EP3363208A1 - Enrichissement contextuel par reconnaissance audio - Google Patents
Enrichissement contextuel par reconnaissance audioInfo
- Publication number
- EP3363208A1 EP3363208A1 EP16791656.8A EP16791656A EP3363208A1 EP 3363208 A1 EP3363208 A1 EP 3363208A1 EP 16791656 A EP16791656 A EP 16791656A EP 3363208 A1 EP3363208 A1 EP 3363208A1
- Authority
- EP
- European Patent Office
- Prior art keywords
- current
- content
- audiovisual
- signatures
- audiovisual content
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Withdrawn
Links
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N21/00—Selective content distribution, e.g. interactive television or video on demand [VOD]
- H04N21/40—Client devices specifically adapted for the reception of or interaction with content, e.g. set-top-box [STB]; Operations thereof
- H04N21/41—Structure of client; Structure of client peripherals
- H04N21/4104—Peripherals receiving signals from specially adapted client devices
- H04N21/4126—The peripheral being portable, e.g. PDAs or mobile phones
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N21/00—Selective content distribution, e.g. interactive television or video on demand [VOD]
- H04N21/40—Client devices specifically adapted for the reception of or interaction with content, e.g. set-top-box [STB]; Operations thereof
- H04N21/43—Processing of content or additional data, e.g. demultiplexing additional data from a digital video stream; Elementary client operations, e.g. monitoring of home network or synchronising decoder's clock; Client middleware
- H04N21/439—Processing of audio elementary streams
- H04N21/4394—Processing of audio elementary streams involving operations for analysing the audio stream, e.g. detecting features or characteristics in audio streams
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N21/00—Selective content distribution, e.g. interactive television or video on demand [VOD]
- H04N21/20—Servers specifically adapted for the distribution of content, e.g. VOD servers; Operations thereof
- H04N21/27—Server based end-user applications
- H04N21/278—Content descriptor database or directory service for end-user access
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N21/00—Selective content distribution, e.g. interactive television or video on demand [VOD]
- H04N21/40—Client devices specifically adapted for the reception of or interaction with content, e.g. set-top-box [STB]; Operations thereof
- H04N21/45—Management operations performed by the client for facilitating the reception of or the interaction with the content or administrating data related to the end-user or to the client device itself, e.g. learning user preferences for recommending movies, resolving scheduling conflicts
- H04N21/462—Content or additional data management, e.g. creating a master electronic program guide from data received from the Internet and a Head-end, controlling the complexity of a video stream by scaling the resolution or bit-rate based on the client capabilities
- H04N21/4622—Retrieving content or additional data from different sources, e.g. from a broadcast channel and the Internet
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N21/00—Selective content distribution, e.g. interactive television or video on demand [VOD]
- H04N21/80—Generation or processing of content or additional data by content creator independently of the distribution process; Content per se
- H04N21/85—Assembly of content; Generation of multimedia applications
- H04N21/858—Linking data to content, e.g. by linking an URL to a video object, by creating a hotspot
- H04N21/8586—Linking data to content, e.g. by linking an URL to a video object, by creating a hotspot by using a URL
Definitions
- the present invention relates to the field of contextual enrichment of audiovisual content, and in particular, but not exclusively, content broadcast by television channels.
- It relates more specifically contextual enrichment implementing auditory recognition of audiovisual content displayed on a user rendering device (a TV for example), by a user terminal (such as a smartphone or "smartphone” for example). ) without requiring a connection between the device and the terminal.
- a user rendering device a TV for example
- a user terminal such as a smartphone or "smartphone” for example
- Auditory recognition techniques are known for the purpose of recognizing extracts of audio contents such as musical tracks.
- the enrichment then consists, after recognition of the extract by using a database storing all the musical tracks (or signatures thereof) recognizable by the service, to return to the user the name of the artist, the music track, and possibly the album from which it is extracted.
- the present invention improves the situation.
- a first aspect of the invention relates to a method for enriching audiovisual content, the method comprising the following steps implemented in a first service platform, the service platform comprising a local database storing associations between audiovisual content signatures and sources of audiovisual content on the one hand and associations between sources of audiovisual content and contextual content on the other hand:
- the request upon receipt of a user request from a first user terminal, the request comprising a candidate signature, identifying a source of audiovisual content by comparison between the candidate signature and the signatures stored in the local database;
- the present invention provides a dynamic update of both contextual content enriching broadcast audiovisual content and both signatures for the recognition of audiovisual content being broadcast. This makes it possible to apply the enrichment to any broadcasted content, even when it is not known in advance (as may be the case when broadcasting television programs).
- the audiovisual content considered which may be a film, a television program, a radio program, a music video, an advertisement, etc.
- Contextual content is any informative data relating to audiovisual content, and can cover any text, audio, video, photo, etc.
