Processing a stereophonic signal for a monophonic listening and monophonic listening

Abstract

Method (100) for processing a stereophonic input signal (SSTER) comprising a left channel (LIN) and a right channel (RIN), comprising the steps of generating a center channel (Mid) representing a sum of the left channel (LIN) and the right channel (RIN) and a first side channel (Side(L), Side(R)) representing a subtraction of one of the left channel (LIN) and the right channel (RIN) from the other of the left channel (LIN) and the right channel (RIN); shifting the phase of the first side channel (Side(L)) by 90° to obtain a first phase-shifted side channel (Sh_Side(L)); generating a monophonic output channel (SRMS) representing a sum of the center channel (Mid) and the phase-shifted side channel (Sh_Side(L)); and send the first monophonic output channel (LOUT) to a sound reproduction system configured to generate a monophonic listening sound signal by a user.

Description

DOMAINE TECHNIQUE DE L'INVENTION

[0001] The invention relates to a method of generating a monophonic listening signal from a stereophonic signal, within the framework of monophonic listening. ARRIERE PLAN TECHNOLOGIQUE

[0002] Recording a piece of music is most often done in such a way as to be able to reproduce that piece using a stereophonic system, that is, a system designed to recreate the spatial distribution of sound sources during recording (spatialization). This "stereophonic" effect is usually achieved by recording a sound scene with two microphones, each microphone being used to generate one of two sound channels, conventionally referred to as the left and right channels. Thus, a piece of music is often found in stereophonic formats, in which an audio signal is composed of two channels.

[0003] These stereophonic formats are designed for stereophonic playback, with each channel reproduced by a dedicated speaker in the stereo system. However, a user may wish to play a track in monophonic format from a stereo file. This is a common situation and occurs, for example, when the user only has a monophonic playback system such as a smartphone or a wireless speaker, which often only have one speaker.

[0004] In such a situation, a common solution is to merge the two channels by summing them and reproduce the signal obtained from the single speaker. However, this operation results in a loss of information corresponding to what is referred to in the field of audio signal processing as the out-of-phase content of the signal.

[0005] This loss of information can be explained using the Figure 1 and of the Figure 2 These figures each illustrate two signals: a signal S1 and a signal S2 corresponding respectively to the two channels of a stereophonic signal. A third signal, S1+S2, represents the sum of these two signals.

[0006] In the specific case illustrated by the Figure 1 The two signals S1 and S2 are identical and in phase. Their sum therefore corresponds to a signal S1+S2 with an amplitude twice that of signals S1 and S2.

[0007] In the specific case illustrated by the Figure 2 The two signals S1 and S2 are identical, but in opposite phase to each other, that is to say, their respective phases are shifted by 180° relative to each other. Their sum is therefore zero.

[0008] In the situation illustrated by the Figure 2 Summing the S1 and S2 signals results in a loss of information, namely the "out-of-phase" content of the stereophonic signal. This out-of-phase content is responsible for our perception of stereo and contains critical sonic information that is lost in traditional monophonic reproduction of a stereophonic signal.

[0009] In general, signals S1 and S2 are not simply either in phase or out of phase, but rather comprise information represented by components of the in-phase signals and other information represented by components of the out-of-phase signals. The level of information loss will depend on each stereophonic signal, and it may be the responsibility of a sound engineer to ensure what is called the "mono compatibility" of the stereophonic signal—that is, to prepare a stereophonic signal in such a way that the sound message retains its coherence despite the information loss when reproduced in monophonic sound. Imposing this compatibility limits the sound engineer's freedom of action. Furthermore, a significant degradation in audio quality will generally be observed when a stereophonic signal is reproduced in monophonic sound compared to when the same signal is reproduced in stereophonic sound. STATEMENT OF THE INVENTION

[0010] An object of the invention is a method for processing a stereophonic input signal, comprising in-phase and out-of-phase content, and for generating a monophonic output signal comprising out-of-phase content of the stereophonic input signal.

[0011] To achieve this goal, a first aspect of the invention is a method for processing a stereophonic input signal comprising a left channel and a right channel, comprising the steps of: generating a center channel and a first side channel, the center channel being representative of a sum of the left channel and the right channel and the first side channel being representative of a subtraction of one of the left channel and the right channel from the other of the left channel and the right channel; shifting the phase of the first side channel by 90° in absolute value in a first direction to obtain a first out-of-phase side channel; and generating a first monophonic output channel representing a sum of the center channel and the first out-of-phase side channel, the method further comprising a step of sending the first monophonic output channel to a sound reproduction system configured to generate a monophonic listening sound signal by a user.

[0012] The two channels of the out-of-phase signal are identical, but their polarities are reversed relative to each other; in other words, there is a 180° phase difference between them. Therefore, by varying the phase of the out-of-phase signal by +90° on one channel and by -90° on the other channel, this signal is brought back into phase and becomes audible in monophonic listening, as the +90° and -90° values ​​have an absolute difference of 180° (a simple phase inversion on one of the two channels only preserves one side of the input stereophonic signal).

[0013] Since the two channels of the out-of-phase signal become identical, it effectively becomes possible to process, from a computational point of view, only one channel to save computing resources.

[0014] This process generates a signal suitable for monophonic sound reproduction that incorporates out-of-phase content from the original stereophonic signal. The resulting sound reproduction is richer in information than a conventional monophonic signal, which lacks this out-of-phase content, resulting in a sound reproduction that is far superior to that of a conventional monophonic signal. The tonal balance achieved during the various stages of the original stereophonic signal's production, and the corresponding emotional impact, are preserved even when converted to monophonic playback.

[0015] Indeed, tests conducted by the inventor have shown that the sound in monophonic playback, based on the monophonic output channel according to the invention, is faithful to the original stereophonic sound. In other words, the listener perceives very distinctly and can affirm that the sonic balance achieved during the various stages of production (recording, mixing and mastering, or the final stage of audio post-production) is restored and the intended emotion is recreated, whether this rebalancing is subtle or very perceptible, depending on the audio content being reproduced.

[0016] Furthermore, the generation of a monophonic signal according to the invention ensures that, regardless of a listener's position relative to a sound source reproducing this signal, the listener always perceives the left and right channels of the stereo signal in their entirety, that is, in the same proportions, as if they were located at the ideal listening point in a stereo configuration (apart from the spatialization), which is the apex of an equilateral triangle with a pair of speakers at the other two vertices. An advantage of the technique according to the invention is therefore the listener's mobility, as they are not constrained to a fixed position predetermined by the positions of the two speakers.This aspect is particularly advantageous compared to the situation of a portable speaker whose left and right speakers are very close together (less than 90 cm, the minimum distance commonly accepted for speakers in stereo configuration), in which the listener can hardly stand at the ideal listening point.

[0017] Thus, the concept of the invention is suitable for businesses, public facilities, and so-called "multiroom" systems that broadcast mono audio signals rather than stereo. Indeed, in addition to a simplified setup, mono broadcasting allows for a uniform distribution of sound in space, unlike stereo broadcasting, which results in variations in sound quality depending on the listener's location. For example, in a stereo setup, a listener near the left speaker may perceive a very different sound from a listener near the right speaker. Processing an input stereophonic signal according to the invention allows, in these situations, for increased efficiency of the sound message reproduced in mono.

[0018] Similarly, this concept is suitable for individuals who desire a uniform spatial distribution of monophonic sound faithful to the original stereophonic sound. The input signal processing according to the invention ensures that the most neutral sound balance is always perceived by the listener in a moving situation, and therefore also by several listeners located at different points around the speaker (whether they are stationary or moving).

[0019] Previously, the average consumer lacked clarity regarding the use of stereo, mono, or pseudo-stereo (speakers with closely spaced drivers less than 90cm apart, the minimum standard recognized by professional studios for the listener to be positioned at the ideal listening point, the apex of an equilateral triangle with the speakers, the position where audio productions are created and where spatialization is correctly perceived). Our concept now provides a logical listening experience for the average consumer: stereo if they desire spatialization (by remaining stationary at the ideal listening point in a suitable stereo configuration), and Real Mono Sound (RMS) processing according to the invention if they wish to be mobile, the other solutions (mono and pseudo-stereo) now being of lesser utility.

