WO2014067968A1

WO2014067968A1 - Method and module for switching from a first programme to a second programme, and corresponding broadcasting method, headend, computer program and storage medium

Info

Publication number: WO2014067968A1
Application number: PCT/EP2013/072649
Authority: WO
Inventors: Mouhtar Haidar; Laurent Cannieux; David Vincent
Original assignee: Tdf
Priority date: 2012-10-30
Filing date: 2013-10-29
Publication date: 2014-05-08
Also published as: FR2997597B1; EP2915330A1; FR2997597A1

Abstract

The invention concerns a method for switching from a first programme to a second programme in a broadcast network implementing a headend and at least one receiver. According to the invention, such a method implements the following steps, in at least one of the receivers: - detection (31), in a signal carrying the first programme, of a specific audio fragment intended to effect a transition between two programmes; - replacement (32), in the signal, of the first programme by the second programme, in case of positive detection.

Description

Method and module for switching from a first program to a second program, broadcast method, headend, computer program and corresponding storage medium.

1. Field of the invention

The field of the invention is that of the transmission and broadcasting of content, especially radio or television data.

More specifically, the invention relates to the "switchover" or "unhooking" of programs in a broadcast network, making it possible to replace, at a user's broadcast site or terminal, a first program with a program. second program in a data stream.

For example, the first program is a national program, broadcast by a national network head. The second program is a local or regional program, a drop-down banner, an alert, a local advertisement, etc., to be inserted as a replacement for the national program in the data stream.

The invention applies in particular, but not exclusively, to a data stream that complies with the DVB-T or DVB-T2 ("Digital Video Broadcasting - Terrestrial") broadcast standards for "digital terrestrial video broadcasting" or an audio data stream.

2. Prior Art

The problem of stalling is well known in the field of digital terrestrial television broadcasting, also called TNT, for example implementing the DVB-T or DVB-T2 standard.

In digital television, stalling makes it possible to substitute for a main program, such as a television program covered nationally on France 3 (registered trademark), a secondary program, such as a television program with regional coverage on France 3. It thus passes from a broadcast by a national network head, broadcasting the main program, to a regional broadcast site, broadcasting the secondary program.

FIG. 1 illustrates an example of a digital broadcasting network, comprising a Tx headend 11 feeding one or more Rx1 121, Rx2 122, Rx3 123 broadcast sites, located at distinct geographical sites. Each broadcast site Rx1 121, Rx2 122, Rx3 123 comprises a receiver and a repeater, which can optionally be combined. The term "broadcast site" is thus understood to mean an equipment or system comprising a receiver followed by a retransmitter or integrated in a retransmitter.

Conventionally, the headend Tx 11 generates a digital signal, also called signal or data stream, transmitted by a network adapter and conveyed to Rxl broadcast sites 121, Rx2 122, Rx3 123 via a distribution network 13, for example a satellite distribution network. Such a signal is for example of type M PEG2-TS (in English "Motion Picture Expert Group 2 - Transport Stream"), and carries a main program.

One or more of the broadcast sites Rx1 121, Rx2 122, Rx3 123, can then "pick up", replacing the main program by a secondary program in the data signal, and then broadcast the signal thus modified to receiving terminals.

Such a stall technique is particularly described in the ANSI / SCTE 35 standard. It relies on the use of the MPEG2-TS transport stream for the transport of stall commands. According to this technique, the transport layer is used to transmit information related to the insertion of regional or local data.

As a result, this stall technique applies exclusively to MPEG2-TS type streams, which is highly limiting.

In addition, it requires the implementation of a national head-end capable of inserting ANSI / SCTE 35-specific stall notification data. It is also noted that this stall notification data must be synchronized with the content. audiovisual to consider a proper transition at the moment of stall.

There is therefore a need for a new failover or failover technique, which is simple to implement and inexpensive, and which is not limited to a particular type of stream, such as MPEG2-TS streams.

3. Presentation of the invention

The invention proposes a new solution that does not have all of these disadvantages of the prior art, in the form of a method of switching from a first program to a second program in a broadcast network implementing a head end and at least one receiver.

According to the invention, such a method implements the following steps, at the level of at least one of the receivers:

detecting, in a signal carrying the first program, a specific audio fragment intended to provide a transition between two programs;

replacement, in the signal, of the first program by the second program, in case of positive detection.

