WO2015144708A1

WO2015144708A1 - Acoustic apparatus comprising at least one electroacoustic microphone, an osteophonic microphone and means for calculating a corrected signal, and associated item of headwear

Info

Publication number: WO2015144708A1
Application number: PCT/EP2015/056261
Authority: WO
Inventors: Patrick Jean François ROBUCHON; Eric Bernard Jacques CLOWEZ; Julie Marie Anne ROSIER
Original assignee: Elno
Priority date: 2014-03-25
Filing date: 2015-03-24
Publication date: 2015-10-01
Also published as: EP3123740A1; EP3123740B1; FR3019422B1; FR3019422A1

Abstract

This acoustic apparatus (10) comprises: a first microphone (12), the first microphone (12) comprising a principal electroacoustic transducer able to receive the acoustic sound waves of a sonic signal originating from the vocal cords and to convert said acoustic waves into a first electrical signal; a second microphone (14), the second microphone (14) comprising an osseous mechanical excitation transducer able to receive by osseous conduction vibratory oscillations of said sonic signal and to convert said vibratory oscillations into a second electrical signal; and means (48) for computing a corrected electrical signal depending on the first electrical signal and the second electrical signal, the corrected electrical signal being able to be delivered as output from the acoustic apparatus (10). The acoustic apparatus (10) furthermore comprises a noise reducing device (20), the noise reducing device (20) being connected at the output of the principal electroacoustic transducer in order to decrease the noise in the first electrical signal and the computing means (48) being connected, on the one hand, to the output of the noise reducing device (20), and on the other hand, to the output of the osseous mechanical excitation transducer.

Description

Acoustic apparatus comprising at least one electroacoustic microphone, an osteophonic microphone and means for calculating a corrected signal, and associated head equipment The present invention relates to an acoustic apparatus comprising:

a first microphone, the first microphone comprising a main electroacoustic transducer capable of receiving acoustic sound waves from a sound signal originating from the vocal cords and transforming said acoustic waves into a first electrical signal,

a second microphone, the second microphone comprising a bone mechanical excitation transducer able to receive, by bone conduction, vibratory oscillations of said sound signal and to transform said vibratory oscillations into a second electrical signal, and

means for calculating an electric signal corrected as a function of the first electrical signal and the second electrical signal, the corrected electrical signal being able to be delivered at the output of the acoustic device.

The present invention also relates to an operator head equipment comprising a protective helmet and such an acoustic device.

US 5,692,059 A discloses an acoustic apparatus of the aforementioned type for insertion into the ear canal of the outer ear. The acoustic apparatus includes an aerial microphone capable of receiving acoustic sound waves and transforming them into an electrical signal, and a vibration sensor, the vibration sensor serving to eliminate background noise. The sounds picked up by the vibration sensor are a mixture of the low frequencies of the sound corresponding to the voice of the user and a small amount of background noise. The acoustic apparatus further comprises an electronic circuit capable of calculating an electric signal corrected according to the electrical signals respectively from the overhead microphone and the vibration sensor.

The document US Pat. No. 7,283,850 B2 also describes an acoustic device of the aforementioned type. The device is a mobile phone including an osteophonic microphone for resting against a lateral side of the skull when the user positions the phone near his ear. The mobile phone further comprises an overhead microphone adapted to receive acoustic sound waves corresponding to the speech signal of the user and transforming said waves into an electrical signal. The osteophonic microphone is used to eliminate noise in the speech signal received by the overhead microphone. The osteophonic microphone is used especially in low frequencies.

However, the electrical signal corresponding to the audio signal of the voice of the user, able to be delivered by such acoustic devices, is not optimal.

The object of the invention is therefore to provide an acoustic device for further reducing the noise in the electrical signal corresponding to the voice of the user, to improve the quality of the signal delivered.

For this purpose, the subject of the invention is an acoustic apparatus of the aforementioned type, in which the acoustic apparatus further comprises a noise reduction device, the noise reduction device being connected at the output of the main electroacoustic transducer to reduce the noise. noise in the first electrical signal and the calculation means being connected, on the one hand, at the output of the noise reduction device, and on the other hand, at the output of the bone mechanical excitation transducer.