[0020] Thus, by switching from a conventional stereo mode to a monophonic mode according to the invention, the user can freely choose between (i) stereophonic listening with spatialization information, in which case he is required to remain at a fixed point so as to form an equilateral triangle with a pair of speakers for optimal listening, and (ii) monophonic mode with out-of-phase content made audible, in which case the listener is mobile since the listening quality does not depend on his location.

[0021] Another advantage of the method according to the invention is that sound engineers and producers can now use out-of-phase content creatively instead of simply mitigating the effects of its disappearance on the audio signal when switching to monophonic playback. For example, they can increase the level of this content to enhance spatialization (the stereo effect) while still achieving faithful mono reproduction. In other words, the method according to the invention also involves, in terms of sound engineering, overcoming mono incompatibility (or the loss of the so-called "sides" signal on a mono speaker).

[0022] According to additional, non-limiting features of the first aspect of the invention, considered individually or in any technically feasible combination: The processing method according to the invention may further include a step of generating an output signal which consists of the first monophonic output channel and therefore a reduced number of channels compared to the stereophonic input signal; the processing method according to the invention may further include the steps of generating a second side channel of polarity opposite to the first side channel; phase shifting the second side channel by 90° in absolute value in a second direction opposite to the first direction to obtain a second phase-shifted side channel;generate a second monophonic output signal (R OUT) representative of a sum of the center channel and the second out-of-phase side channel, wherein the first monophonic output channel and the second monophonic output channel are identical by construction and form a two-channel monophonic output signal, the method further comprising a step of sending the two-channel monophonic output signal to the sound reproduction system configured to generate a monophonic listening sound signal by a user; the first side channel may be representative of a subtraction in which the right channel is subtracted from the left channel; the processing method according to the invention may further comprise a step of reducing the amplitudes of the left and right channels and limiting the monophonic output channel before it is sent in limited form to the sound reproduction system;and the process can be implemented using a monophonic electronic sound reproduction device configured to implement the steps of the process either by software or by hardware.

[0023] The invention extends to: a computer program comprising instructions which, when the program is executed by a digital computer, cause the latter to apply to a stereophonic input signal the steps of the process according to the invention; a computer-readable medium comprising instructions which, when executed by a digital computer, cause the latter to apply to a stereophonic input signal the steps of the process according to the invention; and an electronic chip configured to implement the steps of the process according to the invention.

[0024] The invention extends to an electronic device configured to implement the steps of the process according to the invention by software or by hardware.

[0025] According to additional, non-limiting features of the first aspect of the invention, considered individually or in any technically feasible combination: The electronic device according to the invention can be configured to switch from a first operating mode in which the device generates the first monophonic output channel to a second operating mode in which the device generates a monophonic output channel representative of the center channel; the electronic device according to the invention may include a control interface configured to switch the device from the first operating mode to the second operating mode; the electronic device according to the invention may include a control interface configured to switch the device from the first operating mode to an operating mode in which the first monophonic output channel is representative of the first phase-shifted side channel excluding the center channel;and the electronic device according to the invention can be chosen from a smartphone, a monophonic speaker, a sound recording playback system, and a centralization module configured to distribute the first monophonic output channel to a set of speakers.

[0026] The invention extends to a monophonic listening method comprising (i) the implementation of the processing method according to the invention (ii) the emission of a monophonic sound signal on the basis of the first monophonic output channel, by means of the sound reproduction system configured to generate a monophonic listening sound signal. BREVE DESCRIPTION DES FIGURES

[0027] Other features and advantages of the invention will become apparent from the detailed description of the invention which follows with reference to the accompanying figures in which: [ Fig. 1 ] There figure 1 illustrates the concept of "in-sync content"; [ Fig. 2 ] There figure 2 illustrates the concept of "out-of-phase content"; [ Fig. 3 ] There figure 3 illustrates a first mode of operation of a first mode of implementation of a process according to the invention; [ Fig. 4 ] There figure 4 illustrates a second mode of operation of the first implementation mode; [ Fig. 5 ] There figure 5 illustrates a first mode of operation of a second mode of implementation of a process according to the invention; [ Fig. 6 ] There figure 6 illustrates a second mode of operation of the second implementation mode; [ Fig. 7 ] There figure 7 illustrates a third mode of operation of the second mode of implementation; [ Fig. 8 ] There figure 8 illustrates a variation of the second implementation method; [ Fig. 9 ] There figure 9 illustrates a first mode of operation of a third mode of implementation of a process according to the invention; [ Fig. 10 ] There figure 10 illustrates a second mode of operation of the third implementation mode; [ Fig. 11 ] There figure 11 illustrates a first mode of operation of a fourth mode of implementation of a process according to the invention; [ Fig. 12 ] There figure 12 illustrates a second mode of operation of the fourth implementation mode; [ Fig. 13 ] There figure 13 illustrates a first operating mode of a fifth implementation mode of a process according to the invention; [ Fig. 14 ] There figure 14 illustrates a second mode of operation of the fifth implementation mode; [ Fig. 15 ] There figure 15 illustrates a third mode of operation of the fifth mode of implementation; [ Fig. 16 ] There figure 16 represents a first device configured to implement a process according to the invention; [ Fig. 17 ] There figure 17 represents a second device configured to implement a process according to the invention; [ Fig. 18 ] There figure 18 represents a system configured to implement a process according to the invention; [ Fig. 19 ] There figure 19 illustrates a first variant of a ninth implementation method of a process according to the invention; [ Fig. 20 ] There figure 20 illustrates a second variant of the ninth implementation method; [ Fig. 21 ] There figure 21 illustrates a first variant of a tenth embodiment of a process according to the invention; [ Fig. 22 ] There figure 22 illustrates a second variant of the tenth implementation method; [ Fig. 23 ] There figure 23 summarizes the process of monophonic listening according to the invention. DESCRIPTION DETAILLEE DE L'INVENTION

[0028] The invention aims to improve the monophonic sound reproduction of a stereophonic signal by means of a process known as "RMS processing". To this end, the inventors have developed a method for preserving the out-of-phase content of the stereophonic signal in a derived signal, suitable for high-quality monophonic sound reproduction. First method of implementation

[0029] THE figures 2 And 3 illustrate a first implementation method 100 of a process according to the invention. This first implementation method includes three modes of operation.

[0030] This process involves processing an input stereophonic signal (S STER) comprising a left channel (L IN) and a right channel (R IN), with the aim of transforming it into an output signal suitable for reproduction in an enhanced monophonic mode, known as RMS mono (Real Mono Sound stereophonic), or for reproduction in a conventional stereophonic mode (two operating modes illustrated by the...). Figure 3 ), or to a reproduction in a conventional monophonic mode (illustrated by the Figure 4 ) .

[0031] There figure 3 This illustrates a first operating mode of the first implementation mode, enabling the production of an S RMS output signal suitable for playback according to the enhanced monophonic RMS playback mode; suitable for playback using a monophonic sound system. The S RMS output signal includes the out-of-phase content of the S STER input stereo signal.

[0032] In a first processing stage St1, the stereophonic input signal S STER is transformed or encoded into a so-called Mid / Side signal by means of an MS-Enc functional block. At this stage, the information contained in the stereophonic signal S STER can be distributed into three channels designated by Mid, Side(L) and Side(R).

[0033] The Mid channel is a signal representing the information located in the middle of the soundstage represented by the stereo input signal S STER and corresponds to the sum of the two channels L IN and R IN. The Mid channel is representative of the sum of the two channels L IN and R IN and can be calculated by applying a summation function to these two channels using a summation function block Sum1 configured to sum the signals supplied to it as input and provide their sum as output, here such as Mid = L IN + R IN .

[0034] The Side(L) channel is a signal representing information located on the sides of the soundstage represented by the stereo input signal S STER and corresponds to the difference between the two channels L IN and R IN. The Side(L) channel represents the difference between the two channels L IN and R IN, and can be calculated by applying a subtraction function to these two channels using a Diff1(L) function block configured to perform a difference between two input signals and output their difference, such as Side (L) = L IN - R IN .

[0035] The Side(R) channel is a signal representing information located on the sides of the soundstage represented by the stereophonic input signal S STER and corresponds to the difference between the two channels R IN and L IN. The Side(R) channel is representative of the difference between the two channels L IN and R IN, and can be calculated by applying a subtraction function to these two channels using a Diff1(R) function block configured to perform a difference between two input signals and output their difference, such as Side(R) = R IN - L IN . We can therefore consider that the Side(R) channel and the Side(L) channel are of opposite polarities.