- the set of current audiovisual content signatures respectively associated with the audiovisual source identifiers may be received from a broadcast manager and the method may further include the following steps implemented by the broadcast manager:
- Generating and distributing common signatures is thus centralized, which reduces the complexity and software resources of service platforms.
- the responsiveness of service platforms to compare signatures and extract contextual content is thus improved.
- the current signatures may have a duration D1
- the signature generator may store the set of current signatures of audiovisual contents respectively associated with the identifiers of audiovisual content sources, and the method may furthermore comprise the following steps implemented by the signature generator:
- update of the current signature given by deleting a final period of duration D2 of the current signature given and adding at the beginning of the current signature given the signature extract of duration D2 corresponding to the current audio extract associated with the same audiovisual content source identifier as the current signature given.
- Such a dynamic sliding window generation of signatures makes it possible to ensure that an updated current signature is constantly available relative to the audiovisual content that is being broadcast on a given audiovisual content source.
- the signature generator is dedicated to the generation of signatures, and not to the generation of associated contextual contents, the complexity of the calculations performed is reduced and the reactivity of the generation of signatures is improved.
- the current audio extracts of duration D2 can be received continuously at the end of each period of duration D2.
- the signature generation is carried out continuously and the enrichment service is thus accessible at any time for the user.
- the set of current contextual contents respectively associated with audiovisual content source identifiers can be received from a notification manager, the method further comprising the following steps implemented by the manager notifications:
- broadcasting to a set of service platforms, comprising at least said first service platform, a set comprising at least the current contextual content associated with the identifier of the given audio-visual content source.
- the detection module can store a set of contextual contents, and the method can furthermore comprise the following steps implemented by the detection module:
- the audio-visual excerpt can be an audio extract acquired by microphone, a video extract without sound, or a video and audio extract.
- this embodiment provides a relevant selection of contextual content.
- the detection module is dedicated to the extraction of contextual contents, and not the generation of signatures, the complexity of the calculations performed is reduced and the responsiveness of the extraction of contextual content is improved.
- the detection module may store a set of contextual contents in association with respective reference signatures, further comprising the following steps implemented by the detection module:
- Such a variant makes it possible to pool the current signatures generated by the signature generator between the detection module and the broadcast manager.
- the identifiers of the sources of audiovisual content may be ordered according to a popularity criterion and the identification of a source of audiovisual content may comprise successive comparisons between the candidate signature and the signatures stored in the order of identifiers of the audiovisual content sources associated with them respectively.
- Such an order makes it possible to reduce, on average, the number of comparisons to be made before detecting a correspondence between the candidate signature and a stored signature, which reduces the complexity of the calculations and improves the responsiveness associated with the enrichment of broadcast audiovisual contents. .
- the first user terminal may implement the following steps:
- the extracted contextual content can be transmitted to the first user terminal in association with the identifier of the audiovisual content source identified, the request may furthermore comprise an identifier of the last source of audiovisual content identified and the identification of the A source of audiovisual content may include successive comparisons between the candidate signature and the stored signatures starting with the stored signature in association with the identifier of the last identified audiovisual content source.
- the last source of audiovisual content also makes it possible to reduce, on average, the number of comparisons to be made in identifying the source of audiovisual content. Indeed, it is likely that the user has not changed the source of audiovisual content between two successive requests.
- the identification of the source of audiovisual content may comprise successive comparisons between the candidate signature and the stored signatures, starting with the stored signature in association with the identifier of the last source of audiovisual content identified and then according to the order audiovisual content sources associated with them respectively
- the candidate signature may have a duration less than a duration of the signatures stored in the local database.
- this embodiment makes it possible to ensure that the candidate signature is included in one of the stored signatures, regardless of the technology for transporting the audiovisual content.
- a second aspect of the invention relates to a computer program comprising instructions for the implementation of the method according to the first aspect of the invention, when this program is executed by a processor.
- a third aspect of the invention relates to a service platform for enriching audiovisual content, comprising a local database storing associations between audiovisual content signatures and audiovisual content sources on the one hand and associations between sources of audiovisual content, the service platform further comprising a reception unit and a processor configured for implementing the following steps:
- said request upon reception by the receiving unit of a user request from a first user terminal, said request comprising a candidate signature, identifying a source of audiovisual content by comparison between the candidate signature and the signatures stored in the local database ;
- a fourth aspect of the invention relates to a system comprising a service platform according to the third aspect of the invention, a broadcast manager configured to transmit to the service platform the set of current signatures of audiovisual content respectively associated with the service identifiers. audiovisual content sources, and a notification manager configured to transmit to the service platform said least current contextual content associated with an audiovisual content source identifier.
- FIG. 1 shows a system according to one embodiment of the invention
- FIG. 2 is a diagram showing the steps of a method according to one embodiment of the invention.