According to other advantageous aspects of the invention, the acoustic apparatus comprises one or more of the following characteristics, taken separately or in any technically possible combination:

the noise reduction device comprises a third microphone and first means for determining a first corrected intermediate signal, the third microphone comprising a secondary electroacoustic transducer adapted to receive acoustic sound waves from ambient noise and to transform said waves; acoustic signals into a third electrical signal, the first corrected intermediate signal being a function of the first electrical signal and the third electrical signal, and the corrected electrical signal then being a function of the first corrected intermediate signal and the second signal;

the noise reduction device comprises digital noise processing means adapted to receive the first electrical signal and to deliver a first corrected intermediate signal, the first corrected intermediate signal having a signal-to-noise ratio greater than the signal-to-noise ratio of the first signal;

the noise reduction device further comprises second means for determining a second corrected intermediate signal as a function of the first corrected intermediate signal and the second signal, the second determination means being connected at the output of the first determination means or else at the output of the digital noise processing means;

the acoustic apparatus further comprises first filtering means connected between the noise reduction device and the calculation means, the first filtering means being adapted to decrease the amplitude of the signal for frequencies lower than a first predetermined frequency;

the acoustic apparatus further comprises second filtering means connected between the bone mechanical excitation transducer and the calculating means, the second filtering means being capable of reducing the amplitude of the signal for frequencies greater than a second predetermined frequency; ;

the acoustic apparatus further comprises switching means adapted to switch between a first configuration in which the electrical signal to be delivered at the output of the acoustic device is a function of the corrected electrical signal from the calculation means and a second configuration in which the electrical signal to be delivered at the output of the acoustic device is only a function of the second electrical signal;

- The switching means are controllable means according to a control law, and the control law depends on a signal-to-noise ratio which is a function of the first electrical signal and the second electrical signal;

- The acoustic device further comprises two side acoustic modules resting on the lateral flanks of the skull and adapted to transmit a sound signal to the auditory nerve.

The invention also relates to an operator head equipment comprising a protective helmet and an acoustic device as defined above.

These features and advantages of the invention will become apparent on reading the description which follows, given solely by way of nonlimiting example, and with reference to the appended drawings, in which:

FIG. 1 is an overall perspective view of an acoustic apparatus according to a first embodiment of the invention, the acoustic apparatus comprising first and second overhead microphones, an osteophonic microphone, and a clean processing unit; to deliver a corrected electrical signal from the electrical signals from the different microphones,

FIG. 2 is a diagrammatic sectional view of the first air microphone of FIG. 1,

FIG. 3 is a diagrammatic sectional view of a protective case inside which the osteophonic microphone, the processing unit and the second air microphone of FIG. 1 are arranged,

FIG. 4 is a schematic representation of a processing of the signal implemented by the processing unit to deliver the corrected signal from the signals from the microphones, FIG. 5 is a view similar to that of FIG. 4 according to a first variant embodiment,

FIG. 6 is a block diagram of digital noise processing means implemented in this first variant,

FIGS. 7 to 12 are views similar to that of FIG. 4 according to other embodiments,

FIG. 13 is a view similar to that of FIG. 1 according to a second embodiment of the invention, and

- Figure 14 is a view similar to that of Figure 1 according to a third embodiment of the invention.

In FIG. 1, an acoustic apparatus 10 comprises a first microphone 12, also called the first air microphone, capable of receiving acoustic sound waves and transforming them into a first electrical signal E1, and a second microphone 14, also called an osteophonic microphone, adapted to receive vibratory oscillations by bone conduction and to transform them into a second electrical signal E2.

The acoustic apparatus 10 comprises a processing unit 16 disposed inside a protective casing 18, the processing unit 16 being able to deliver a corrected electrical signal Se in particular as a function of the first electrical signal E1 and the second electrical signal E2.

The acoustic apparatus 10 comprises a noise reduction device 20, the noise reduction device 20 being connected at the output of the first microphone 12 to reduce the noise in the first signal E1, and the processing unit 16 is then connected, on the one hand, at the output of the noise reduction device 20, and on the other hand, at the output of the second microphone 14.

The acoustic apparatus 10 also comprises two lateral acoustic modules 22, an upper arch 24, a rear arch 26 connecting the acoustic modules and a connection cable 28, the connection cable being equipped at its end with a connector, not shown. .

The first microphone 12, visible in Figure 2, comprises a main electroacoustic transducer 32 adapted to receive acoustic sound waves of a sound signal from the vocal cords and transforming said acoustic waves into the first electrical signal E1.

