FR2770675A1 - Device for decoding a digital numerical signal - Google Patents

Abstract

Means for combining the primary and secondary transforms of a digital audio signal The electrical analog output at the interface (50) is suitable for output of the first transformation (TP) of the digital audio obtained according to the chosen primary transformation function. This output can be reconstructed by the primary reconstruction device (RES1) as the first audio acoustic signal, the audio quality of which is not very high. The decoder also comprises a secondary transformation mechanism (60) which sets up a reciprocal or complementary function to the primary function. The secondary function has the primary function (TP) or the audio-numerical signal (SAN) as operand and outputs the second transform (TS). The control for reconstruction of primary and secondary transforms operates so as to provide an audio signal of the required quality.

Description

Device for decoding a digital audio signal capable of satisfactory listening according to selected acoustic criteria
The present invention relates to the decoding of a digital audio signal capable of satisfactory listening according to selected acoustic criteria, in particular for a normal ear.

It finds a particular application in the fight against unauthorized analog copying of a digital audio signal, also called "sonogram" and carrying music, speech, sound, or the like.

A system for distributing sonograms, for example online, is already known, via a communication network connecting listening terminals (or decoders) and a server containing said sonograms to be listened to.

The decoder is for example a multimedia personal microcomputer equipped with appropriate software and a communication interface connected to the communication network connecting the decoder and the server. The communication network can be a cable network, a satellite distribution network, the Internet, a telephone or data transmission network.

In this type of installation not limited to online installations, a digital communication protocol (coding forming a digital lock) allows only listening without the possibility of digital recording or digital storage of the digital audio signal thus transmitted.

Conversely, a previously stored digital file cannot be decoded by the decoder to be listened to.

In practice, the decoder includes an analog electrical output, capable of being connected to a high fidelity acoustic chain or to active acoustic speakers (with amplification), as well as to conventional, inexpensive and quality recording means. satisfactory registration.

 As a result, the user of the decoder can easily produce audio-analog copies of satisfactory acoustic quality. Thus, an indelicate user can reproduce and market fraudulently audio data for which he has only paid a listening right for private use.

The Applicant has already proposed, in its Patent FR 95 13089, a repression tool based on a marking of digital audio data making it possible to unambiguously identify the origin and the content of said data.

This solution provides physical evidence on unauthorized or fraudulent audio-analog copying, but it does not prevent audio-analog copying in itself.

There are also known toll data transfer installations in which a decoder receives an encrypted signal, not capable of satisfactory use, to then be decrypted at the level of the decoder after payment or access control and thus being capable of satisfactory use. .

This type of installation gives a degree of security in the transfer of data, but it does not prevent analog audio copying in itself either.

The present invention provides a solution to this problem.

Thus, it aims to prevent any audio-analog copy of a digital audio signal by rendering such audio-analog copy without market value.

It also aims to prevent fraudulent or unauthorized behavior that results from the dematerialization of the digital audio signal.

It starts from a device decoding a digital audio signal capable of satisfactory listening according to selected acoustic criteria, the decoder being of the type comprising a communication interface whose analog electrical output is capable of being connected to first means of acoustic reproduction capable of reproducing a first audio-acoustic signal.

According to a general definition of the invention, the analog electrical output of the interface is capable of delivering the primary transform of the audio-digital signal obtained according to a selected primary transformation function, said primary transform being capable of being restored by the first acoustic restitution means as first audio-acoustic signal the listening to which alone is not satisfactory according to said chosen acoustic criteria, and - the decoder further comprises - secondary transformation means suitable for establishing a secondary transformation function reciprocal and / or complementary to the primary function, having the primary transform or the audio-digital signal as operand, and capable of delivering the secondary transform of said digital audio signal according to said secondary transformation function, - second acoustic restitution means connected to the means secondary processing and specific to restore a second audio-acoustic signal representative of the secondary transform, and - control means suitable for controlling the first and second acoustic reproduction means such that the simultaneous listening of said first and second audio-acoustic signals thus restored corresponds to satisfactory listening of the digital audio signal according to said chosen acoustic criteria.

It should be noted that the restitution of the digital audio signal is done here only at the acoustic level and not electric by simultaneous combination (in phase or phase opposition) of two sound waves (acoustic pressure). As a result, this acoustic reproduction can only be picked up by the ear or else by acousticoelectric transducers which have the drawback of being very expensive when it is desired to maintain good quality of the signal.

This only acoustic and non-electric restitution therefore prevents making an audio-analog copy of quality with simple and inexpensive recording means.