- Figure 3 shows a service platform according to one embodiment of the invention
- FIG. 4 shows a broadcast manager according to one embodiment of the invention
- FIG. 5 illustrates a notification manager according to one embodiment of the invention
- FIG. 6 presents a first user terminal according to one embodiment of the invention
- FIG. 7 illustrates a signature generator according to one embodiment of the invention
- FIG. 8 shows a detection module according to one embodiment of the invention.
- FIG. 9 illustrates the generation of a current signature and a candidate signature as a function of time, according to certain embodiments of the invention.
- Figure 1 shows a system according to one embodiment of the invention.
- the system comprises a first user terminal 10 and a second user terminal 11.
- the first user terminal 10 may be a smartphone-type mobile phone, a laptop, a desktop computer, a tablet touch, or more generally any user terminal allowing access to a network, such as an Internet type network for example.
- the first user terminal 10 can access an access point 12 via a wired interface (Ethernet cable for example) or wireless (Wi-fi, Bluetooth, etc.).
- the first user terminal 10 is able to acquire audio data (audio content) from the second user terminal 11.
- the first user terminal 10 may be equipped with at least one microphone. This audio data can be acquired continuously, or over a period of time, on activation of the user for example, via the use of a user interface (touch screen, keyboard, mouse, etc.).
- the second user terminal 11 may be a terminal capable of receiving audiovisual contents from one or more sources of audiovisual content (television channels, radio stations, Netflix-type Internet channels, for example) and reproducing at least the audio component of the contents. audiovisual.
- the second user terminal 1 1 may be a TV or a laptop or desktop.
- the example of a television receiving television channels is considered for illustrative purposes.
- the first user terminal is able to acquire audio content reproduced by the TV 1 1 (from the audiovisual content received on the current television channel) and to generate a candidate signature based on the audio content.
- signature refers to any set of characteristics determined from audio content. The determination of such signatures is well known and is not described in more detail in the following.
- the candidate signature may have a duration D3 equal to, for example, 10 seconds.
- the first user terminal 10 may for example generate a signature of 10 seconds every 10 seconds, and transmitting each time the generated signature to a first service platform 13.1 via the access point 12, as detailed in the following.
- Candidate signature generation variants are described in the following.
- a plurality of service platforms 13.1, 13.2 ... 13.n, including the first service platform 13.1 is included in the system according to the invention.
- Each of the service platforms may for example cover a geographical area of its own, or may be dedicated to a group of users of its own.
- Each service platform 13.1 -13. n is able to access, via an Internet-type network for example, to a broadcast manager 14 connected to a signature generator 15 and to a notification manager 16 connected to a detection module 17.
- Each service platform 13.1 -13. n may include a local database storing associations between audiovisual content signatures and audiovisual content sources on the one hand and associations between audiovisual content sources and contextual content on the other hand. These associations will be better understood from the description below.
- the signature generator 15 is able to generate audio signatures in association with sources of audiovisual content in parallel.
- a signature generator can be used for each television channel.
- each television channel (more generally each source of audiovisual content) is identified by an identifier.
- the signature generator 15 thus stores a current signature of a duration D1, D1 being for example equal to 30 seconds, in association with each television channel identifier.
- the current signature of duration D1 associated with the identifier of a given television channel is thus representative of the last period of duration D1 of the audio stream broadcast on the given television channel.
- the signature generator 15 can receive in parallel the audio streams coming from all the television channels, and continuously extract a current audio extract (the last period of duration D2 of the audio stream) in order to generate continuously (all periods D2) signature extracts of duration D2, from the current audio extract of duration D2 of the audio stream, D2 may be equal to one second for example.
- Each current audio extract (and the corresponding signature extract) is associated with the television channel identifier from which it is derived. Then, for each given current signature, the current signature given is updated by deleting a final period of duration D2 of the given current signature and adding, at the beginning of the given current signature, the generated signature extract corresponding to the current signature.
- current audio extract associated with the same TV channel identifier as the current signature given.
- the current signatures are updated by sliding window, which makes it possible to maintain with high granularity (for example 1 second) signatures representative of the last period of duration D1 (for example 30 seconds) broadcast on each television channel. .
- the set of current signatures thus updated is then transmitted, preferably at the end of each period of duration D2 (every second for example) to the broadcast manager 14.
- the broadcast manager 14, preferably to the after each period of duration D2, can thus broadcast to all service platforms 13.1 -13.
- the set of current signatures so that they store the set of current signatures. No restriction is attached to the distribution of the set of current signatures (of the "multicast” or "broadcast” type for example).
- the broadcast manager 14 may further be in charge of managing the number of authorized user connections per service platform 13.1 - 13.n.
- each service platform 13.1 -13. n can update its local database by modifying the stored signatures based on the received current signatures. For example, previously stored signatures are all deleted and replaced by the current signatures received.
- the service platform may store the last N signatures associated with a given TV channel identifier, where N is an integer greater than 1.
- N is an integer greater than 1.