The invention thus proposes an original solution for stall based on the recognition of a particular audio fragment. Such a particular audio fragment, if it is detected in the signal, is therefore likened to a stall order. Note that such an audio fragment is conventionally provided in the signal to ensure a transition between two programs, hiding the imperfections related to the transition. In other words, such an audio fragment is conventionally used to ensure the linking between two programs. According to the invention, this audio fragment thus has a dual function, since it makes it possible both to ensure a transition between two programs and to achieve a stall.

In other words, the invention proposes to analyze the audio portion of the signal (which may be a simple audio signal, an audio / video signal, a multimedia signal, etc., digital or analog), to detect whether the signal carries a particular audio fragment.

The invention thus makes it possible for an audio receiver (belonging for example to a broadcasting site or to a terminal of the smart phone, tablet, computer, car radio or, more generally, any terminal adapted to receive a broadcast signal) type to achieve a stall. a first source to a second source on detecting a fragment of sound.

The invention is therefore at the level of the physical layer, which makes it possible to process a wide range of data streams, as long as they include an audio part. In particular, since the stall information is part of the audio according to the invention, the proposed solution is applicable independently of the chosen transport layer (M PEG2-TS, IP / RTP, etc.) and coding. In contrast, the stall technique of the ANSI / SCTE 35 standard of the prior art was implemented at the transport layer and therefore applied exclusively to MPEG2-TS streams.

Moreover, the invention provides more precision on the moment of stall in relation to the first source. Indeed, as soon as the receiver recognizes a specific audio fragment, it knows that it must switch to the second source, for example just at the end of the specific audio fragment.

Finally, it should be noted that, in order to guarantee the effectiveness of the proposed solution and correct detection, the particular audio fragment must respect a minimum duration, for example at least equal to 1.5 seconds.

In particular, the specific audio fragment is of the sonic comma or sonal ("jingle") type.

The technique according to the invention, according to at least one embodiment, therefore does not require the modification of the contents, since sonic commas or jingles are already inserted at the beginning and end of program ranges during production. . Thus, classically, commas surround a range of advertising. It is thus possible according to the invention to switch from a first source (for example broadcasting a national program) to a second source (broadcasting for example a local advertisement) to detect a first sound point, then to switch back from the second one. source to the first source to detect a second decimal point.

In addition, such an audio fragment is easily detectable, and allows to limit the size of databases, since we find substantially the same programs at the same times every day (series, news, games, etc.).

In particular, the second program (or the second source broadcasting the second program) is chosen taking into account at least one criterion belonging to the group comprising:

a specific type of audio fragment

a specific audio fragment content;

a clock ;

profile information of a user;

a location information of a user.

Another source is thus selected according to the detected specific audio fragment and / or information relating to the user of the terminal when the detachment is implemented at the terminal.

According to a particular aspect, the detection step is implemented over a predetermined time range.

It is thus possible to implement the detection step only over a specific time range, during which it is expected to receive a stall sequence (for example between Hh58 and 12h02), which makes it possible to optimize the resources of the receiver.

For example, this information is set at the receiver of the broadcast site or the terminal. It can also be transmitted from the headend to the receiver. This information can also be updated regularly.

According to a first embodiment, the detection step implements the following steps:

determining at least one signature associated with an audio fragment of the signal; comparing the signature to a base of known signatures of the receiver. In other words, at the broadcast site or the terminal, the signal is analyzed as and when the signal is regularly calculated (for example every 2 or 3 seconds) a new signature on the basis of a predetermined number audio samples of the signal.

At each determination, the resulting signature is compared to a signature database (also called a signature bank) known to the receiver.

This first embodiment therefore implements a calculation algorithm of signatures, also called "fingerprinting" in English.

An advantage of such a technique is that it is resistant to signal processing, including compression and coding, or slight signal alteration. Thus, the signature associated with an audio fragment is unchanged whether the signal is compressed or not, which allows the detection of a specific audio fragment even if it is encoded or compressed, or slightly degraded.

In addition, this first embodiment does not require the modification of the contents to be broadcast during their production.

According to a second embodiment, the detection step detects a tattoo representative of a specific audio fragment, previously inserted into the specific audio fragment.

In other words, a tattoo is inserted in an audio fragment broadcast with the first program (either in the studio, during the creation of the first program, or at the head of the network), and we try to detect at the receiver a specific audio fragment comprising a tattoo.

According to this second embodiment, the stall occurs when the receiver receives a tattoo.

This second embodiment therefore implements a tattoo algorithm, also called "watermarking" in English.

Again, an advantage of such a technique is that it is resistant to signal processing, including compression and coding, or slight signal corruption. Thus, the tattoo is invariant even if the signal is coded or compressed, or slightly degraded.