According to a first embodiment of the invention, the communication interface comprises primary transformation means suitable for establishing the primary transformation function, an input to the communication interface receiving the audio-digital signal to be listened to.

According to a second embodiment of the invention, the primary transformation function is established outside the decoder, an input to the communication interface receiving the primary transform thus obtained outside the decoder.

In practice, the primary transformation function is established using generator means capable of generating a disturbing acoustic signal, and of incorporating, at a selected rate, said disturbing signal into the audio-digital signal to be listened to in order to obtain the primary transform, while the secondary transformation function is established using the secondary transformation means which are able to control the second acoustic restitution means to generate an attenuation signal corresponding to the secondary transform and able to attenuate the disturbing signal incorporated in the audio-digital signal so as to make it acoustically non-annoying.

Thus, the analog copy of the digital audio signal available at the analog electrical output of the decoder interface is made without market value due to the incorporation of the disturbing signal whose acoustic parameters are chosen to make listening to the signal thus disturbed completely unsatisfactory according to conventional acoustic criteria for a normal ear. The attenuation of the disturbing signal is done here at the acoustic level by generation of an audio-acoustic signal in phase opposition with the disturbing signal.

According to another embodiment of the invention, in order to make the analog copy of the digital audio signal without market value, the digital audio signal is no longer disturbed but amputated part of its useful information in a chosen spectrum. This mode is a variant of the previous one where the disturbance is no longer additive, but subtractive, which means that the combination of the primary and secondary transforms is here additive (complementary function) instead of being subtractive as in the previous mode.

In this embodiment, the primary and secondary transformation functions are established using spectrum separator means capable of separating, dynamically and / or statically, if necessary beyond a chosen cut-off frequency, the spectrum of the digital audio signal to be listened to in at least first and second distinct spectra, the separating means having a first output corresponding to the analog electrical output of the communication interface and delivering the first spectrum corresponding to the primary transform and a second output delivering the second spectrum corresponding to the secondary transform, the acoustic listening of only one of said first and second spectra not being satisfactory according to the chosen acoustic criteria, while the simultaneous and phase acoustic listening of the first and second spectra corresponds to satisfactory listening of the digital audio signal according to said chosen acoustic criteria.

Other characteristics and advantages of the invention will become apparent in the light of the detailed description below and the appended drawings in which - FIG. 1 schematically represents the means constituting the decoder according to the invention, in which the disturbance of the digital audio signal is carried out at the level of the decoder - FIG. 2 schematically represents the means constituting the decoder according to the invention, in which the disturbance of the digital audio signal is carried out outside the decoder, for example at the level of a server intended to be in communication with the decoder; and FIG. 3 schematically represents the means constituting a decoder according to the invention in which the digital audio signal is separated into two distinct spectra at the level of the decoder, only the simultaneous listening and in phase of the two spectra allowing satisfactory listening of the digital audio signal.

In the following description, the invention is implemented in an installation for online distribution of sonograms. Obviously, this example is not limiting. In fact, the invention can find an application in any online or offline installation in which the digital audio signal must be processed to prevent its fraudulent or unauthorized analog copying.

With reference to FIG. 1, an installation for online distribution of sonograms generally comprises three entities: a SER server, a COM communication network and DEC decoders.

The SER server contains sonograms intended for listening in real time or in delayed time at the level of the decoder.

The user of the decoder makes a request to the server to listen to a chosen sonogram. The request is processed according to an appropriate business transaction.

Sonograms are digital audio signals SAN stored in a mass memory of the hard disk type DR for example.

The COM communication network is a network capable of routing sonograms from the server to the client decoder in response to a chosen request. The communication network can be a cable network, a satellite distribution network, the Internet, a digital or analog telephone network or a data transmission network or the like.

On the client side, the DEC decoder includes an INT communication interface comprising an input 52 connected to the network
COM to communicate with the SER server.

For example, the DEC decoder is a personal multimedia microcomputer provided with appropriate communication software or a dedicated decoder. For example, within the framework of an Internet type network, the microcomputer is equipped with a modem and software allowing an Internet connection, as well as software making it possible to select a chosen sonogram and to receive, via the modem, said sonogram thus chosen.

In practice, the interface INT of the decoder DEC comprises an analog electrical output 50 which makes it possible to connect the decoder to a high-fidelity acoustic chain RES1 comprising amplification and / or processing means 32 and acoustic speakers 30.

The applicant company has posed the problem of providing such an installation with means making it possible to prevent any audio-analog copy of the sonogram thus transmitted.

A first solution according to the invention consists in including in the digital audio signal to be listened to SAN, a disturbing acoustic signal modulated in frequency, in phase, and / or in amplitude and acoustically attenuating this disturbance at the level of the acoustic restitution.