[0036] The Sum1, Diff1(L) and Diff1(R) functions are part of the MS-Enc function block, which can be implemented using conventional stereophonic audio signal processing techniques.

[0037] The Mid channel is split here (the Sum1 function block outputs two identical Mid channels) in anticipation of subsequent processing: as explained later, the Mid signal will be combined with two other signals. To maintain the relative amplitudes of these signals throughout the processing, it is necessary to balance their amplitudes. This is represented in the figures by splitting the Mid channel to prevent its amplitude from being halved when divided into two identical parts. Of course, this is a representation given for a specific implementation of the process, and alternatives with equivalent effects will undoubtedly be apparent to those skilled in the art.In general, this description outlines signal transmissions and combinations, and those skilled in the art will understand that, depending on the specific implementation, it may be necessary to balance the amplitudes of the signals involved. These adjustments are generally omitted to maintain conciseness.

[0038] The signals generated by the first processing stage St1 are transmitted to a second processing stage St2 via two selection functional blocks SelL and SelR are interposed between the Diff1(L) and Diff1(R) functional blocks and the Ph+ and Ph- functional blocks, respectively.

[0039] In the second processing stage, St2, phase shift functions are applied to the Side(L) and Side(R) channels using a Ph-Shift phase shift function block. In this implementation, phase shifts of +90° and -90° are applied to the Side(L) and Side(R) channels to obtain the Sh_Side(L) and Sh_Side(R) channels using Ph+ and Ph- phase shift function blocks, respectively, each configured to apply a phase shift to an input signal and output the phase-shifted signal. By their construction, the Sh_Side(L) and Sh_Side(R) channels are identical: Sh_Side(L) = Sh-Side(R).

[0040] Conversely, phase shifts of -90° and +90° can be applied to the Side(L) and Side(R) channels, respectively. However, the inventor has observed a loss of sonic brilliance in a few audio tracks when phase shifts of -90° and +90° are applied to the Side(L) and Side(R) channels, respectively. Thus, in the inventor's experience, the results are, in the vast majority of cases, more satisfactory in terms of listening quality, and particularly sonic brilliance, when phase shifts of +90° and -90° are applied to the Side(L) and Side(R) channels, respectively. For this reason, it is this latter configuration that is illustrated in the figure 3 and which is used for the description of this implementation method.

[0041] The signals processed by the second stage are transmitted to a first recombination stage.

[0042] At the Real recombination stage, two summations of the Mid, Sh_Side(L) and Sh_Side(R) channels are performed to obtain two A1 and A2 channels at the stage output.

[0043] The first channel A1 is representative of the sum of the two channels Mid and Sh_Side(L) and can be calculated by applying a summation function to these channels using a summation function block Sum Rec1L configured to sum the signals supplied to it as input and provide their sum as output, such that A1 = Mid + Sh_Side (L) = L IN + R IN + Sh_Side(L).

[0044] The second channel A2 is representative of the sum of the two channels Mid and Sh_Side(R) can be calculated by applying to these channels a summation function by means of a summation function block Sum Rec1R configured to sum the signals supplied to it as input and provide their sum as output, such that A2 = Mid + Sh-Side (R) = L IN + R IN + Sh-Side (R) .

[0045] The channels generated at the first stage of Real recombination are transmitted to a second recombination stage Rec2.

[0046] At the Rec2 recombination stage, a summation and two subtractions of channels A1 and A2 are performed to obtain three channels B1, B2 and B3.

[0047] The first channel B1 represents the sum of the two channels A1 and A2 and can be calculated by applying a summation function to these channels using a summation function block Sum2(L) configured to sum the input signals and output their sum, such that B1 = A1 + A2 = 2(L IN + R IN + Sh_Side(L)). Channel B1 is duplicated (the Sum2 function block outputs two identical B1 channels).

[0048] The second channel B2 represents the difference between the two channels A1 and A2, and can be calculated by applying a subtraction function to these two channels using a Diff2(L) function block configured to subtract two input signals and output their difference, such that B2 = A1 - A2 = Sh_Side(L) - Sh_Side(R), which is zero in the case modeled by the Figure 2 .

[0049] The third channel B3 represents the difference between the two channels A1 and A2, and can be calculated by applying a subtraction function to these two channels using a Diff2(R) function block configured to subtract two input signals and output their difference, such that B3 = A2 - A1 = Sh_Side(R) - Sh_Side(L), which is zero in the case modeled by the Figure 2 .

[0050] The Sum2, Diff2(L) and Diff3(R) functions can be part of a second MS-Enc functional block identical to the MS-Enc functional block of the first processing stage St1. These two functional blocks apply the same operations to the signals supplied to them respectively as input.

[0051] The B1, B2 and B3 channels generated at the second recombination stage Rec2 are transmitted to a third processing stage St3 via switching function blocks ComL and ComR, linking the Diff2(L) and Diff2(R) function blocks to two summation blocks Sum2(L) and Sum2(R) of the third processing stage, respectively.

[0052] In the third processing stage St3, two identical output channels L OUT and R OUT are generated so as to be adapted for transmission to a monophonic sound reproduction system, by means of an MS-Dec functional block, based on channels B1, B2 and B3. These two channels L OUT and R OUT form the output signal S RMS.

[0053] The output channel L OUT is representative of the sum of the two channels B1 and B2, and can be calculated by applying a summation function to these channels using the summation function block Sum2(L) configured to sum the signals supplied to it as input and provide their sum as output, such that L OUT = B1 + B2 = 2(L IN + R IN + Sh_Side(L)).

[0054] The output channel R OUT is representative of the sum of the two channels B1 and B3 and can be calculated by applying a summation function to these channels using the summation function block Sum2(R) configured to sum the signals supplied to it as input and provide their sum as output, such that R OUT = 2(L IN + R IN + Sh_Side(R)).

[0055] The two output channels, L OUT and R OUT, are identical by construction, as the channels Sh_Side(L) and Sh_Side(R) are identical: Sh_Side(L) = Sh_Side(R), and can therefore be transmitted to a monophonic sound reproduction system without additional processing. Indeed, since the channels are identical by construction, they will be identical regardless of the information content of the signal they form. The phase shifts applied to the Side(L) and Side(R) channels, which contain out-of-phase information content, have made it possible to equalize these signals, so that the so-called "out-of-phase" information content of the Side(L) and Side(R) channels is no longer out of phase at the level of the phase-shifted channels Sh_Side(L) and Sh_Side(R), and the information loss phenomenon described using the figure 2 is avoided.

[0056] The monophonic S RMS signal formed by the two output channels L OUT and R OUT can be transmitted to, for example, a conventional monophonic loudspeaker equipped with a system for converting a two-channel stereo input signal into a single-channel monophonic signal, which will then be reproduced by a loudspeaker. Conventionally, merging the two channels results in the loss of out-of-phase information, as explained above. However, when the stereophonic signal has been processed according to this operating mode, the information is no longer out of phase and is therefore not lost when the monophonic loudspeaker performs the conversion of the input S RMS monophonic signal.The sound reproduction of the S RMS signal has a much richer content and is closer to the ear of the original S STER stereophonic signal than the reproduction of a conventional monophonic signal obtained by a simple fusion of the channels of the original stereophonic signal.

[0057] There figure 3 also illustrates a second mode of operation of the first mode of implementation of the invention, allowing to recover at the output of processing an S STEREO signal identical to the original S STER signal.

[0058] This second mode of operation differs from the first mode of operation in the following respect: instead of transmitting the Side(L) and Side(R) channels to the second processing stage St2 and thus to the Ph+ and Ph- blocks, the two selection function blocks SelL and SelR transmit these two channels to the Real recombination stage via simple DerL and DerR branches connected to the Sum Rec1L and Sum Rec1R function blocks, respectively, without processing or modifying the content of the transmitted signals, and in particular without applying any phase shift to these channels. Alternatively, the Ph+ and Ph- blocks could be deactivated so as to transmit the channels supplied to them without modification. Apart from this identical transmission of the channels from the Diff1(L) and Diff1(R) blocks to the Real stage, the first and second modes of operation are identical.