In addition, another advantage of this technique is that it does not require complex calculations at the receiver. Indeed, the tattoo is calculated and inserted in an audio fragment before broadcast, and the receiver implements a simple tattoo detection (for example by means of a correlator).

According to a third embodiment, the detection step implements a step of comparing at least two samples of the signal with at least two samples of at least one audio fragment known to the receiver.

This third embodiment can particularly apply if the signal is very simple, and has not undergone any coding or compression processing.

In this case, it is considered that the receiver has at its disposal a base of audio fragments, and that it compares audio samples of the signal with audio samples of audio fragments. known to the receiver.

According to another aspect of the invention, the switching method comprises a preliminary step of decoding the signal, feeding the detection step.

Of course, the different characteristics of this tilting process can be combined.

In another embodiment, the invention relates to a module for switching from a first program to a second program in a broadcast network implementing a headend and at least one receiver.

According to the invention, such a failover module comprises:

means for detecting, in a signal carrying the first program, a specific audio fragment intended to provide a transition between two programs;

replacement means, in the signal, of the first program by the second program, in case of positive detection;

the failover module being adapted to be implanted in at least one of the receivers.

Such a failover module, also called a switch, is particularly suitable for implementing the switching method described above. It is for example integrated in a receiver of a broadcast site or a user terminal. This module can of course include the various characteristics relating to the switching method according to the invention, which can be combined or taken separately. Thus, the features and benefits of this module are the same as those of the failover method and are not detailed further.

The invention also relates to a smartphone-type terminal, tablet, computer, car radio, or more generally any terminal adapted to receive a broadcast signal (over IP, radiofrequency, etc.), equipped with such a switching module. In particular, such a terminal comprises means for restoring the modified signal.

The invention also relates to a method of rediffusion of a modified signal, in a broadcast network implementing a headend and at least one broadcast site comprising a receiver and a re-transmitter.

According to the invention, such a method implements the following steps, at the level of at least one of the diffusion sites:

receiving a signal broadcast by the head end, the signal carrying a first program; detecting, in the signal, a specific audio fragment intended to provide a transition between two programs;

in case of positive detection, replacement, in the signal, of the first program by a second program, delivering a modified signal;

broadcast of the modified signal.

Such a rediffusion method thus implements the switching method described above to replace a first program with a second program in a signal, and to broadcast the modified signal thus obtained.

It therefore has the same characteristics and advantages as those of the tilting process.

In another embodiment, the invention relates to a site for broadcasting a modified signal, in a broadcast network implementing a headend, the broadcast site comprising a receiver and a re-transmitter.

According to the invention, the receiver comprises:

means for receiving a signal broadcast by the head end, the signal carrying a first program;

means for detecting, in the signal, a specific audio fragment intended to provide a transition between two programs;

means for replacing, in the signal, the first program by a second program, delivering a modified signal, activated in the event of positive detection;

and the retransmitter comprises means for broadcasting the modified signal.

Such a broadcast site is particularly adapted to implement the replay method described above. It includes a receiver, which can be followed by a repeater or integrated into the transponder.

The invention further relates to a method of broadcasting a program in a broadcast network implementing a headend and at least one receiver.

According to the invention, such a method implements the following steps, at the head end:

generating a signal carrying the program, said first program, and at least one specific audio fragment intended to provide a transition between two programs;

broadcasting information representative of the audio fragment (s) specific;

signal diffusion;

so as to allow a switchover of the first program to a second program detection by at least one of said receivers, or the specific audio fragments.

In particular, according to a first embodiment, the information representative of a specific audio fragment is a signature (also called acoustic fingerprint) obtained from the specific audio fragment. This signature is transmitted beforehand, during a first broadcast step (IP or radio frequency) to the receiver (s) broadcast sites or terminals to power a bank of signatures. Then, during a second broadcast step, the signal carrying the program and the specific audio fragment is also transmitted to the receiver (s) of the broadcast sites or terminals.

As already described, at the level of a broadcasting site or a terminal, the signal is then analyzed progressively and a new signature is regularly calculated, and at each determination, the signature obtained is compared with the bank of known signatures of the receiver.

According to a second embodiment, the information representative of a specific audio fragment is a tattoo representative of the specific audio fragment.

In this case, the tattoo is inserted into the specific audio fragment, and the steps of broadcasting information representative of the specific audio fragment and of the signal diffusion are implemented simultaneously (since the tattoo is carried by the specific audio fragment). even worn by the signal).