- Contextual content means any information, any data, of any format whatsoever (audio, text, URL link, video, photo) related to a main content (audiovisual content broadcast on television channels).
- a main content audiovisual content broadcast on television channels.
- contextual content related to the advertisement may be a URL link allowing a redirection to the merchant site to buy the product.
- contextual content related to the film may be a subtitle file, a summary of the film, a URL link to an article criticizing the film, a video summarizing the film, a photo of the movie poster, etc.
- the detection module 17 is able, upon acquisition of an audiovisual extract of a stream broadcast by a given television channel, to identify audiovisual content being broadcast and to extract contextual content related to the audiovisual content. currently being broadcast.
- the audio-visual clip may be an audio clip, a video clip, or an excerpt that includes video data and audio data.
- the audiovisual extract comprises at least video data, which facilitates the identification of the audiovisual content being broadcast.
- Video and / or audio identification algorithms are well known and are not detailed in what follows.
- audiovisual content such as a movie may be associated with director, actor, or other metadata, and all contextual content associated with the same metadata (or some of that metadata) may be retrieved. , or one of them can be selected.
- the detection module 17 When current contextual content is extracted according to the audiovisual content being broadcast on the given television channel, the detection module 17 transmits the current contextual content in association with the identifier of the given television channel, to the notification manager 16. In addition, the detection module 17 can determine current contextual content for each of the television channels, and transmit to the notification manager 16 all of the current contextual content respectively associated with TV channel identifiers.
- the current contextual content to be associated with a television channel identifier may be imposed by a server of the television channel (or by the server of an advertiser) able to communicate with the notification manager 16.
- the detection module 17 is not used and the server of the television channel can instruct the notification server 17 to broadcast to the service platforms 13.1 -13.
- n contextual content to associate with the identifier of the television channel may be imposed by a server of the television channel (or by the server of an advertiser) able to communicate with the notification manager 16.
- the notification manager 16 may broadcast a set comprising at least the current contextual content associated with the identifier from the television channel given to service platforms 13.1 -13. not.
- each service platform 13.1 -13. n updates its local database by modifying the contextual contents according to the set of at least one current contextual content received. For example, the last contextual content associated with the identifier of the given television channel is deleted and replaced by the current contextual content.
- the current contextual content is also associated with a period of validity, and, at the expiration of the period of validity, each service platform 13.1 -13. n can delete the current contextual content stored in association with the identifier of the given television channel.
- Each service platform 13.1 -13. n thus has dynamically updated associations between audiovisual content signatures and sources of audiovisual content on the one hand and between audiovisual content sources and contextual content on the other hand.
- the user terminal 10 can transmit to the first service platform 13.1 a request comprising the candidate signature.
- the sending of the request, and the preliminary determination of the candidate signature, can be triggered by the launching of a dedicated application on the first user terminal 10.
- the service platform 13.1 On receiving the request comprising the candidate signature, the service platform 13.1 compares the candidate signature (of duration D3) with the signatures stored in its local database (of duration D1 greater than D3). In the case where a match is detected between the candidate signature and a given stored signature, the associated audiovisual content source in the local database at the given stored signature is identified. The contextual content associated with the source of audiovisual content identified is thus extracted from the local database by the service platform 13.1 and transmitted to the first user terminal 10.
- the user thus has on his first user terminal 10 contextual content enriching the audiovisual content displayed on the second user terminal 11.
- the duration D1 is preferably greater than D3. Indeed, depending on the technology of transport of the audiovisual stream displayed on the second User terminal 1 1, the transport time varies (for example terrestrial broadcasting and satellite broadcasting involves different transport times). In order to ensure that, whatever the transport technology used for the audiovisual stream, the candidate signature can be included in one of the signatures stored in the local database of the service platform 13.1, the duration D1 is longer. large than D3 (for example a multiple of D3).
- the candidate signature generated by the first user terminal 10 can be updated at the end of each period D2 (every second for example, or at the same frequency as the update of the signatures in the signature generator 15).
- the first user terminal 10 can acquire every second an audio extract from the second user terminal 1 1 and determine a signature extract on this basis.
- the last generated candidate signature is then modified by deleting the final period of duration D2 of the signature (the oldest second) and inserting the signature extract at the beginning of the candidate signature. This makes it possible to ensure that the candidate signature of duration D3 (which can be 10 seconds as detailed above) is dynamically updated at each period D2.
- the candidate signature thus updated can be transmitted to the service platform every m * D2 periods, where m is an integer greater than or equal to 1.
- the sending of each request comprising the updated candidate signature is at the initiative of the user.
- Figure 2 is an exchange diagram illustrating the steps implemented by the entities of the system.
- the signature generator 15 acquires a current audio extract of duration D2 for each audiovisual content source identifier.
- the signature generator generates for each current audio extract a signature extract of duration D2.