This acoustic disturbance can be generated either at the SER server (Figure 1), or more simply at the DEC decoder (Figure 2).

With reference to FIG. 1, the server SER comprises transformation means 10 capable of establishing a primary transformation function of the digital audio signal to be listened to
SAN. The transformation means 10 comprise an input 12 suitable for receiving the digital audio signal SAN. The transformation means 10 comprise an input 14 suitable for receiving the disturbing signal SP intended to be included in the digital audio signal to listen to SAN at the transformation means 10 which thus play the role of jammer.

The transformation means 10 comprise an output 16 delivering the primary transform TP of the digital audio signal
SAN. The output 16 is connected to the input 18 of the COM communication network.

In practice, the disturbing signal SP comes from a disturbance generator 20 whose output 22 delivers said disturbing signal SP.

The disturbance generator 20 includes an input 24 receiving signals representative of the disturbance law
LO which can be for example one or more frequencies modulated in all or nothing. This LO law is advantageously unpredictable. It is known to the server but not to the user of the decoder.

Most simply, the law of disturbance is used only by a limited number of frequencies 150 to 300 Hz for example, insofar as it is more difficult, but not impossible, to cancel acoustically beyond 300 Hz.

Different modes of disturbance can be used simultaneously or independently of one another, or even partially combined with each other depending on the degree of protection that one wants to provide.

The disturbance generator 20 includes an output 26 connected to the input 28 of the communication network COM. The output 26 delivers information INFO relating to the law of disturbance LO. In practice, the network inputs 18 and 28 form a single multiplexed input.

On the decoder side, the primary transform TP is delivered by the output 40 of the communication network COM and applied to the input 52 of the interface INT of the decoder which comprises, as seen above, an analog electrical output 50.

According to the invention, listening and / or recording the primary transform TP available on the analog electrical output 50 of the communication interface is completely devoid of market value since the TP signal is highly acoustically disturbed.

To allow satisfactory listening to the digital audio signal while preventing fraudulent or unauthorized analog copies, provision is made to equip the decoder with secondary transformation means 60 comprising an input 62 receiving the primary transform TP, an input 64 receiving the information INFO relating to the disturbance law via output 42 of the COM network, input 55 and output 57 of the communication interface.

In practice, the secondary transformation means 60 are suitable for delivering to the output 66 an attenuation signal according to the secondary transformation function corresponding to the secondary transform TS of the digital audio signal to be listened to.

This secondary transform is applied to a sound or acoustic source RES2 capable of acoustically restoring the secondary transform TS.

A microphone 72 is arranged near the sound source
RES2 to receive the signals emitted by the sound sources 30 and RES2. The microphone 72 is connected to the input 68 of the secondary transformation means 60 which further comprise processing and filtering means (not shown) capable of adapting, in real time, the signal delivered by the sound source RES2 as a function of the signal picked up by the microphone 72 to attenuate the disturbing signal SP included in the digital audio signal SAN.

 It should be noted that the secondary transform TS must be in phase opposition (TS = -SP) with the disturbing signal SP to cancel it in the disturbed signal TP according to the known principle of active noise attenuation.

In practice, the acoustic system RES1 is of the stereophonic type with at least two acoustic speakers 30 placed in suitable locations.

In this case, provision is made to associate with each of these conventional speakers 30, a small annex acoustic enclosure CAP called here a canceling capsule and which is intended to be fixed close to an acoustic enclosure 30 to attenuate the disturbing signal. Each capsule CAP comprises the secondary transformation means 60, a sound source RES2 and a microphone 72.

For example, the canceling capsule CAP is placed at a small distance from the center of the speaker 30 restoring the frequency bands corresponding to the disturbing signal of the primary transform.

Advantageously, the sound sources RES2 are connected to the secondary transformation means 60 through a particular analog or digital electrical interface not compatible with an audio recording on cassette or magnetic tape or the like. For example, the sound sources RES2 are produced from piezoelectric resonators or speakers associated with control electronics (not shown) having a proprietary digital electrical interface, for example encrypted. As a result, the sound sources have an electrical interface which is not compatible with those of conventional acoustic electrical transducers and therefore with the electrical signal TS to be recorded.

In addition, the CAP capsules are molded in a secure molding against external aggressions and the particularity of which is to destroy or deteriorate the means constituting the capsule if an attempt is made to open the molding to directly access the electrical signal TS. For example, the molding is filled with a hard resin trapping the electronic components making them inaccessible when the capsules are opened.