In the embodiment of FIG. 2, the first microphone 12 comprises a protuberance 34, for example integral with the protective housing 18 and extending in a longitudinal direction X. The protrusion 34 comprises a cavity 36 receiving the main electroacoustic transducer and a cylindrical duct 38 extending longitudinally between the cavity 36 and a free end 40 of the protrusion, the free end 40 being located opposite the protective housing 18. The longitudinal direction X is oriented substantially towards the mouth of the user when the acoustic device 10 is positioned on the head of the user. The longitudinal direction X forms, for example, an angle between 45 ° and 75 °, preferably equal to 60 °, with a horizontal plane in which the rear arch 26 is arranged when it is positioned against the skull of the user.

The first microphone 12 additionally comprises an element 42 for holding the main electroacoustic transducer inside the cavity 36. The holding element 42 is able to dampen any mechanical vibrations of the main electroacoustic transducer 32, and is produced, for example, rubber or silicone.

The second microphone 14 comprises a bone mechanical excitation transducer 44, visible in FIG. 3, disposed in the protective casing 18, and adapted to receive, by bone conduction, in particular through a corresponding bone of the skull, the vibratory waves of the sound signal. from the vocal cords of the user and to transform it into the second electrical signal E2. The second microphone 14 is electrically connected to the processing unit 16.

The processing unit 16 is, for example, embodied as an electronic circuit 46, as shown in FIG. 3. The electronic circuit 46 comprises, for example, a processor 47A and a memory 47B associated with the processor.

The processing unit 16 comprises means 48 for calculating a corrected electrical signal Se as a function of the first electrical signal E1 and the second electrical signal E2, the corrected electrical signal Se being able to be delivered at the output of the acoustic device 10, possibly after being amplified by an amplifier 49. The calculation means 48 are, for example, made in the form of a calculation software capable of being stored in the memory 47B of the processing unit.

The processing unit 16 additionally comprises the amplifier 49 adapted to amplify the corrected signal Se and deliver an amplified signal S. at the output of the acoustic device 10.

The housing 18 has a wall 50 for bearing on an area of the skull and an outer protective shell 52, the protective shell 52 for protecting both the treatment unit and the bone mechanical excitation transducer of the second microphone, and being located on the opposite side to the bearing wall 50, as shown in FIG. The noise reduction device 20 is connected at the output of the main electroacoustic transducer 32 to reduce the noise in the first signal E1.

In the embodiment of Figures 1, 3 and 4, the noise reduction device 20 comprises a third microphone 54 and first means 56 for determining a first corrected intermediate signal Intl. The third microphone 54, also called second air microphone, comprises a secondary electroacoustic transducer 58 adapted to receive acoustic sound waves of ambient noise and to transform said acoustic waves into a third electrical signal E3. The third microphone 54 is omnidirectional for capturing acoustic waves coming from several directions. The first determination means 56 are for example made in the form of a first determination software able to be stored in the memory 47B of the processing unit.

The first corrected intermediate signal Int1 is a function of the first electrical signal E1 and the third electrical signal E3, the corrected electrical signal Se then being a function of the first corrected intermediate signal Int1 and the second signal E2.

The first corrected intermediate signal Int1 is, for example, equal to the subtraction of the third electrical signal E3 from the first electrical signal E1, and then verifies the following equation:

lnt1 = E1 - E3 (1)

In addition, a filter, not shown, is used in order to phase the signals E1 and E3 before subtracting the signal E3 from the signal E1 according to equation (1).

In the embodiment of FIG. 4, the corrected electrical signal Se is equal to the subtraction of the second signal E2 from the first corrected intermediate signal Int1, and then verifies the following equation:

Se = Int1 - E2 (2)

The corrected electrical signal Is thus verified the following equation:

Sc = E1 - E3 - E2 (3)

In addition and analogously, a filter, not shown, is used to phase the signals E1, E2 and E3 before subtracting the signals E2 and E3 signal E1 according to equation (2).

Each acoustic module 22, visible in FIG. 1, comprises a support plate 60 on a skull lateral flank and a bone mechanical excitation transducer 62. Each plate 60 comprises a support plate 64 adapted to bear on the skull above an ear, the support plate 64 having a clearance passage of the ear in its lower part. The protective housing 18 is, for example, fixed in the rear part of the support plate 64 which is intended to be positioned above the ear right. The support plates 64 are made of plastic material and injection molded.