- the signature generator 15 can update the current signature given by deleting a final period of duration D2 of the given current signature and adding at the beginning of the given current signature the signature extract of duration D2 corresponding to the current audio extract associated with the same audiovisual content source identifier as said given current signature.
- the signature generator delays for a period D2 before executing steps 200 to 202 again.
- the signature generator transmits to the broadcast manager 14 the set of current signatures of audiovisual contents respectively associated with the identifiers of audiovisual content sources.
- the set of current signatures respectively associated with the identifiers of audiovisual content sources can also be transmitted to the detection module 17.
- the broadcast manager 14 can check the availability of the service platforms 13.1 -13. not.
- the broadcast manager 14 broadcasts to all of the service platforms 13.1 -13. n (or at least at the first service platform 13.1) the set of current signatures of audiovisual content respectively associated with the identifiers of audiovisual content sources.
- the service platform 13.1 updates its local database by modifying the stored signatures according to the signatures. received, as detailed above.
- the detection module 17 acquires an audiovisual extract corresponding to at least the given audiovisual content source (see description below). before, with reference to Figure 1).
- the detection module 17 extracts a current contextual content from the set of contextual contents stored in the detection module 17, as a function of the audio-visual extract corresponding to the given audio-visual content source. 7
- step 209 can be implemented by considering each current signature as a candidate signature for a search in a database of reference signatures associated with contextual contents. In this case, the preliminary step 208 is not implemented.
- This embodiment makes it possible to pool the current signatures generated by the signature generator 15 between the detection module 17 and the broadcast manager 14.
- the detection module 17 transmits to the notification manager 16 a set comprising at least the current contextual content in association with the identifier of the given audio-visual content source.
- the detection module can delay during a period D4, before repeating steps 208 to 210.
- the notification manager 16 stores the set comprising at least the current contextual content in association with the identifier of the given audiovisual content source, received from the detection module 17.
- the notification manager 16 can receive directly from a server of a television channel contextual content to associate with an identifier of the television channel.
- the notification manager 16 broadcasts to the set of service platforms 13.1 -13. n (or at least at the first service platform 13.1) a set comprising at least the current contextual content associated with the given audiovisual content source identifier.
- a step 214 upon reception of the set comprising at least the current contextual content associated with an audiovisual content source identifier, update of the local database by modifying the contextual contents as a function of the set comprising the less current contextual content.
- an application is launched at an optional step 215 on the first user terminal 10, the application being dedicated to the contextual enrichment according to the invention .
- the first user terminal 10 can acquire, by a microphone, audio content from the second user terminal 11.
- the audio content may be content of duration D3 for generating a candidate signature, or content of duration D2 for updating a previously generated candidate signature.
- a candidate signature is generated according to the audio content acquired by the first user terminal 10.
- a request comprising the candidate signature is generated by the first user terminal 10.
- the generated request is transmitted to the first service platform 13.1 by the first user terminal 10.
- the first user terminal 10 can delay during a period D2 before repeating the steps 216 and 217 making it possible to generate a new candidate signature.
- a request is not necessarily generated, as described above, since the transmission of a request may occur preferentially for all m * D2 periods, where m is an integer.
- the first service platform 13.1 On receiving the user request from the first user terminal 20, the first service platform 13.1 identifies, at a step 221, a source of audiovisual content by comparison between the candidate signature and the signatures stored in the local database.
- the first service platform 13.1 extracts from its local database the contextual content associated with the source of audiovisual content identified.
- the extracted contextual content is transmitted to the first user terminal 10, which thus has a contextual content enabling the enrichment of the audiovisual content displayed on the second user terminal 11.
- the contextual content may be transmitted with the identifier of the audiovisual content source identified. So the first user terminal 10 also has the identifier of the audiovisual content source that broadcasts the audiovisual content displayed on the second user terminal 11, and can integrate this identifier when transmitting a new request.
- This enables the first service platform 13.1, during the signature comparison step 221, to start by comparing the candidate signature with the stored signature in association with the audiovisual content source identified in the request. Indeed, the probability that the user has not changed the source is high, and software resources of the service platform are thus saved (the average number of comparisons to be performed is reduced).
- the identifiers of the audiovisual content sources are ordered in the service platforms 13.1 -13. n according to a popularity criterion (from the most consulted to the least consulted). There are no restrictions on the popularity criteria: for example, it may be the number of views of the source of audiovisual content for a given time slot, or a ranking by the user himself.
- the identification of a source of audiovisual content comprises successive comparisons between the candidate signature and the stored signatures according to the order of the identifiers of the audiovisual content sources which are respectively associated with the audiovisual content sources. signatures.
- the popularity criterion can moreover be used in combination with the audiovisual content source identifier inserted into the user's request: the identification of the audiovisual content source comprises successive comparisons between the candidate signature and the stored signatures starting with the stored signature in association with the identifier of the last source of audiovisual content identified in the request, then according to the order of the audiovisual content source identifiers that are respectively associated with the signatures.