[0059] Thus, in this second operating mode, channels A1 and A2, B1 to B3, and L OUT and R OUT are derived from the Mid, Side(L), and Side(R) channels, instead of the Mid, Side(L), and Side(R) channels used in the first operating mode. The same operations are applied to these channels as in the first operating mode. Therefore, we obtain A1 = 2L IN; A2 = 2R IN; B1 = 2(L IN + R IN); B2 = L IN - R IN; B3 = R IN - L IN; L OUT = 2L IN; and R OUT = 2R IN. In this second operating mode, the two output channels L OUT and R OUT combine to form a stereo S-channel signal at the output of the third processing stage. This S STEREO signal is a stereophonic signal, comprising two differentiated channels containing together an informational content identical in substance to the original S STER signal and sufficient to reconstruct the spatial distribution of sound sources during recording.

[0060] There figure 4 illustrates a third mode of operation of the first mode of implementation of the invention, allowing to obtain an S MONO output signal suitable for possible reproduction by means of a monophonic sound system but not including the out-of-phase content of the original stereophonic signal, unlike the S RMS signal of the first mode of operation.

[0061] This third operating mode differs from the first operating mode in the following respect: the ComL and ComR switching function blocks are configured to prevent the transmission of output channels from the Diff2(L) and Diff2(R) blocks to the Sum2(L) and Sum2(R) blocks. Apart from this point, the first and second operating modes are identical.

[0062] Thus, at the output of the Sum2(L) and Sum2(R) function blocks, the L OUT and R OUT channels are identical and each has the same value as the Mid channel: L OUT = R OUT = Mid = L IN + R IN. Whether the Ph-Shift function block is disabled or left active does not significantly affect this second mode of operation: a mono signal (S MONO) is formed from the two L OUT and R OUT channels.

[0063] The S MONO signal can be supplied as input to a monophonic speaker. Since the out-of-phase content of the stereophonic S STER signal is absent from the Mid channel, it is also absent from the sound reproduction produced by the speaker, which therefore has reduced information content compared to the original S STER signal and compared to the S RMS signal of the first operating mode.

[0064] The Ph+, Ph-, ComL and ComR blocks can be conventionally controlled by a user to compare the sound reproductions of the different operating modes and to observe the relative richness of content, with in particular the S RMS and S STEREO signals being much richer in information than the S MONO signal. Second method of implementation

[0065] THE figures 5 à 8 illustrate a second embodiment 200 of a process according to the invention. This second embodiment includes three modes of operation, respectively illustrated by the figures 5 à 7 , and a variant illustrated by the figure 8 .

[0066] This process is a treatment of an input stereophonic signal S STER comprising a left channel L IN and a right channel R IN with the aim of transforming it into an output signal suitable for stereophonic reproduction ( Figure 7 ) or a monophonic reproduction, and in the latter case to a reproduction in a conventional monophonic mode ( Figure 6 ) or an enhanced monophonic mode ( Figure 5 ) said RMS mono mode for Real Mono Sound.

[0067] There figure 5 illustrates a first mode of operation of the second mode of implementation, allowing to obtain an S RMS output signal suitable for possible reproduction by means of a monophonic sound system and including the out-of-phase content of the S STER input stereo signal.

[0068] This implementation includes the first, second, and third processing stages St1, St2, and St3 of the first implementation, but omits the first recombination stage Real and the second recombination stage Rec2. Therefore, refer to the description of the first implementation for its common elements with the second implementation.

[0069] Here, the output channels of block Sum1 are directly transmitted to blocks Sum2(L) and Sum2(R) and the output channels of blocks Ph+ and Ph- are directly transmitted to blocks ComL and ComR, considered to be part of the third processing stage St3, respectively.

[0070] In the third processing stage, the identical output channels L OUT and R OUT are generated, using an MS-Dec function block, to be suitable for transmission to a monophonic sound reproduction system, based on the Mid, Sh_Side(L) and Sh_Side(R) signals. These two channels, L OUT and R OUT, form the RMS output signal S.

[0071] The output channel L OUT is representative of the sum of the two channels Mid and Sh_Side(L) and can be calculated by applying a summation function to these channels using the summation function block Sum2(L) configured to sum the signals supplied to it as input and provide their sum as output, such that L OUT = Mid + Sh_Side(L).

[0072] The output channel R OUT is representative of the sum of the two channels Mid and Sh_Side(R) and can be calculated by applying a summation function to these channels using the summation function block Sum2(R) configured to sum the signals supplied to it as input and provide their sum as output, such that R OUT = Mid + Sh_Side (R) .

[0073] Two switching function blocks, ComL and ComR, are configured to, in this first mode of operation, transmit the Sh_Side(L) and Sh_Side(R) channels to the Sum2(L) and Sum2(R) function blocks, respectively.

[0074] As with the first operating mode of the first implementation, the two output channels L OUT and R OUT are identical by construction and form the monophonic signal S RMS, which can be transmitted to a monophonic playback sound system without additional processing. The phase shifts applied to the Side(L) and Side(R) channels, which contain out-of-phase information, have made it possible to equalize these signals, so that the so-called "out-of-phase" information content of the Side(L) and Side(R) channels is no longer out of phase at the phase-shifted channels Sh_Side(L) and Sh_Side(R), and the information loss phenomenon described using the figure 2 is avoided. The monophonic S RMS signal has an informational content enriched with out-of-phase content compared to a conventional monophonic signal, in which the out-of-phase content is lost.

[0075] There figure 6 illustrates a second mode of operation of the second mode of implementation of the invention, allowing to obtain an S MONO output signal suitable for possible reproduction by means of a monophonic sound system but not including the out-of-phase content of the original stereophonic signal, unlike the S RMS signal of the first mode of operation.

[0076] This second operating mode differs from the first in the following respect: the ComL and ComR switching function blocks are configured to prevent the transmission of the Sh_Side(L) and Sh_Side(R) channels to the Sum2(L) and Sum2(R) function blocks. Apart from this, the first and second operating modes are identical. Whether the Ph-Shift function block is active or not does not affect the second operating mode.

[0077] Thus, at the output of the Sum2(L) and Sum2(R) functional blocks, the L OUT and R OUT channels are identical by construction and each corresponds to the Mid channel: L OUT = R OUT = Mid. The S MONO signal is formed from the two L OUT and R OUT channels and can be supplied as input to a monophonic speaker. Since the out-of-phase content of the stereophonic S STER signal is absent from the Mid channel, it is also absent from the sound reproduction produced by the speaker, which therefore has reduced information content compared to the original S STER signal.

[0078] There figure 7 illustrates a third mode of operation of the second mode of implementation of the invention, allowing to recover at the output of processing an S STEREO signal identical to the original S STER signal.

[0079] This third operating mode differs from the first operating mode in the following respect: the Ph-Shift function block is deactivated, transmitting the Side(L) and Side(R) channels from the first processing stage to the third processing stage without applying any phase shift to these channels. Apart from this point, the first and third operating modes are identical.

[0080] Thus, at the output of processing, the output channel L OUT is representative of the sum of the two channels Mid and Side(L) and can be calculated by applying a summation function using the Sum2(L) function block to these two channels, such that L OUT = Mid + Side (L).

[0081] Similarly, the output channel R OUT is representative of the sum of the two channels Mid and Side (R) and can be calculated by applying a summation function using the Sum2(R) function block to these two channels, such that R OUT = Mid + Sh_Side_(R).

[0082] Thus, the S STEREO signal at the output of the third processing stage is a stereophonic signal, comprising two differentiated channels containing together sufficient information content to reconstruct the spatial distribution of sound sources during recording.

[0083] There figure 8 illustrates a variant of the second implementation mode applied to the third operating mode, also allowing the output of the processing to be an S STEREO signal identical to the original S STER signal.

[0084] With the application of this variant, the fourth operating mode changes from its version illustrated by the figure 7 On the following point: instead of disabling the Ph-Shift function block, the Ph+ and Ph- function blocks it contains are bypassed when a stereo output is desired.