It should be noted that the capsules include a microphone which is connected to associated electronics (processing and filtering of the adaptive type) to auto-calibrate the cancellation signal as a function of the signal picked up by the microphone.

To confer an additional degree of security, the acoustic disturbance of the digital audio signal as well as its acoustic attenuation are implemented in association with an access control suitable for authorizing / prohibiting the simultaneous control of the first and second acoustic restitution means in response to predetermined operating conditions.

For example, access control uses conventional cryptological methods, so that knowledge of the law of disturbance by the user requires the use of heavy and complex means making it possible to make knowledge of this law ineffective and uneconomic. .

The access control mechanism is for example articulated around the user's request for a sonogram of his choice. The decoder has, for example, a smart card reader allowing the introduction of a smart card capable of carrying out conventional active authentication.

The law of disturbance may depend on the user and on a pseudo-random code associated with the commercial transaction and supplied by the server.

Advantageously, the canceling capsules are designed so as to include the access control device to authorize / prohibit the acoustic restitution of the cancellation signal of the disturbing signal.

With reference to FIG. 2, instead of being performed on the server side, the disturbance is generated at the level of the decoder according to a law depending on the user and a pseudo-random code X associated with the transaction and provided by the SER server in the event that an access control is implemented with the disturbance.

In this embodiment, the digital audio signal SAN is sent in clear from output 100 of the server SER to input 102 of the decoder DEC via input 104 and output 106 of the COM network.

At the level of the decoder, the signal SAN is applied to the input 102 of the primary transformation means 110, the analog electrical output 112 of which is capable of delivering the primary transform TP of the digital audio signal SAN.

For the rest, the signal corresponding to the primary transform
TP is applied to the acoustic reproduction chain RES1 in the same way as that described with reference to Figure 1.

Similarly, the signal corresponding to the primary transform TP is applied to secondary transformation means 60 capable of attenuating the disturbing signal acoustically according to the principle of active noise attenuation as described with reference to FIG. 1.

Access control makes it possible, if necessary, to authorize / prohibit the operation of the secondary transformation means 60 as described with reference to FIG. 1. Here, the pseudo-random code X is routed to the input 64 of the means secondary transformation via output 120 from the server, input 122 from the network, and output 124 from the network.

The secondary transformation means calculate the transform of the pseudo-random code X according to an encryption key personal to the decoder.

With reference to FIG. 3, the signal is no longer disturbed as described with reference to FIGS. 1 and 2, but divided into at least two complementary parts whose border can be dynamic and / or static beyond a chosen cutoff frequency .

For example, the signals representative of the low and mid frequencies leave at the output 112 in analogical towards the standard acoustic chain RES1. These signals in fact represent the primary transform TP as defined above.

For example, the cutoff frequency here is of the order of 1000 Hz.

In this variant, the decoder comprises dynamic and / or static separator means 200 of spectra capable of separating dynamically and / or statically, beyond a chosen cutoff frequency, the spectrum of the digital audio signal to be listened to SAN in at least first TP and second TS complementary spectrum s whose border can be dynamic and / or static.

Here, the term static border means a cutoff limit fixed at a chosen frequency, while the term dynamic border means a limit which can move according to a law for cutting the spectrum of the temporal or frequency type chosen.

In practice, this cutting law is known only to the server and not to the user.

The separator means 200 have an input 201 receiving the signal SAN, a first analog electrical output 202 delivering the first spectrum corresponding to the primary transform TP, and a second output 204 delivering the second spectrum corresponding to the secondary transform TS.

According to the invention, the acoustic listening of only one of said first and second spectra is not satisfactory according to conventional acoustic criteria for a normal ear, while the acoustic listening in phase and simultaneous of the first and second spectra is satisfactory according to the acoustic criteria chosen.

Such a solution thus makes it possible not to disturb the signal to be listened to, but to make it incomplete, which generates such poor quality that its analog copy has no market value when it is thus amputated.

In practice, the treble spectrum corresponds to the second TS spectrum. It is returned by preferably tamper-resistant RES2 speakers connected to the decoder by a digital or analog interface, advantageously of a confidential or secure nature, of wired or non-wired link type, for example in infrared light.

Advantageously, the separation means 200 are designed so as to include an access control device for authorizing / prohibiting the acoustic reproduction in phase of the two spectra.

Advantageously, the second acoustic restitution means
RES2 are connected to the separation means 200 through a particular analog or digital interface which is not compatible with recording on magnetic tape.

In addition, the second acoustic restitution means are molded as described with reference to FIG. 1 to secure the interface of said second acoustic restitution means with the separating means, against external aggressions.