In another embodiment, the invention aims to protect a headend broadcasting a program in a broadcast network implementing the headend and at least one receiver.

According to the invention, the headend comprises:

means for generating a signal carrying the program, said first program, and at least one specific audio fragment intended to ensure a transition between two programs;

means for broadcasting information representative of the specific audio fragment or fragments;

signal broadcasting means;

so as to allow a switch from the first program to a second program detection by at least one of the receivers, the specific audio fragment or fragments. Such a headend is particularly adapted to implement the broadcasting method described above.

The invention also relates to one or more computer programs comprising instructions for implementing a switching method and / or a replay method and / or a broadcasting method as described above, when this or these programs are executed by a processor.

The methods according to the invention can therefore be implemented in various ways, in particular in hard-wired form or in software form.

This or these programs can use any programming language, and be in the form of source code, object code, or intermediate code between source code and object code, such as in a partially compiled form, or in n ' any other desirable form.

The invention also relates to one or more computer-readable and non-transitory storage media storing one or more computer programs comprising processor executable instructions for implementing a failover method and / or a replay method and / or or a diffusion method as described above.

A storage medium can be any entity or device capable of storing the computer program (CD-ROM, hard disk, etc.).

4. List of figures

Other features and advantages of the invention will appear more clearly on reading the following description of a particular embodiment, given as a simple illustrative and nonlimiting example, and the appended drawings, among which:

Figure 1, described in relation to the prior art, shows the structure of a broadcast network;

FIG. 2 presents the main steps of a diffusion method according to one embodiment of the invention;

FIG. 3 illustrates the main steps of a switching method according to one embodiment of the invention;

FIG. 4 illustrates the operation of a receiver of a broadcasting site according to a particular embodiment of the invention;

Figure 5 shows a first embodiment of the detection step of Figure 2;

FIG. 6 presents a second embodiment of the step of detecting the Figure 2;

FIGS. 7 and 8 show examples of a simplified structure of a failover module implementing a switching technique and a headend implementing a broadcasting technique according to a particular embodiment of the invention.

5. Description of an embodiment of the invention

The general principle of the invention is based on the use of the audio part (or a portion of the audio part) of a signal to switch from a first program carried by the signal to a second program, in a broadcasting network. Such a switchover may be implemented either at a broadcast site or directly at the terminal of a user, if such a terminal has the ability to integrate an audio recognition algorithm.

A stall (or hang-up) order according to the invention is therefore obtained when a specific audio fragment is detected in the signal.

In relation to FIG. 2, the main steps implemented by a headend broadcasting a program in a broadcast network implementing the headend and at least one receiver (belonging to a broadcast site or to a broadcast site) are presented. a user terminal), according to a particular embodiment of the invention. It will be recalled here that the term "broadcast site" refers to an equipment or system comprising a receiver followed by a retransmitter or integrated in a retransmitter.

During a generation step 21, a signal carrying a first program and at least one specific audio fragment (sound point, jingle, etc.) is generated. As already indicated, such an audio fragment is conventionally provided in the signal to ensure a transition between two programs.

During a broadcast step 22, the signal is broadcast or transmitted to the different broadcast sites and / or terminals, so as to allow a switch from the first program to a second detection program, by the receiver of at least one broadcast site or terminal, the specific audio fragment.

In particular, information representative of the specific audio fragment is also broadcast or transmitted (20) to the various broadcast sites and / or terminals.

According to the embodiment considered, this piece of information representative of the specific audio fragment may be a signature (also called acoustic fingerprint) obtained at from the specific audio fragment, transmitted beforehand to the signal to feed a bank of signatures, or a tattoo representative of the specific audio fragment, inserted into the specific audio fragment, and transmitted in the signal, or other.

The main steps implemented by a method of switching from a first program to a second program in a broadcast network as described above are now presented in relation to FIG.

During a first detection step 31, a receiver of a broadcast site or a user terminal analyzes the signal carrying the first program, which can be broadcast by the headend Tx 11, and tries to detect the presence of a specific audio fragment (such as a sonic comma, a jingle, etc.) in this signal.

In case of positive detection, the detected audio fragment is likened to a stall order, and during a second stall step 32, the receiver replaces the first program with a second program in the signal. According to the invention, the specific audio fragment thus has a dual function, since it makes it possible both to ensure a transition between two programs and to achieve a stall.

For example, the first program is a main program, broadcast by a national head end. A sounding point is inserted at the end of the main program, during the production of this program in the studio or when it is broadcast by the head end. When it detects this sound point, the receiver of a broadcast site or the user terminal knows that it can immediately switch to a second program, also called secondary program, broadcast for example by a local source.