[0085] More specifically, as with the implementation method illustrated by the Figure 4 The two SelL and SelR selection function blocks are interposed between the Diff1(L) and Diff1(R) function blocks and the Ph+ and Ph- function blocks, respectively. These function blocks are configured to transmit the Side(L) and Side(R) channels to the second processing stage when the desired signal is S RMS, as in the first operating mode. Conversely, in the current operating mode, where the desired signal is S STEREO, as in the third operating mode, these function blocks are configured to transmit the Side(L) and Side(R) channels directly to the third processing stage, where the phase shifts of the second processing stage are applied by means of the DerL and DerR derivations, respectively.

[0086] This mode of operation is particularly suitable for a hardware implementation of the process, but also represents a good implementation of the processing applied in software to the original stereo S STER signal. Third method of implementation

[0087] THE figures 9 And 10 illustrate a third embodiment 300 of a process according to the invention. This third embodiment includes two modes of operation, respectively illustrated by the figures 9 And 10 .

[0088] The third implementation method is particularly suited to minimal digital processing during software implementation, which has the advantage of limiting the consumption of resources (computing time, battery) of the device used to run the software implementation of the process.

[0089] This third implementation differs from the second implementation in the following ways: (i) the Side(R) channel is transmitted directly to the third processing stage, without any phase shift, regardless of the operating mode; (ii) the Sh_Side(L) channel is amplified by a Mult function block configured to double the amplitude of an input signal, which outputs the amplified Ash_Side(L) signal; (iii) the third processing stage includes two switching function blocks, Com3 and Com4, responsible for controlling the transmission of the ASh_Side(L) and Side(R) channels to the Sum2(R) function block, respectively. Apart from these three points, the second and third implementations are identical, and the explanations for the second implementation should therefore be used.

[0090] There figure 9 illustrates a first mode of operation of this third mode of implementation, allowing to obtain an S RMS output signal suitable for possible reproduction by means of a monophonic sound system and including the out-of-phase content of the S STER input stereo signal.

[0091] The Ph+ functional block can be implemented by means of a phase-shifting function by convolution of the Side(L) channel with an impulse response defined such that the convolution generates the Sh_Side(L) channel phase-shifted from the Side(L) channel by 90°, and more specifically by +90° in this example.

[0092] The Mid channel is transmitted to the inputs of the two function blocks Sum2(L) and Sum2(R), the ASh_Side(L) channel is transmitted to the Sum2(L) function block, and the switching function block Com3 transmits the ASh_Side(L) channel to the Sum2(R) function block, while the switching function block Com4 blocks the transmission of the Side(R) channel to the Sum2(R) function block. The ASh_Side(L) channel is divided into two equal parts and combined with the two Mid signals in such a way that the relative amplitude between the Mid and Side(L) signals, on the one hand, and the relative amplitude between the two Mid channels and each of the two signals originating from the ASh_Side(L) channel and supplied to the inputs of the Sum2(L) and Sum2(R) blocks, on the other hand, are equal, maintaining the original balance between the different information contained in the stereophonic S STER signal.

[0093] Thus, the output channels L OUT and R OUT are defined by L OUT = R OUT = Mid + Sh_Side(L), each output channel being obtained by summing the two Mid and Ash_Side(L) channels. The S RMS output signal therefore comprises two identical channels, with the advantages and applications already mentioned for the S RMS signal in the first and second implementation modes.

[0094] There figure 10 illustrates a second mode of operation of the third mode of implementation of the invention, allowing to recover at the output of processing an S STEREO signal identical to the original S STER signal.

[0095] This second mode of operation differs from the first mode of operation in the following ways: (i) the Ph-Shift function block is disabled, transmitting the Side(L) and Side(R) channels from the first processing stage to the third processing stage without applying any phase shift or amplification to these channels, (ii) the switching function block Com3 is configured to block the transmission of the Side(L) channel to the Sum2(R) function block and the switching function block Com4 is configured to transmit the Side(R) channel to the Sum2(R) function block.

[0096] Thus, we find ourselves in the situation of the third mode of operation of the second mode of implementation, illustrated by the Figure 7 : L OUT = Mid + Sh_Side(L) and R OUT = Mid + Sh_Side(R), the S STEREO signal at the output of the third processing stage is a stereophonic signal, comprising two differentiated channels containing together sufficient information content to reconstruct the spatial distribution of sound sources during recording.

[0097] Of course, instead of disabling the Ph-Shift block, one could proceed as described in the operating variant of the second implementation mode, illustrated by the figure 8 , in which selection functional blocks and derivations are used to transmit the Side(L) and Side(R) channels to the third processing stage without applying a phase shift to them, i.e. by bypassing the Ph+ and Ph- phase-shift functional blocks. Fourth method of implementation

[0098] THE figures 11 And 12illustrate a fourth embodiment 400 of a process according to the invention. This fourth embodiment includes two modes of operation, respectively illustrated by the figures 11 And 12 .

[0099] The fourth implementation mode is optimized for a device not designed to reproduce a stereophonic audio signal. This is the case, for example, with a monophonic speaker, such as a so-called "Bluetooth" speaker, configured to reproduce an audio signal using a single speaker. In such a case, the loss of the ability to provide a stereophonic output signal is not a problem, and the processing can therefore be simplified. In particular, it is not necessary to maintain an operating mode that produces a stereophonic signal.

[0100] Thus, this fourth implementation differs from the third implementation in the following ways: (i) the MS-Enc function block does not include the Diff1(R) function block, (ii) the MS-Dec function block does not include the ComL, ComR, Com3, and Com4 switching blocks, and (iii) a Com5 switching block is configured to control the transmission of a signal originating from the Diff1(L) function block to the third processing stage. Apart from these points, the third and fourth implementations are identical, and further details can be found in the description of the third implementation.

[0101] In this implementation mode, switching block Com5 is configured to control the transmission of the Side(L) channel to the Ph-Shift function block, but it could be located at another level, such as between the Ph-Shift function block and the MS-Dec function block, the main thing being to ensure that only the Mid signal is transmitted to the Sum2(L) and Sum2(R) function blocks.

[0102] There figure 11 illustrates a first mode of operation of this fourth mode of implementation, allowing to obtain an S RMS output signal suitable for possible reproduction by means of a monophonic sound system and including the out-of-phase content of the S STER input stereo signal.

[0103] This first mode of operation is similar to the first mode of operation of the third implementation mode, illustrated by the Figure 9 : the Side(L) channel is passed to the Ph-Shift function block by means of the Com5 function block, and the Ash_Side(L) channel is passed to the Sum2(R) function block.

[0104] Consequently, the output channels L OUT and R OUT are identical, with L OUT = R OUT = Mid + Sh_Side(L), each output channel being obtained by summing the two Mid and Ash_Side(L) channels. The S RMS output signal therefore comprises two identical channels, with the advantages and applications already mentioned for the S RMS signal in the first and second implementation modes.

[0105] There figure 12 illustrates a second mode of operation of the fourth mode of implementation of the invention, allowing to obtain an S MONO output signal suitable for reproduction by means of a monophonic sound system but not including the out-of-phase content of the original stereophonic signal, unlike the S RMS signal of the first mode of operation.

[0106] Function block Com5 is configured to block transmission of the Side(L) channel to the second and third processing stages. Consequently, the Sum2(L) and Sum2(R) function blocks are fed only by the Mid channel, and at the output of the Sum2(L) and Sum2(R) function blocks, the L OUT and R OUT channels are identical by construction and each has the same value as the Mid channel: L OUT = R OUT = Mid. The S MONO signal is formed from the two L OUT and R OUT channels and can be supplied as input to a monophonic speaker. Since the out-of-phase content of the stereophonic S STER signal is absent from the Mid channel, it is also absent from the sound reproduction produced by the speaker, which therefore has reduced information content compared to the original S STER signal and the S RMS signal. Fifth method of implementation

[0107] THE figures 13 à 15 illustrate a fifth embodiment 500 of a process according to the invention. This fifth embodiment includes two modes of operation, respectively illustrated by the figures 13 And 14 .

[0108] This fifth implementation is a simplified variant of the fourth implementation in the following ways: (i) the Sum1 function block outputs only a single Mid channel, (ii) the Ph-Shift function block does not include the Mult function block, and (iii) the MS-Dec function block does not include the Sum2(R) function block. Furthermore, a Com6 switching block is configured to control the transmission of the Mid channel to the Sum2(L) function block of the third processing stage, St3. Apart from these points, the fourth implementation can be used as a reference.