- FIG. 3 illustrates a first service platform 13.1 according to one embodiment of the invention.
- the first service platform 13.1 includes a RAM 303 and a processor 302 for storing instructions for implementing steps 207, 214, 221, 222 and 223 of the method described above.
- the service platform 13.1 may further include a local database 304 for storing the associations between audiovisual content signatures and audiovisual content sources on the one hand and the associations between audiovisual content sources and contextual content of somewhere else.
- the first service platform 13.1 also comprises an input interface 301 intended to receive the set of current signatures of audiovisual contents respectively associated with audio-visual content source identifiers of the broadcast manager 14, the set of at least current contextual content associated with an audiovisual content source identifier of the notification manager 16 and the request of the first device of the user.
- the first service platform 13.1 also comprises an output interface 305 able to transmit to the first user terminal 10 the contextual content extracted from the local database 304.
- FIG. 4 illustrates a broadcast manager 14 according to one embodiment of the invention.
- the broadcast manager 14 comprises a random access memory 403 and a processor 402 for storing instructions enabling the implementation of step 205 of the method described above.
- the broadcast manager 14 may further include a local database 404 for storing the associations between current audiovisual content signatures and audiovisual content sources.
- the broadcast manager 14 further comprises an input interface 401 intended to receive the set of current signatures of audiovisual contents respectively associated with audio-visual content source identifiers of the signature generator 15 and an output interface 405 able to broadcast to service platforms 13.1 -13. n the set of current signatures of audiovisual contents respectively associated with identifiers of audiovisual content sources.
- Figure 5 illustrates a notification manager 16 according to an embodiment of the invention.
- the notification manager 16 comprises a random access memory 503 and a processor 502 for storing instructions enabling the implementation of step 212 of the method described above.
- the notification manager 16 may further include a local database 504 for storing associations between audiovisual content sources and contextual contents.
- the notification manager 16 furthermore comprises an input interface 501 intended to receive the set of contextual contents respectively associated with audio-visual content source identifiers of the detection module 17 (or directly from a television channel server or from the an advertiser's server) and a 505 output interface capable of broadcasting to the service platforms 13.1 -13. n the set of contextual contents respectively associated with audiovisual content source identifiers.
- FIG. 6 illustrates a first user terminal 10 according to one embodiment of the invention.
- the first user terminal 10 comprises a random access memory 604 and a processor 603 for storing instructions for carrying out the steps 215, 216, 217, 218 and 220 of the method described above.
- the first user terminal 10 may further include a local database 606 for storing the identifier of the last source of audiovisual content received from the first service platform 13.1 and for storing the generated candidate signatures.
- the local database 606 can also store the application dedicated to the enrichment of audiovisual content according to the invention.
- the first user terminal 10 further comprises a microphone 601 for acquiring audio content from the second user terminal 10 and a user interface 602 for receiving commands from the user (launching the dedicated application, reading the received contextual content, etc. ).
- the audio content may be from an audio stream acquired directly by wire from the second user terminal 20 (DLNA feature for example), and in this case, the microphone 601 is optional.
- the first user terminal 10 further includes a screen 605 for displaying a visual component of the contextual content and a speaker 608 for rendering the audio component of the contextual content.
- the first user terminal 10 further comprises an input interface 601 intended to receive the contextual content, optionally accompanied by the identifier of the source of audiovisual content given, from the first service platform 13.1, and an output interface 607 capable of transmit the generated request to the first service platform 13.1.
- Figure 7 illustrates a signature generator 15 according to one embodiment of the invention.
- the signature generator 15 comprises a random access memory 703 and a processor 702 for storing instructions enabling the implementation of steps 200 to 203 of the method described above.
- the signature generator 15 may further include a local database 704 for storing the current audiovisual content signatures in association with the audiovisual content sources.
- the signature generator 15 furthermore comprises an input interface 501 intended to receive the audiovisual streams of the various audio-visual content sources and an output interface 705 capable of transmitting to the broadcast manager 14 (and optionally to the detection module 1 7) the associations between current signatures and identifiers of audiovisual content sources.
- FIG. 8 illustrates a detection module 17 according to one embodiment of the invention.
- the detection module 17 comprises a random access memory 803 and a processor 802 for storing instructions for carrying out the steps 208, 209 and 21 1 of the method described above.
- the detection module 17 may further include a local database 804 for storing the contextual contents for extracting current contextual content.
- each contextual content is associated with a reference signature
- the processor 802 is able to search the current signatures received from the signature generator 15, among the reference signatures, in order to extract contextual content to be associated with a channel identifier.
- the detection module 17 furthermore comprises an interface 801 that can be a module for acquiring an audiovisual extract (camera, microphone, for example) or that can be a network interface capable of receiving the current signatures and the identifiers of content sources.