The same steps can be implemented at the end of the secondary program: as for the main program, a sound point is inserted at the end of the secondary program, during the production of this program in the studio or when it is broadcast by a source local. When it detects this sound point, the receiver of the broadcast site or the user terminal knows that it can immediately return to the main program (or switch to a third program, depending on the content and / or type of the detected audio fragment and / or according to at least one information related to the user). The technique according to the invention can therefore be used to switch from a first program to a second program, or to switch from the second program to the first program.

In particular, the invention can be implemented in a method of rediffusion of a modified signal. In this case, in addition to the detection and stalling steps 31 implemented by a receiver of a broadcast site, the invention also implements a step of broadcast 33 by the repeater of the broadcast site, broadcasting the modified signal obtained by replacing the first program by the second program in the signal.

In the following, with reference to FIG. 4, the operation of a receiver according to a particular embodiment of the invention will be described. By way of example, an audio receiver having n sources S1, S2, S3, Sn, that is to say capable of receiving contents (programs) originating from n different sources, such as a national headend SI, is considered as an example. a local studio S2, files on a S3 d isque, etc. For example, the national head end SI broadcasts a national stream on satellite contribution, the local studio S2 broadcasts a local stream over IP, S3 audio files are recorded on a disk, and so on. It is recalled that the chosen transport layer (M PEG2-TS, I P / RTP, etc.) does not affect the tilting technique proposed by the invention.

Such a receiver is particularly adapted to receive a program from a first source among the n sources S1, S2, S3, Sn, to pick up to receive a program from a second source among the sources S1, S2, S3 , Sn, then hang up to receive a program from the first source again, using the failover technique according to the invention.

In this exemplary embodiment, the tilting or stalling technique is based on four main modules of the audio receiver:

a stall module 41;

a stall command manager module 42;

a decoding module 43; and

an audio recognition module 44.

More precisely, according to this exemplary embodiment, it is considered that the stall command manager module 42 manages the programming of stalls. Thus, the stall command manager module 42 ensures in particular that the algorithm for detecting an audio fragment is implemented only during a predetermined time range Wtemp during which a stall is expected, in activating the audio recognition module 44 only during the predetermined time range. This time range Wtemp, also called security window, can be parameterized and updated at the level of the stall command manager 42, or parameterized and updated at different sources.

When, during this security window, the expected audio fragment is detected by the audio recognition module 44, the stall command manager module 42 informs the stall module 41 on the new source on which the receiver must switch, in depending on the type and / or the content of the detected audio fragment and / or according to at least one information related to the user. For example, the receiver must switch from the source S2 to the source Sn, or from the source S3 to the source S1, etc.

The stall module 41 thus selects the source (s) to be listened to on a stall command transmitted by the stall command manager module 42. If the data stream coming from this new source is coded, it is transmitted. to the decoding module 43 (optional).

The decoding module 43 decodes the streams and feeds the output of the receiver. The audio portion of the decoded data stream is also presented to the audio recognition module 44, which implements the audio fragment detection algorithm. For example, as the data stream received by the receiver decodes, a predetermined number of audio samples of the decoded data stream are temporarily stored in a buffer memory 45 to compare these audio samples with audio samples known to the audio receiver. receiver.

As indicated above, the audio recognition module 44 starts or stops the audio fragment recognition algorithm on order of the stall command manager module 42. When it detects the expected audio fragment, it informs the manager module 42. For example, according to a first embodiment, the audio recognition module 44 implements a signature calculation algorithm, also called "fingerprinting" in English, in order to identify the audio fragment sought. According to a second embodiment, the audio recognition module 44 implements a tattoo detection algorithm, also called "watermarking" in English.

Indeed, as already indicated, such algorithms have a high resistance to coding.

In particular, the audio recognition module 44 can integrate a space storing the audio signature (s) to be found in order to trigger the stall, in the form of a base or bank of signatures. This module can also manage the audio signatures to be detected, such as adding, deleting or updating signatures in the signature database, for example.

Note that these modules can be used in a similar way to switch from a second source to a first source. In particular, the stall command manager module 42 is capable of managing hang-up commands by using the mechanism described above for the processing of the stall commands.

With reference to FIG. 5, a first embodiment of FIG. the invention according to which the audio recognition module 44 implements a signature calculation algorithm.

According to this example, only two sources, denoted SI and S2, are considered.