[0109] There figure 13 illustrates a first mode of operation of the fifth implementation mode, allowing an S RMS output signal to be obtained suitable for reproduction using a monophonic sound system and including the out-of-phase content of the S STER input stereo signal.

[0110] In the first processing stage, the stereophonic input signal S STER is encoded into a Mid / Side signal using an MS-Enc function block. At this stage, the information contained in the stereophonic signal S STER is distributed into two channels designated Mid and Side(L), as explained for the second implementation mode.

[0111] Switching block Com5 is configured to transmit the Side(L) channel to the Ph-Shift function block. Switching block Com6 is configured to transmit the Mid channel to the Sum2(L) function block.

[0112] In the second processing stage, a phase shift function is applied to the Side(L) channel using the Ph-Shift function block. In this implementation, a phase shift of +90° is applied to the Side(L) channel to obtain the Sh_Side(L) channel using the Ph+ phase shift function block, configured to apply a phase shift to an input signal and output the phase-shifted signal.

[0113] In the third processing stage, an L OUT output channel is generated to be suitable for transmission to a monophonic sound reproduction system, using an MS-Dec functional block, based on the Mid and Sh_Side(L) signals. This L OUT channel forms the S RMS output signal.

[0114] The output channel L OUT represents the sum of the two channels Mid and Sh_Side(L) and can be calculated by applying a summation function to these channels using the Sum2(L) function block, configured to sum the input signals and output their sum, such that L OUT = Mid + Sh_Side(L). The S RMS signal of this implementation has the advantages and applications already mentioned for the S RMS signal of the first and second implementations.

[0115] There Figure 14 This illustrates a second operating mode of the fifth implementation mode, allowing for specific listening to the out-of-phase signal. This second operating mode is identical to the first operating mode, except that the Com6 switching block is now configured to block transmission of the Mid channel to the Sum2(L) functional block.

[0116] The resulting output is an S OP signal, corresponding to the S RMS signal minus the Mid channel, meaning it contains only the out-of-phase content of the original S STER signal. This allows the listener to verify the audible content of the out-of-phase signal and better appreciate the advantage of the first operating mode compared to a conventional monophonic signal, which lacks the out-of-phase content. This operating mode can be applied to all the implementation modes described above by controlling the transmission and blocking of the Mid channel(s) at the output of the first processing stage, St1.

[0117] There Figure 15 illustrates a third mode of operation of the fifth mode of implementation, allowing to obtain an S MONO output signal suitable for reproduction by means of a monophonic sound system but not including the out-of-phase content of the original stereophonic signal, unlike the S RMS signal of the first mode of operation.

[0118] Unlike the first and second operating modes, the Com5 switching block is configured to block transmission from the Side(L) channel to the Ph-Shift function block. Furthermore, the Ph-Shift function block can be disabled.

[0119] Consequently, the output channel L OUT represents only the Mid channel, with L OUT = Mid since only the Mid channel is supplied as input to the Sum2(L) function block. The S RMS signal of this implementation has the advantages and applications already mentioned for the S RMS signal of the second implementation.

[0120] The S MONO signal consists of the single L OUT channel and can be supplied as the input to a monophonic speaker. Since the out-of-phase content of the stereophonic S STER signal is absent from the Mid channel, it is also absent from the sound reproduction produced by the speaker, which therefore has reduced information content compared to the original S STER signal and the S RMS signal. Sixth method of implementation

[0121] A sixth implementation method is illustrated by the Figure 16 .

[0122] The first, second, third, fourth, and fifth implementation modes can be implemented in software, for example, to apply a method according to the invention using a smartphone or other portable or non-portable sound reproduction device equipped with a monophonic speaker and supplied with stereophonic sound files. In this case, switching from one operating mode to another can be done manually, according to the user's choice, by entering a command via a control interface, for example, in the form of a dedicated physical button or a button that is part of a graphical interface of an application executing the method.

[0123] There Figure 16 This illustrates an example of a sound reproduction device using a Smart smartphone configured to implement the method according to the invention via software. The Smart smartphone includes physical control buttons Bt1, Bt2, and Bt3, a computing module Mod, a computer memory Mem in functional communication with the module Mem, and a speaker Spk. The Smart smartphone can employ a graphical control interface Int displaying virtual buttons with the same respective functions as the physical buttons Bt1, Bt2, and Bt3.

[0124] Thus, the Smart smartphone can include instructions in its memory (Mem) which, when executed by a computer processing module (Mod), cause the latter to apply the steps of the process according to the invention to a stereophonic input signal (S STER), corresponding, for example, to one of the first through fourth implementation modes. The third implementation mode represents an interesting compromise for a smartphone because it allows for resource-efficient processing while maintaining the possibility of stereophonic listening, which remains a viable option even with a smartphone.

[0125] The Bt1 button can be used to control the sound volume emitted by the Spk speaker.

[0126] The Bt2 button can be configured to switch to the operating mode described by the Figure 14 In this mode, the mid channel(s) are blocked at the output of the first processing stage, with the other operations performed such that the resulting output signal includes the out-of-phase content of the original stereo signal without the mid channel being present. The Bt2 button can be considered a mid channel mute button in an operating mode designed to generate a monophonic signal containing the out-of-phase content of an original stereo signal.

[0127] The Bt3 button can be configured to switch the device implementing the process from one operating mode to another. Considering, for example, the first implementation mode, the first, second, and third operating modes allow, from the same stereophonic signal S STER, the generation of (i) a monophonic signal S RMS containing out-of-phase content from the stereophonic signal, (ii) a monophonic signal S MONO devoid of the out-of-phase content from the stereophonic signal, and (iii) the recovery of a stereophonic signal S STEREO. Switching between these modes allows the user to test which mode best suits their needs and to listen to the difference between the reproduction of a monophonic signal containing the out-of-phase content (S RMS) and that of a monophonic signal devoid of this out-of-phase content (S MONO).With the help of stereo headphones connected to the smartphone, the user can also compare the reproduction of these two signals with that of the corresponding stereo signal (S STEREO). Thus, the user can determine which operating mode of the method according to the invention best suits their preferences and the track being listened to.

[0128] The Bt1, Bt2, and Bt3 buttons can be configured so that their respective actuations generate a control signal or signals for the software or hardware implementations of the phase-shifting, switching, or selection functional groups, thus controlling the operating mode used according to conventional methods. The physical Bt1, Bt2, and Bt3 buttons could be replaced respectively by Bt1', Bt2', and Bt3' buttons in a graphical interface. Seventh method of implementation

[0129] A seventh method of implementation is illustrated by the Figure 14 .

[0130] The first, second, third, fourth and fifth modes of implementation can be implemented in hardware using discrete components or using a so-called "DSP" chip (Digital Signal Processor in English terminology, or digital signal processor in French) integrated into a sound reproduction device, such as the monophonic speaker taken as an example in this mode of implementation.

[0131] There figure 17 represents a Spk MONO monophonic speaker, such as a so-called "Bluetooth" speaker, configured to implement the method according to the invention in a physical manner. The Spk Mono monophonic speaker comprises the physical control buttons Bt1, Bt2, and Bt3 and a Spk loudspeaker like those mentioned in the sixth embodiment illustrated by the figure 16 The Bt3 button can be configured to switch the device implementing the process between the two operating modes, allowing the user to choose the one that best reproduces the currently playing piece of music, as previously explained. The Bt2 button allows the user to specifically listen to the out-of-phase content of the original stereo signal, thus enabling them to perceive its audio content.

[0132] The Spk Mono monophonic speaker also includes a Chp DSP chip configured to apply one of the first through fourth implementation modes to an input stereophonic signal. The fourth and fifth implementation modes are particularly suitable: since the monophonic speaker is inherently specialized in reproducing a monophonic signal, operating modes designed to generate a stereophonic signal are superfluous. The fifth implementation mode (illustrated by the figures 13 And 14 ) can be considered more economical due to its greater simplicity compared to the fourth implementation method (illustrated by the figures 11 And 12 ).

[0133] In the case where the Spk MONO speaker applies the fourth implementation mode, the two-channel output signal (S RMS or S MONO consisting of the two L OUT and R OUT channels) is transmitted to the Spk speaker via a module (not shown) merging the two channels in a conventional manner and possibly amplifying the resulting signal, then transmitting it to the Spk speaker.