It is important to note that the signals SAN, SP, -SP,
TS or -TS are never available on an analog electrical interface or output, compatible with magnetic recording on tape or analog cassette. These signals are also not accessible to the user, to prevent him from recording said signals either directly or after an electrical combination carried out using a simple device.

Here, only the TP signal is available on an analog electrical output and accessible to the user. It is therefore recordable, but it cannot be heard satisfactorily because it is disturbed or incomplete.

Claims (13)

Claims 1. Decoder device for a digital audio signal capable of satisfactory listening according to selected acoustic criteria, the decoder (DEC) being of the type comprising a communication interface (INT) whose analog electrical output (50) is capable of be connected to first acoustic restitution means (RES1) capable of restoring a first audio-acoustic signal, characterized in that - the analog electrical output (50) of the interface is suitable for delivering the primary transform (TP) of the signal audio-digital obtained according to a selected primary transformation function, said primary transformed (TP) being capable of being reproduced by the first acoustic restitution means (RES1) as the first audio-acoustic signal whose listening alone is not satisfactory according to said chosen acoustic criteria, and in that the decoder further comprises - own secondary transformation means (60, 200) s to establish a reciprocal and / or complementary secondary function of the primary function, having the primary transform (TP) or the audio-digital signal (SAN) as operand, and capable of delivering the secondary transform (TS) of said digital audio signal according to said secondary transformation function, - second acoustic restitution means (RES2) connected to the secondary transformation means (60, 200) and suitable for restoring, on command, a second audio-acoustic signal representative of the secondary transform (TS), and - control means suitable for controlling the first and second acoustic reproduction means such that the simultaneous listening of said first and second audio-acoustic signals thus restored corresponds to satisfactory listening of the audio-digital signal according to said chosen acoustic criteria. 2. Device according to claim 1, characterized in that the communication interface comprises primary transformation means (110) suitable for establishing the primary transformation function, an input (102) of the communication interface (INT) receiving the digital audio signal to listen to. 3. Device according to claim 1, characterized in that the primary transformation function is established outside the decoder, an input (52) of the communication interface (INT) receiving the primary transform (TP) thus obtained outside the decoder. 4. Device according to any one of the preceding claims, characterized in that the control means comprise access control means suitable for authorizing / prohibiting the control of the first and second acoustic restitution means (RES1 and RES2) in response at predetermined conditions. 5. Device according to one of the preceding claims, characterized in that the interface between the secondary transformation means (60, 200) and the second acoustic restitution means (RES2) comprises securing means capable of preventing electrical copying analog of the secondary transform. 6. Device according to claim 5, characterized in that the securing means comprise means of the secure molding type and capable of protecting the second acoustic restitution means (RES2) against external aggressions intended to electrically intercept the secondary transform (TS) . 7. Device according to claim 1, characterized in that the primary transformation means (10) comprise generator means (20) able to generate a disturbing acoustic signal (SP) according to the primary transformation function, and to incorporate at selected rate the disturbing signal to the audio-digital signal to be listened to in order to obtain the primary transform (TP). 8. Device according to claim 7, characterized in that the secondary transformation means (60) are capable of controlling the second acoustic restitution means (RES2) to generate an attenuation signal (-SP) according to the secondary transformation function corresponding to the secondary transform (TS) of the digital audio signal to be listened to, said attenuation signal being in phase opposition with the disturbing signal (SP) to cancel it acoustically in the primary signal (TP). 9. Device according to claim 8, characterized in that the second acoustic restitution means (RES2) are fixed near the first acoustic restitution means (RES1). 10. Device according to claim 9, characterized in that the second acoustic restitution means (RES2) comprise a sound source, a microphone (72) and processing and filtering means adapted in real time to eliminate the disturbing signal. 11. Device according to claim 1, characterized in that the primary and secondary transformation means comprise dynamic and / or static separator means (200) of spectrum capable of separating, if necessary, beyond a cut-off frequency chosen, the spectrum of the audio-digital signal to be listened to in at least first and second spectra, the separating means (200) having a first output (202) delivering the first spectrum corresponding to the primary transform (TP) and a second output (204 ) delivering the second spectrum corresponding to the secondary transform (TS), listening to only one of said first and second spectra not being satisfactory according to the acoustic criteria chosen. 12. Device according to claim 11, characterized in that the cut-off frequency beyond which the spectral separation takes place is of the order of 1000 Hz. 13. Device according to claim 11, characterized in that the connection between the second output (204) of the separation means (200) and the second acoustic restitution means (RES2) is secured to prevent an analog electric copy of the secondary transform ( TS).