For example, the source SI generates a digital signal carrying a first program comprising an audio part. It is considered according to this example that sound V commas are inserted into the digital signal before broadcast of the first program, for example at the beginning and end of this program. As illustrated in dashed line, the signatures associated with these commas V can be calculated in a signature calculation module 51 and transmitted to the terminal and / or to the broadcasting site (via a radiofrequency or IP broadcast for example) for storage in a database. signatures 52, advantageously asynchronously to the production and broadcasting of the first program. Such a terminal or broadcast site therefore offers a data storage capacity, such as an audio signature bank.

According to this example, in a first step, the receiver (of the broadcast site or of the user terminal) receives a digital signal broadcast by the source S1. As described above, the audio portion of the digital signal is decoded if necessary, and presented to the audio recognition module 44. As the digital signal is received and decoded, the audio recognition module 44:

calculates a new signature associated with a predetermined number of audio samples of the digital signal in an audio signature calculation module 441,

compares the obtained signature with signatures previously stored in the signature database 52 in an audio signature comparison module 442.

When a specific audio fragment, such as the sound V comma, is detected (ie when the signature associated with this fragment calculated in the module 441 is identical to a signature present in the signature database 52), the audio recognition module 44 informs the stall command manager 42. The stall command manager 42, taking into account a base of actions to be performed according to the detected signature 53, generates a stall (or on-hook) order and informs the stall module. 41 of the new source on which the receiver must switch, such as the source S2, by means of an audio source selector for example. In a second step, the receiver therefore receives a program broadcast by the source S2 (which can be rebroadcast by the repeater of the broadcast site in the case where the receiver is the receiver of a broadcast site, or restored on the terminal by example by playing the audio part).

In particular, the stall command manager 42 may notably take into account at least one criterion belonging to the group comprising:

a specific type of audio fragment

specific audio fragment content;

a clock ;

profile information of a user;

location information of a user;

to decide which source the receiver should switch to.

For example, if we consider a receiver of a broadcast site having three sources SI (satellite for example for the national stream), S2 (IP for example for regional information) and S3 (internal storage or files available on the network, for example for local advertising), the following actions can be defined in the action base to be performed: regional stall (satellite to IP), advertising stall (satellite to internal memory), return to the main source ( IP to satellite or internal memory to satellite), etc. Each action will be associated with a particular signature (trigger).

According to another example, if a receiver of a terminal is considered, the secondary source may be a stream broadcast over an IP network, a stream broadcast by radio wave (FM), a remote file available in a network or a file. present in a storage space of the terminal, previously downloaded for example.

In particular, an action / signature association scheme for associating an action with a particular signature according to data from the headend, for example, can be provided to the terminal and / or the broadcast site, in order to assist the stall manager 42 in deciding which source to switch to. Advantageously, this signature / action association scheme can be transmitted asynchronously from an action / signature association module 57 from the headend / production site to the action base module 53.

When the failover module is implanted in a user terminal, optional clock 54, geolocation 55 and / or user profile 56 modules can be used to allow the stall manager 42 to customize the choice of the second one. source with respect to the time of stall execution, the location of the user, and / or the user's preferences. Thus, a user terminal may switch to a second program corresponding to an advertising page on the restaurants around the user at lunch time.

For example, the choice of the replacement audio source is firstly dependent on the detected signature, as defined in an action / signature association scheme of the module. basic action 53. Then, this choice can be refined, adapted to the geographical location and / or customized according to the technologies supported by the terminal (geolocation, user profile, etc.) from the clock modules 54, geolocation 55 and / or user profile 56.

It should be noted that the technique used to calculate the signatures is a conventional technique, such as those described in the document "A Review of Algorithms for Audio Fingerprinting" (P. Cano et al., In International Workshop on Multimedia Signal Processing, US Virgin Islands , December 2002) for example.

In particular, the audio recognition module 44 can implement a calculation algorithm and comparison of signatures as described in the patent application WO2005101292 filed March 18, 2005.

FIG. 6 also shows a second embodiment of the invention according to which the audio recognition module 44 implements a tattoo detection algorithm.

According to this second embodiment, it is determined, before broadcasting a program, at least one tattoo representative of a specific audio fragment as a sonic comma. Thus, during a preliminary step, a tattoo is inserted in the specific audio fragment, in order to obtain, for example, a tattooed sound point Vw. It is noted that these tattoo determination and insertion operations in the audio fragment can be performed in the studio. The insertion of a tattooed audio fragment in a digital signal carrying a program can be done in the studio, during the creation of the program, or by the source, prior to the broadcast of the program.