[0134] In the case where the Spk MONO speaker uses the fifth implementation mode, the output signal (S RMS or S MONO consisting of the L OUT channel) comprises only a single channel and does not require channel merging. It can therefore be directly transmitted to the Spk speaker, possibly via a module (not shown) that amplifies the signal and then transmits it to the Spk speaker. Eighth method of implementation

[0135] An eighth implementation method is illustrated by the Figure 18 .

[0136] This implementation method applies to the distribution of music through multiple monophonic speakers distributed throughout a space or multiple rooms, and is most often found in buildings accessible to the public, such as shopping centers or nightclubs, but also in private homes using so-called "multiroom" audio systems. In such situations, stereophonic listening is impractical due to the nature of the listening experience: speakers must be distributed in different rooms or at different points within the same room without meeting the conditions necessary for stereophonic sound reproduction (in which the listener is equidistant from two speakers directed towards them, each broadcasting one from a right channel and one from a left channel, respectively). Furthermore, the listener is required to move around, thus moving away from some speakers in the system and closer to others.

[0137] Here, monophonic sound reproduction is the only feasible method, but, as we saw earlier, a conventional monophonic signal lacks the out-of-phase content of the original stereophonic signal. Applying the generation of a monophonic signal that includes the out-of-phase content of the original stereophonic signal significantly improves the sound quality of the music being played.

[0138] There figure 18 illustrates such a music broadcasting system, which includes a centralization module incorporating a control device configured to receive and process an input stereophonic signal S STER so as to generate a monophonic signal S RMS including out-of-phase content of the stereophonic signal, according to any one of the implementation modes described above, the third and fourth being particularly suitable for the reasons set forth in relation to the seventh implementation mode illustrated by the figure 17 The Sys system also includes a Dist distribution module, configured to receive the monophonic S RMS signal and distribute it in parallel, via wired or wireless means, possibly in a modified or amplified form, to a SpkEns array consisting of a plurality of monophonic Spk speakers. These speakers are all powered by the same monophonic S RMS signal and thus produce an audio signal intended for monophonic listening by a user.

[0139] The control unit can use the DSP Chp chip configured to apply one of the first to fourth implementation modes to the input stereophonic signal, and is configured to transmit the signal, S RMS or S MONO, generated by the DSP Chp chip to the Dist distribution module. If necessary, the latter can be configured to merge the two channels of the signal supplied to it before transmission to the Spk monophonic speakers.

[0140] Like the smartphone of the figure 16 or the enclosure of the figure 17 The Sys system can be equipped with Bt1, Bt2, and Bt3 buttons, either physical or virtual. As with the SMART smartphone and the Spk MONO speaker, the Bt2 button can be configured to allow the listener to switch to a mode specifically for listening to the out-of-phase content of the source signal, separate from the Mid channel; the Bt3 button can be configured to switch between the different operating modes of the implementation, with their associated advantages. Ninth method of implementation

[0141] A ninth implementation method is illustrated by the Figure 19 and the Figure 20 .

[0142] The processing of a stereophonic input signal S STER generating an S RMS signal as illustrated by the figures 3 à 15 RMS processing, also known as RMS processing, induces an increase in the dynamic range of the sound. In other words, the amplitude of the processed RMS signal can exceed the maximum amplitude of the original stereo signal. RMS processing can therefore advantageously employ an audio limiter (LIMIT) to optimize the transients appearing in the RMS signal.

[0143] A limiter is a processor that prevents the signal from exceeding a certain level or threshold. This limiter can have the following characteristics: a "lookahead" function for better sound transparency, a "brickwall" behavior (absence of the "attack" parameter, no transient exceeding the limiting threshold), and an automatic release time (absence of the "release" parameter, shorter release for fast transients and longer release for long transients).

[0144] Almost all transients generated by RMS processing have amplitudes between 0 and +3 dB FS (decibels relative to full scale). The LIMIT limiter is therefore preferably configured for two operating modes: mode 1, known as "maximum volume," and mode 2, known as "high sound quality."

[0145] In Mode 1, the original signal amplitude is preserved, and the LIMIT limiter threshold is set to 0 dB FS. Therefore, any amplitude peaks generated by the RMS processing will be limited and will not exceed 0 dB FS. This mode has the advantage of preserving the original signal amplitude, and thus the corresponding volume when reproduced by a loudspeaker, optimizing the signal-to-noise ratio, and reducing signal distortion. This mode can be the default setting.

[0146] Mode 2 is recommended when using high-quality audio equipment. It involves reducing the amplitude in the input channels by 3 dB before RMS processing, while maintaining the limiter threshold at 0 dB FS. This preserves the amplitude peaks generated by the RMS processing, resulting in a natural dynamic range for the RMS signal (no peak clipping as in Mode 1). Only the occasional transient exceeding 3 dB, if any, is limited. This mode therefore offers the advantage of improved sound quality, as transients are reproduced naturally.

[0147] In the explanations above, a 3 dB reduction in the amplitude of the audio signal was indicated. However, a larger reduction (-6 dB, for example) could be used if the resolution of the audio environment does not allow the signal to exceed 0 dB FS. The threshold level of the LIMT limiter can then be adjusted to obtain the desired mode (-6 dB FS in our example for maximum volume mode 1 and -3 dB FS for high-quality sound mode 2).

[0148] THE figures 19 And 20 illustrate the application of the ninth implementation method to the first and fifth implementation methods respectively. Of course, the ninth implementation method could be applied similarly to the other implementation methods.

[0149] There figure 19 This illustrates a Mode 100, a variant of Mode 100 that also includes two Red Mod modules to reduce the amplitude of the input channels L IN and R IN before they are transmitted to the first processing stage, St1, when Mode 2 is applied. Associated with these reducers are two Sel selectors, configured to bypass the two Red Mod modules for the input channels L IN and R IN when Mode 1 is applied. Here, it is the maximum volume Mode 1 that is implemented, with the input channels bypassing the two Red Mod modules that reduce the amplitude of the input channels thanks to the Sel selectors. The output channels L OUT and R OUT are sent to the input of the LIMIT limiter, which limits their amplitude as explained above.

[0150] Similarly, The figure 20 This illustrates a '500' mode, a variant of the 500 mode that also includes two Red Mod modules for amplitude reduction of the input channels L IN and R IN before they are sent to the first processing stage, St1, when mode 1 is applied. Associated with these reducers are two Sel selectors, configured to bypass the two Red modules for the input channels L IN and R IN respectively when mode 1 is applied. Here, the high-quality sound mode 2 is implemented, with the two input channels being sent to the two Red Mod amplitude reduction modules by the Sel selectors. The output channel L OUT is sent to the input of the LIMIT limiter, which limits its amplitude as explained above. Tenth method of implementation

[0151] THE figures 21 And 22illustrate the application of the tenth implementation method to the first and fifth implementation methods respectively. Of course, the tenth implementation method could be applied similarly to the other implementation methods.

[0152] According to this implementation method, RMS processing can be produced entirely by digital convolution. This involves recording a test signal through the process described in the figure 3 and to apply this sonic signature in real time to a stereophonic input signal, using methods well known to those skilled in the art. The limiter of the ninth implementation can also be reproduced, provided it is a dynamic convolution. This tenth implementation can be considered a model or emulation of the RMS processing according to the invention.

[0153] However, any method of modeling or emulating RMS processing has the disadvantage of using more resources than those required by the optimized processing for optimal numerical processing such as that illustrated by the figure 9 .

[0154] This convolution modeling can reproduce RMS processing (which makes the out-of-phase content of a stereophonic signal audible in monophonic listening with the associated advantages which are discussed in this document) and also reproduce other situations (stereo, mono, and the other modes mentioned above).

[0155] There Figure 21 illustrates a method 100" corresponding to the application of mode 100 according to the tenth implementation mode. It includes Model L and Model R models of the processing applied to the input channels L IN and R IN, to obtain the input signals of the Sum2(L) and Sum2(R) blocks of the third stage St3 of implementation mode 100.

[0156] There Figure 22 This illustrates a 500'' process corresponding to the application of mode 500 according to the tenth implementation mode. It includes Model L and Model R models of the processing applied to the input channels L IN and R IN, to obtain the input signals of the Sum2(L) block of the third stage St3 of implementation mode 500.

