90 (structure comprising a RAM, a processing unit which for example equipped with a processor and driven by a computer program stored in a ROM). As for the terminal, it is only a particular way, among several possible, to achieve server side different algorithms detailed above. Indeed, the technique of the invention, server side, is carried out indifferently: on a reprogrammable calculation machine (a PC computer, a DSP processor or a microcontroller) executing a program comprising a sequence of instructions, or on a machine dedicated calculation (for example a set of logic gates such as an FPGA or an ASIC, or any other hardware module).

Claims (18)

A method of measuring an audience of content broadcast in streaming mode or by downloading over a computer network, the method being implemented with an audience measurement server and a plurality of terminals, characterized in that a terminal given the following steps: a) it receives a tracing block previously associated with the content, or associates a tracing block with the content; b) during the consumption of the content with a multimedia player included in the given terminal, the tracing block performs the following steps: * obtain from the media player an audio component of the content; extracting, from the audio component obtained, at least one digital tattoo and / or at least one digital fingerprint comprising at least one absolute time index; * transmit to the audience measurement server tracing messages each comprising a unique tracing identifier, stored by the given terminal, and said at least one digital tattoo and / or said at least one digital imprint; and in that in step a): The given terminal receives at least one web page in which the tracking block is integrated, the at least one web page referencing at least part of the content and being executable with a browser included in the given terminal; or The given terminal receives at least one web page referencing at least part of the content and executable with a browser included in the given terminal, and an extension module, included in the given terminal and completing the browser, integrates the tracking block said at least one web page, before the execution of said at least one web page by the browser; or The given terminal receives a third-party application making it possible to access the content when it is executed by the given terminal, the third-party application being written in native code and encapsulating the trace block written in executable code with a browser included in the given terminal. ; or The given terminal receives a third-party application allowing access to the content when it is executed by the given terminal, the third-party application being written in native code and integrating the tracing block also written in native code. Method according to claim 1, characterized in that , in step a), the given terminal receives at least one web page in which the tracking block is integrated, said at least one web page referencing at least part of the content and being executable with a browser included in the given terminal,
and in that , in step b), the tracking block, when executed by the browser, communicates with the media player to obtain the audio component of the content, the media player being integrated with or accessed by the browser. Process according to claim 1, characterized in that in step a): the given terminal receives at least one web page referencing at least part of the content and executable with a browser included in the given terminal; and an extension module, included in the given terminal and completing the browser, integrates the tracking block with said at least one web page, before the execution of said at least one web page by the browser;
and in that , in step b), the tracking block, when executed by the browser, communicates with the media player to obtain the audio component of the content, the media player being integrated with or accessed by the browser. Method according to claim 1, characterized in that , in step a), the given terminal receives a third-party application for accessing the content when it is executed by the given terminal, the third-party application being written in native code and encapsulating the trace block written in executable code with a browser included in the given terminal,
and in that , in step b), the tracking block, when executed by the browser, communicates with the media player to obtain the audio component of the content, the media player being integrated with or accessed by the browser. Method according to claim 1, characterized in that , in step a), the given terminal receives a third-party application for accessing the content when it is executed by the given terminal, the third-party application being written in native code and integrating the trace block also written in native code,
and in that , in step b), the trace block, when executed by the terminal, communicates with the media player to obtain the audio component of the content, the media player being integrated with or accessible by the application third. Method according to any one of claims 1 to 5, characterized in that said unique tracing identifier is a unique identifier of the given terminal, stored by the terminal. Method according to any one of claims 1 to 5, characterized in that said unique tracing identifier is an identifier generated by the tracing block, then stored by the terminal. Method according to any one of claims 1 to 7, characterized in that it is implemented with a group of panellists distributed by households and having a first set of terminals included in said plurality of terminals, each home comprising least one panelist, and in that the given terminal, if it is part of said first set of terminals, also performs the following steps when it is used by at least one panelist of a given household: i) it is registered as a panellist terminal with the audience measurement server; ii) at least once, it transmits to the audience measurement server a declaration message comprising: * the unique tracing identifier; an identifier of the given terminal, or information relating to the type of the given terminal; and * a list of panelist (s) declared. Method according to claim 8, characterized in that in step i): the given terminal transmits to the audience measurement server the identifier and the type of the given terminal; and the audience measurement server associates the identifier and the type of the given terminal with a unique identifier linked to the given focus and a list of possible panelists related to the given focus. Method according to claim 8, characterized in that in step i): the given terminal transmits to the audience measurement server the type of the given terminal; and the audience measurement server generates an identifier of the given terminal, and associates the identifier and the type of the given terminal with a unique identifier linked to the given focus and a list of possible panelists related to the given focus. A method according to any one of claims 8 to 10, characterized in that the given terminal executes a declaration application for performing steps i) and ii), and in that the tracing block transmits the unique tracing identifier to the declaration application. A method according to claim 4 and any one of claims 8 to 10, characterized in that the given terminal executes a declaration application for performing step i), in that the tracing block performs step ii), and in that the unique tracing identifier is a unique identifier of the terminal, which is stored by the given terminal and accessible by the declaration application and the tracing block. A method according to any one of claims 2 to 4 and any one of claims 8 to 10, characterized in that the given terminal executes a declaration application for performing step i), in that the tracing block performs step ii), in that the unique tracing identifier is a unique identifier of the terminal, and in that , in step i), the declaration application launches the browser on a URL containing said unique identifier of the terminal and that when it is executed deposits on the terminal a cookie containing said unique identifier of the terminal, said cookie being accessible by the tracing block when it is executed by the browser. Method according to any one of claims 8 to 10, characterized in that the browser performs step i), during which is deposited on the terminal a cookie containing said unique identifier of the terminal, said cookie being accessible by the block tracing when executed by the browser, in that the unique tracing identifier is said unique identifier of the terminal, and in that the tracing block performs step ii). Process according to any one of Claims 8 to 14, characterized in that it is carried out with: said group of panellists distributed by households and having said first set of terminals included in said plurality of terminals; and - non-panelist users with a second set of terminals
included in said plurality of terminals; and in that the audience measurement server performs a recovery step: first audience measurement data, obtained from a correlation between the tracing messages and the declaration messages transmitted by the terminals of the first set, according to second audience measurement data, obtained from the tracing messages transmitted by the terminals of the second set. A computer program product, comprising program code instructions for implementing the method according to at least one of claims 1 to 15, when said program is run on a computer. A computer-readable and non-transitory storage medium storing a computer program comprising a set of instructions executable by a computer or a processor for carrying out the method according to at least one of claims 1 to 15. Terminal included in a system for measuring an audience of content broadcast in streaming mode or by downloading on a computer network, the system comprising an audience measurement server and a plurality of terminals to which said terminal belongs, characterized in what said terminal comprises: a) means for receiving a tracing block previously associated with the content, or means for associating a tracing block with the content; b) the following means, included in the trace block and activated during the consumption of the content with a multimedia player included in the terminal: means for obtaining from the multimedia player an audio component of the content; means for extracting, from the audio component obtained, at least one digital tattoo and / or at least one digital fingerprint comprising at least one absolute time index; means for transmitting, to the audience measurement server, tracing messages each comprising a unique tracing identifier, stored by the given terminal, and said at least one digital tattoo and / or said at least one digital imprint; and in that the means for receiving or associating a tracing block comprises: Means for receiving at least one web page in which the tracking block is integrated, the at least one web page referencing at least part of the content and being executable with a browser included in the given terminal; or Means for receiving at least one web page referencing at least part of the content and executable with a browser included in the given terminal, and an extension module, included in the given terminal and completing the browser, which integrates the tracing block to said at least one web page, prior to the execution of said at least one web page by the browser; or Means for receiving a third-party application that makes it possible to access the content when it is executed by the given terminal, the third-party application being written in native code and encapsulating the trace block written in executable code with a browser included in the given terminal; or Means for receiving a third-party application making it possible to access the content when it is executed by the given terminal, the third-party application being written in native code and integrating the tracing block also written in native code.

Images