As illustrated in FIG. 6, two sources, denoted SI and S2, are again considered according to this example.

For example, the source SI generates a digital signal carrying a first program comprising an audio part. According to this example, tattooed sound points Vw are inserted by the source S1 into the digital signal carrying the first program before it is broadcast, for example at the beginning and the end of this program.

According to this embodiment, the digital signal broadcast by the source SI, for example a national headend, is modified, since it carries at least one tattoo.

In a first step, the receiver of the broadcast site or the user terminal therefore receives a digital signal broadcast by the source SI, carrying at least one tattooed audio fragment. As described above, the audio portion of the digital signal is decoded if necessary, and presented to the audio recognition module 44. When a specific audio fragment, such as the Vw soundtrack carrying a tattoo, is detected (ie when a tattoo is received), the audio recognition module 44 informs the stall manager 42.

The stall command manager 42, taking into account a base of actions to be performed according to the tattoo detected 61, generates a stall (or hang-up) order and informs the stall module 41 of the new source on which the receiver must switch, like source S2, by means of an audio source selector for example. For example, it is possible to insert different tattoos depending on the source on which you want to switch. In a second step, the receiver therefore receives a program broadcast by the source S2 (which can be rebroadcast by the repeater of the broadcast site in the case where the receiver is the receiver of a broadcast site, or restored on the terminal by example by playing the audio part).

In particular, the stall command manager 42 can in particular take account of at least one criterion belonging to the group comprising:

a specific type of audio fragment

specific audio fragment content;

a clock ;

profile information of a user;

location information of a user;

to decide which source the receiver should switch to.

For example, if we consider a receiver of a broadcast site having three sources SI (satellite for example for the national stream), S2 (IP for example for regional information) and S3 (internal storage or networked files for local advertising), we can define the following actions in the action database to be performed: regional stall (satellite to IP), advertising stall (satellite to internal memory), return to the main source (IP to satellite or internal memory to satellite), etc. Each action will be associated with a particular tattoo (trigger).

In particular, an action / tattoo association scheme for associating an action with a particular tattoo according to data from the headend, for example, can be provided to the terminal and / or the broadcast site, in order to help the stall manager 42 to decide which source to switch to. Advantageously, this tattoos / actions association scheme can be transmitted asynchronously from an action / tattoo association module 65 from the headend / production site to the basic action module 61. failover module is implemented in a user terminal, optional clock 62, geolocation 63 and / or user profile 64 modules may be used to allow the stall manager 42 to customize the choice of the second source by report at the time of stall execution, the location of the user, and / or the preferences of the user. Thus, a user terminal may switch to a second program corresponding to an advertising page on the restaurants around the user at lunch time.

For example, the choice of the replacement audio source is firstly dependent on the tattoo detected, as defined in an action / tattoo association scheme of the basic action module 65. Then, this choice can be refined, adapted to the geographical location and / or customized according to the technologies supported by the terminal (geolocation, user profile, etc.) from the clock modules 62, geolocation 63 and / or user profile 64.

It is noted that the technique used to calculate the tattoo is a conventional technique, for example implementing a spread spectrum, as described in the document "Detecting Hidden Information from Watermarked Signal using Granulation Based Fitness Approximation", M. Davarynejad et al. , Applications of Soft Computing: From Theory to Praxis, Springer, Series: Advances in Intelligent and Soft Computing, Volume 58/2009. Other techniques may also be envisaged, such as those described in documents EP 0 658 989 ("Device for the frequency multiplexing of a data signal and a sound signal"), EP 0 829 145 ("Broadcasting system of data using the properties of the human ear "), or FR 2 812 503 (" Method and system for coding-decoding digital information in a sound signal transmitted by a reverberant channel "), in the name of the same Applicant as the this patent application.

Finally, in connection with FIGS. 7 and 8, examples are given of the simplified structure of a headend and a failover module, or switch, intended to be implemented in a user terminal or in a receiver. a broadcast site, implementing a switching technique according to an embodiment described above. These Figures 7 and 8 illustrate only one particular way, among several possible, to achieve the various embodiments detailed above.

The failover module, illustrated in FIG. 7, comprises a RAM memory 70, a processing unit 71, equipped for example with a processor, and driven by a computer program stored in a ROM 72, implementing the switching method according to one embodiment of the invention.