- the detection module includes an output interface 805 capable of transmitting to the notification manager 16 the set of at least one current contextual content associated with the identifier of the given audio-visual content source.
- FIG. 9 illustrates the generation of a current signature 900 and a candidate signature 903 as a function of time, according to some embodiments of the invention.
- the signature generator 15 has a current signature 900 which has been transmitted to the service platforms 13.1 -13. not.
- the current signature 900 has a duration D1, which can be equal to 30 seconds for example.
- a current audio extract is received with an audiovisual content source identifier and the signature generator generates a signature extract 901 of duration D2.
- the current signature 900 is thus updated by erasing a final period 902 of duration D2 and adding to the top of the current signature (between ti and t 2) given to extract a signature 901 of length D2.
- the duration D2 may be equal to one second.
- the first service platform 13.1 can receive a candidate signature 903 of duration D3 of the first user terminal 10.
- the duration D3 may be equal to 10 seconds.
- the current signature received at the instant ti may have a delay ⁇ indicating that the candidate signature 903 corresponds to an audio extract of the stream retarded delay ⁇ compared to current signatures 900.
- the fact of having a duration D1 that is substantially greater (for example a multiple) than the duration D3 allows a comparison to be made between the candidate signature 903 and the current signatures 900 whatever the delay ⁇ between 0 and (D1-D3).
- the candidate signature 903 can be updated in the same way as the current signatures 900.
Abstract
Description
Claims
Applications Claiming Priority (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
FR1559688A FR3042369B1 (fr) | 2015-10-12 | 2015-10-12 | Enrichissement contextuel par reconnaissance audio |
PCT/FR2016/052599 WO2017064400A1 (fr) | 2015-10-12 | 2016-10-07 | Enrichissement contextuel par reconnaissance audio |
Publications (1)
Publication Number | Publication Date |
---|---|
EP3363208A1 true EP3363208A1 (fr) | 2018-08-22 |
Family
ID=55299612
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
EP16791656.8A Withdrawn EP3363208A1 (fr) | 2015-10-12 | 2016-10-07 | Enrichissement contextuel par reconnaissance audio |
Country Status (3)
Country | Link |
---|---|
EP (1) | EP3363208A1 (fr) |
FR (1) | FR3042369B1 (fr) |
WO (1) | WO2017064400A1 (fr) |
Families Citing this family (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
FR3101451B1 (fr) * | 2019-09-26 | 2021-10-01 | Tdf | Procédé d’identification de flux audio provenant d’une pluralité de sources, système, récepteur et programme associé au procédé |
Family Cites Families (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
MXPA04004645A (es) * | 2001-11-16 | 2004-08-12 | Koninkl Philips Electronics Nv | Metodo, cliente y servidor para actualizar base de datos de huellas digitales. |
US7707221B1 (en) * | 2002-04-03 | 2010-04-27 | Yahoo! Inc. | Associating and linking compact disc metadata |
FR2927183B1 (fr) * | 2008-01-31 | 2010-02-26 | Alcatel Lucent | Procede de generation de donnees permettant la recherche de complements de contenus, systeme, terminal et serveur pour la mise en oeuvre du procede |
US20100057527A1 (en) * | 2008-08-29 | 2010-03-04 | Disney Enterprises, Inc. | System and method for personalized action based on a comparison of delivered content with a content fingerprint database |
FR2983672A1 (fr) * | 2011-12-06 | 2013-06-07 | France Telecom | Notification relative a des contenus diffuses |
FR2997597B1 (fr) * | 2012-10-30 | 2015-12-18 | Tdf | Procede et module de basculement d'un premier programme vers un deuxieme programme, procede de diffusion, tete de reseau, programme d'ordinateur et medium de stockage correspondants. |
US9542488B2 (en) * | 2013-08-02 | 2017-01-10 | Google Inc. | Associating audio tracks with video content |
FR3016720B1 (fr) * | 2014-01-20 | 2016-02-05 | Tdf | Procede et systeme de delivrance de coupons de reduction et de gestion desdits coupons. |
-
2015
- 2015-10-12 FR FR1559688A patent/FR3042369B1/fr active Active
-
2016
- 2016-10-07 WO PCT/FR2016/052599 patent/WO2017064400A1/fr active Application Filing
- 2016-10-07 EP EP16791656.8A patent/EP3363208A1/fr not_active Withdrawn
Also Published As
Publication number | Publication date |
---|---|
FR3042369B1 (fr) | 2017-12-08 |
WO2017064400A1 (fr) | 2017-04-20 |
FR3042369A1 (fr) | 2017-04-14 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
EP2811749B1 (fr) | Synchronisation de contenus multimédia sur deuxième écran | |
JP5828501B2 (ja) | 番組コンテキストに基づくモバイルコンテンツの提示 | |
US9734153B2 (en) | Managing related digital content | |
US9703781B2 (en) | Managing related digital content | |