[0157] There figure 23 summarizes the process of monophonic listening, according to the invention, by a user from a stereophonic signal.

[0158] In variant 23 (A), the two-channel stereophonic S STER signal, L IN and R IN, is supplied as input to an S RMS GEN system for generating an S RMS output signal suitable for enhanced monophonic reproduction. The S RMS GEN system is configured to implement any one of the processes illustrated by the figures 3 à 15 And 19 has 22The S RMS GEN system generates a monophonic S RMS output signal, which is transmitted wirelessly or via a wired connection to a MonoRep monophonic sound reproduction system. Unlike stereophonic systems or systems with more than two listening channels, such a monophonic sound reproduction system is configured to distribute one or more signals carrying the same information to the speaker(s) in the system. This can be a single-speaker system or a multi-speaker system. In the latter case, all speakers simultaneously active in the MonoRep monophonic sound reproduction system are fed the same monophonic S RMS signal. In addition to the speaker(s), this system may include, for example, one or more filtering or amplification devices upstream of the speaker(s).The MonoRep monophonic sound reproduction system emits Loud RMS sound waves which form a monophonic sound signal, that is to say it is based on the sole monophonic SRMS signal, so that the user of the system benefits from monophonic listening.

[0159] Variant 23(B) corresponds to the application of the ninth implementation mode to variant 23(A). This variant 23(B) differs from variant 23(A) in that the two input channels, L IN and R IN, have their amplitude reduced by means of an amplitude reduction device Red, which can consist of two Red Mod modules. The resulting two-channel signal S STER.Red is supplied as input to the S RMS GEN system instead of the S STER signal in variant S3(A). Another point of differentiation between variant 23(B) and variant 23(A) is that the output S RMS signal passes through a limiter, preferably with a threshold of 0 dB FS, to form a limited output signal SRMS.Lim, which is sent to the MonoRep monophonic sound reproduction system. General considerations

[0160] In this description, a monophonic audio signal is defined as one that does not contain the information necessary to produce a spatially defined sound scene. This would be the case for a signal with only one channel or a signal with multiple channels containing the same information, for example, two identical channels.

[0161] Also, in this description, the term "channel" refers to a signal that can be considered in association with another channel. This is the case for a stereophonic signal composed of two channels.

[0162] THE figures 9 à 14These illustrate situations in which signals are generated from a subtraction LIN - RIN, where the right channel RIN is subtracted from the left channel LIN, LIN and RIN representing the left and right channels of a stereophonic signal, respectively. This is an arbitrary choice used for explanatory purposes; a subtraction RIN - LIN could just as easily be used.

[0163] The first, second, third, and fourth implementation methods can be implemented in an electronic device via software and managed, for example, by means of a dedicated software application. Alternatively, they can be implemented in an electronic device via hardware, using discrete electronic components or an electronic chip integrating all the functions necessary for implementing the method.

[0164] The summation function blocks mentioned in this description are configured to sum the signals provided to them as inputs. If two signals are provided to these blocks, they output their sum. If only one signal is provided, that signal is output unchanged from the function block.

[0165] The stereophonic input signal under consideration can be a digital signal or an analog signal, the implementation of the process being of course adapted to the nature of the signal.

[0166] In this document, the figures are not necessarily to scale. Some features and components may be shown exaggerated relative to other components or in a somewhat schematic form, and some details of conventional elements may not be shown for the sake of clarity and conciseness.

[0167] Regarding an element of a figure defined within a given embodiment, reference may be made to the description of the characteristics and advantages of that same element in another embodiment. Similarly, unless otherwise indicated, the described embodiments may be combined, as a person skilled in the art will understand.

[0168] Of course the invention is not limited to the implementation methods described and alternative embodiments can be made without departing from the scope of the invention as defined by the claims.

Claims

1. Method (100) for processing a stereophonic input signal (S STER ) including a left channel (L IN ) and a right channel (R IN ), comprising the steps of: - generating a central channel (Mid) and a first lateral channel (Side(L), Side(R)), the central channel being representative of a sum of the left channel (L IN ) and the right channel (R IN ) and the first side channel (Side(L), Side(R)) being representative of a subtraction of one from the left channel (L IN ) and the right channel (R IN ) of the other of the left channel (L IN ) and the right channel (R IN ); - phase shift by 90° in absolute value in a first direction the first side channel (Side(L)) to obtain a first phase-shifted side channel (Sh_Side(L)); and - generate a first monophonic output channel (L OUT) representative of a sum of the central channel (Mid) and the first phase-shifted side channel (Sh_Side(L)), the method further comprising a step of sending the first monophonic output channel (L OUT ) to a sound reproduction system (Spk, SpkEns) configured to generate a monaural listening sound signal by a user.

2. The method according to claim 1, generating an output signal (S RMS ) which consists of the first monophonic output channel (L OUT ) and therefore a reduced number of channels compared to the stereophonic input signal (S Ster ).

3. The method according to claim 1, further comprising the steps of: - generating a second side channel (Sh_Side(R)) of opposite polarity to the first side channel (Sh_Side(L)); - shifting the second side channel (Sh_Side(R)) in phase by 90° in absolute value in a second direction opposite to the first direction to obtain a second phase-shifted side channel (Sh_Side(R)); - generating a second monophonic output signal (R OUT ) representative of a sum of the center channel (Mid) and the second phase-shifted side channel (Sh_Side(R)), in which the first monophonic output channel (L OUT ) and the second monophonic output channel (R OUT ) are identical by construction and form a monophonic output signal (S RMS ) to two channels, the method further comprising a step of sending the monophonic output signal (S RMS) two-channel to the sound reproduction system (Spk, SpkEns) configured to generate a monaural listening sound signal by a user.

4. The method according to any one of claims 1 to 3, in which the first side channel (Side(L)) is representative of a subtraction in which the right channel (R) is subtracted. IN ) of the left channel (L IN ).

5. The method according to any one of claims 1 to 4, further comprising a reduction (Red) of the amplitudes of the left channel (L IN ) and the right channel (R IN ) and a limitation (LIMIT) of the monophonic output channel (L OUT ) before it is sent in limited form (S RMS.Lim ) to the sound reproduction system (Spk, SpkEns).

6. The processing method according to any one of claims 1 to 5, the method being implemented by means of an electronic monophonic sound reproduction device (SMART, SpkMONO , Sys, Cent) configured to implement the process steps by software or hardware.

7. A computer program comprising instructions which, when executed by a digital computer, cause the computer to apply to a stereophonic input signal (S STER ) the steps of the method according to any one of claims 1 to 5.

8. A computer-readable medium comprising instructions which, when executed by a digital computer, cause the computer to apply to a stereophonic input signal (S STER ) the steps of the method according to any one of claims 1 to 5.

9. Electronic chip (Chp) configured to implement the steps of the method according to any one of claims 1 to 5.

10. Electronic device (SMART, Spk MONO, Sys, Cent) comprising at least one loudspeaker (Spk), configured to implement the steps of the method according to any one of claims 1 to 5 by software or by hardware.

11. The device of claim 10, configured to switch from a first operating mode in which the device generates the first monophonic output channel (L OUT ) to a second mode of operation in which the device generates a monophonic output channel representative of the central channel (S MONO ).

12. The device according to claim 11, comprising a control interface (Bt3, Bt3') configured to switch the device from the first operating mode to the second operating mode.

13. The device according to claim 11 or 12, comprising a control interface (Bt2, Bt2') configured to switch the device from the first operating mode to an operating mode in which the first monophonic output channel (S OP ) is representative of the first phase-shifted side channel (Sh_Side(L)) excluding the center channel (Mid).

14. The device according to any one of claims 10 to 13, chosen from a smartphone (SMART), a monophonic speaker (Spk MONO ), a system (Sys) for broadcasting sound recordings, and a centralization module (Cent) configured to distribute the first monophonic output channel (L OUT ) to a set (SpkEns) of speakers (Spk).

15. A monophonic listening method comprising (i) implementing the processing method according to any one of claims 1 to 6 and (ii) emitting a monophonic sound signal (Loud RMS) based on the first monophonic output channel, by means of the sound reproduction system (Spk, SpkEns) configured to generate a monophonic listening sound signal.

Images