At initialization, the code instructions of the computer program 72 are for example loaded into the RAM memory 70 before being executed by the processor of the processing unit 71. The processing unit 71 receives as input a signal carrying a first program, possibly in which at least one tattoo has been inserted. The processor of the processing unit 71 implements the steps of the switching method described above, according to the instructions of the computer program 72, to replace the first program with a second program when a specific audio fragment is detected. . For this, the failover module comprises, in addition to the memory 70, means for detecting, in a signal carrying the first program, a specific audio fragment, and means for replacing, in the signal, the first program by the second program, in case of positive detection. These means are controlled by the processor of the processing unit 71. This failover module can be an integral part of a receiver of a broadcast site, itself associated with a retransmitter or an integral part of the retransmitter, or a receiver of a terminal.

The headend, illustrated in FIG. 8, comprises a RAM 80, a processing unit 81, equipped for example with a processor, and driven by a computer program stored in a ROM 82, implementing the method of diffusion according to one embodiment of the invention.

At initialization, the code instructions of the computer program 82 are for example loaded into the RAM 80 before being executed by the processor of the processing unit 81. The processing unit 81 receives as input a program to be broadcast and one or more specific audio fragments. The processor of the processing unit 81 implements the steps of the broadcasting method described above, according to the instructions of the computer program 82, for broadcasting the signal and information representative of the specific audio fragment or fragments. For this, the head end includes, in addition to the memory 80, means for generating a signal carrying the program and the specific audio fragment or fragments, means for broadcasting information representative of the specific audio fragment or fragments, means of broadcasting the signal.

These means are controlled by the processor of the processing unit 81.

Claims

1. Method of switching from a first program to a second program in a broadcast network implementing a headend and at least one receiver,

characterized in that it implements the following steps, at the level of at least one of said receivers:

detecting (31), in a signal carrying said first program, a specific audio fragment intended to provide a transition between two programs;

replacing (32), in said signal, said first program by said second program, in case of positive detection.

2. Switching method according to claim 1, characterized in that said specific audio fragment is of the sonic or sonic comma type.

3. switching method according to any one of claims 1 and 2, characterized in that said second program is selected taking into account at least one criterion belonging to the group comprising:

a specific type of audio fragment

specific audio fragment content;

a clock ;

profile information of a user;

a location information of a user.

4. Switching method according to any one of claims 1 to 3, characterized in that said detecting step is implemented over a predetermined time range.

5. Switching method according to any one of claims 1 to 4, characterized in that said detection step implements the following steps:

determining (441) at least one signature associated with an audio fragment of said signal; comparing (442) said signature with a known signature database of said receiver.

6. Switching method according to any one of claims 1 to 5, characterized in that said detecting step detects a tattoo representative of a specific audio fragment, previously inserted into said specific audio fragment.

7. Switching method according to any one of claims 1 to 6, characterized in that said detecting step implements a step of comparing at least two samples of said signal with at least two samples of at least one fragment audio belonging to a base of audio fragments of said receiver.

8. switching method according to any one of claims 1 to 7, characterized in that it comprises a preliminary step of decoding said signal, feeding said detection step.

9. Module for switching from a first program to a second program in a broadcast network implementing a headend and at least one receiver,

characterized in that it comprises:

detection means (31), in a signal carrying said first program, of a specific audio fragment intended to provide a transition between two programs; replacement means (32), in said signal, of said first program by said second program, in case of positive detection;

said module being able to be implanted in at least one of said receivers.

Terminal characterized in that it comprises a switching module according to claim 9.

11. Broadcast site characterized in that it comprises a switching module according to claim 9.

12. Method for broadcasting a program in a broadcast network implementing a headend and at least one receiver,

characterized in that it implements the following steps, at said headend: generation (21) of a signal carrying said program, said first program, and at least one specific audio fragment intended to ensure a transition between two programs; broadcasting (20) information representative of said at least one specific audio fragment;

diffusing (22) said signal;

to allow a switchover of said first program to a second program detection, by at least one of said receivers, of said at least one specific audio fragment.

13. Head end broadcasting a program in a broadcast network implementing said headend and at least one receiver,

characterized in that said headend comprises:

means (21) for generating a signal carrying said program, said first program, and at least one specific audio fragment intended to provide a transition between two programs;

means for broadcasting (20) information representative of said at least one specific audio fragment;

broadcasting means (22) of said signal;

Computer program comprising instructions for implementing a switching method according to any one of claims 1 to 8 or a method of broadcast according to claim 12 when this program is executed by a processor.

A computer-readable and non-transient storage medium storing a computer program comprising processor-executable instructions for implementing a failover method according to any one of claims 1 to 8 or a broadcast method according to the present invention. claim 12.