EP3646548B1 (fr) | Procédé de transmission d'un contenu audio interrompu dans un récepteur hybride, système, récepteur et programme associé au procédé | |
WO2012131258A1 (fr) | Procede d'acces a un service, notamment un portail web, par un terminal de restitution d'un flux multimedia | |
FR3028631A1 (fr) | Procede de classement d'un contenu et recommandation de contenu dans un guide electronique des programmes | |
EP3363208A1 (fr) | Enrichissement contextuel par reconnaissance audio | |
US8234158B1 (en) | Analyzing text streams for cue points of advertisements in a media stream | |
US20150020125A1 (en) | System and method for providing interactive or additional media | |
EP4161081A1 (fr) | Procédé de génération d'une chaîne de télévision personnalisée pour un utilisateur d'un terminal configuré pour accéder à au moins un service de diffusion de contenus audiovisuels, dispositif, équipement de service, système et programme d'ordinateur correspondants. | |
WO2017158274A1 (fr) | Acquisition d'extraits d'un flux multimédia sur un terminal | |
FR2927183A1 (fr) | Procede de generation de donnees permettant la recherche de complements de contenus, systeme, terminal et serveur pour la mise en oeuvre du procede | |
FR3005386A1 (fr) | Procede et dispositif de fourniture d’une partie deja diffusee d’un flux multimedia, terminal utilisateur, programme d’ordinateur et medium de stockage correspondants | |
WO2001091344A2 (fr) | Procede de diffusion d'elements d'information multimedia | |
FR2917553A1 (fr) | Procede de diffusion d'un element complementaire, serveur et terminal correspondants | |
WO2009112556A1 (fr) | Procede de restitution d'au moins un contenu multimedia personnalise, terminal et programme d'ordinateur correspondants | |
FR2956787A1 (fr) | Procede et serveur pour detecter un programme video recu par un usager | |
FR3009103A1 (fr) | Generation de listes de reproduction de contenus personnalisees | |
FR2983605A1 (fr) | Dispositif et procede de selection et de mise a jour du profil d'un utilisateur. | |
EP2915330A1 (fr) | Procédé et module de basculement d'un premier programme vers un deuxième programme, procédé de diffusion, tête de réseau, programme d'ordinateur et medium de stockage correspondants | |
FR3032584A1 (fr) | Acces ameliore a un contenu numerique | |
WO2020216926A1 (fr) | Commande d'un service utilisant le traitement d'un flux comprenant des donnees multimedias | |
EP4254968A1 (fr) | Procédé de génération d'une chaîne de télévision virtuelle pour un utilisateur d' au moins un service de diffusion de contenus audiovisuels, dispositif de génération, équipement de service et programme d ordinateur correspondants | |
WO2016156386A1 (fr) | Système de diffusion de contenus audio et/ou vidéo par un réseau wifi local, et appareils mettant en œuvre le procédé |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
STAA | Information on the status of an ep patent application or granted ep patent |
Free format text: STATUS: UNKNOWN |
|
STAA | Information on the status of an ep patent application or granted ep patent |
Free format text: STATUS: THE INTERNATIONAL PUBLICATION HAS BEEN MADE |
|
PUAI | Public reference made under article 153(3) epc to a published international application that has entered the european phase |
Free format text: ORIGINAL CODE: 0009012 |
|
STAA | Information on the status of an ep patent application or granted ep patent |
Free format text: STATUS: REQUEST FOR EXAMINATION WAS MADE |
|
17P | Request for examination filed |
Effective date: 20180410 |
|
AK | Designated contracting states |
Kind code of ref document: A1 Designated state(s): AL AT BE BG CH CY CZ DE DK EE ES FI FR GB GR HR HU IE IS IT LI LT LU LV MC MK MT NL NO PL PT RO RS SE SI SK SM TR |
|
AX | Request for extension of the european patent |
Extension state: BA ME |
|
DAV | Request for validation of the european patent (deleted) | ||
DAX | Request for extension of the european patent (deleted) | ||
STAA | Information on the status of an ep patent application or granted ep patent |
Free format text: STATUS: EXAMINATION IS IN PROGRESS |
|
17Q | First examination report despatched |
Effective date: 20210204 |
|
STAA | Information on the status of an ep patent application or granted ep patent |
Free format text: STATUS: EXAMINATION IS IN PROGRESS |
|
GRAP | Despatch of communication of intention to grant a patent |
Free format text: ORIGINAL CODE: EPIDOSNIGR1 |
|
STAA | Information on the status of an ep patent application or granted ep patent |
Free format text: STATUS: GRANT OF PATENT IS INTENDED |
|
INTG | Intention to grant announced |
Effective date: 20220315 |
|
STAA | Information on the status of an ep patent application or granted ep patent |
Free format text: STATUS: THE APPLICATION IS DEEMED TO BE WITHDRAWN |
|
18D | Application deemed to be withdrawn |
Effective date: 20220726 |