Each acoustic module 22 comprises a hinge 66 provided between the support plate 60 and the transducer 62. A spring, not shown, equips the hinge 66 and is capable of ensuring a return in rotation, around the hinge 66, of the transducer 62 relative to the plate 60 to a rest position.

The upper arch 24, also called headband, is of adjustable length and adapted to be positioned on the top of the head. In the example of Figure 1, the upper bow 24 is in the form of a strap of adjustable length.

The rear arch 26, made of a rigid material, is a mechanical holding bar of the osteophonic microphone 14 and each acoustic module 22 resting on a corresponding lateral side of the skull. The holding bar 26 is of adjustable length, and able to be positioned under the bone of the rock behind the head, near the nape of the neck.

The bone mechanical excitation transducer 44 is an accelerometer, preferably a one-axis accelerometer, oriented substantially in a direction normal to the bearing wall 50, in order to allow a homogeneous reproduction of the vibratory wave transmitted by bone conduction. according to the different users. The bone mechanical excitation transducer 44 is oriented towards the inside, that is to say towards the skull of the user when the acoustic apparatus 10 is positioned on the skull.

The calculation means 48 are connected, on the one hand, at the output of the noise reduction device 20, and on the other hand, at the output of the bone mechanical excitation transducer 44.

The wall 50 is in the form of a skin in contact with the accelerometer 44 of the second microphone. The skin 50, visible in section in FIG. 3, has a value hardness of between 45 and 85 A shores, so that the skin 50 is of mechanical impedance substantially equal to the mechanical impedance of the corresponding zone of the skull. The value of the hardness of the skin 50 is preferably between 55 and 80 Shore A, more preferably between 65 and 75 Shore A. The thickness E of the skin 50 is between 0.4 mm and 0.6 mm. mm, preferably equal to 0.5 mm. The skin 50 is made of synthetic rubber, for example rubber based on polychloroprene.

The outer shell 52 has a U-shaped cross section, and is made of plastic.

The secondary electroacoustic transducer 58 is fixed inside the outer shell 52 of the protective casing, and is adapted to receive acoustic waves. ambient noise via an orifice 68 formed through said outer shell 52, as shown in Figure 3.

The secondary electroacoustic transducer 58 is oriented substantially in the same direction as the bone mechanical excitation transducer 44, but in the opposite direction. In other words, the secondary electroacoustic transducer 58 is oriented substantially in a normal direction to the support wall 50 and outwards, that is to say in the opposite direction to the user's skull when the acoustic device 10 is positioned on the skull.

The bone mechanical excitation transducer 62 of each acoustic module 22 comprises an emissive element, not shown, able to transform an electrical signal received into vibratory waves representative of the sound signal and to transmit them to the auditory nerve by bone conduction.

Thus, a user of the acoustic apparatus 10 begins by positioning the apparatus 10 on his head, by placing the strap 24 on top of his skull, the rear arch 26 behind his head, near his neck and the turntables. 60 support on the lateral flanks of his skull, above each respective ear. The clearance formed in the lower part of the support plates 64 ensures good ergonomics of the plate 60 relative to the upper part of the flag of the ear.

After positioning the apparatus 10 on his head, the user adjusts the upper bow 24 on his head, then the rear arch 26 on his neck. The adjustment of the position of each bone mechanical excitation transducer 62 on the corresponding temple takes place naturally and automatically by each articulation 66.

When the acoustic apparatus 10 is positioned in this manner on the head of the user, the first microphone 12 picks up a sound signal comprising both a component corresponding to the voice of the user, a component corresponding to the ambient noise, and than an osteophonic component. The second microphone 14, also called osteophonic microphone, captures only the osteophonic component of the sound signal, and the third microphone 54 essentially captures, by its orientation, the component corresponding to the ambient noise.

According to equation (3), therefore, the corrected electrical signal Se essentially contains the only component corresponding to the voice of the user, since the subtraction of the third electrical signal E3 makes it possible to eliminate the component corresponding to the ambient noise. , then that the subtraction of the second electrical signal E2 makes it possible to eliminate the osteophonic component of the sound signal.

In other words, with the acoustic apparatus 10 according to the invention, the component corresponding to the ambient noise is eliminated from the signal delivered at the output of the apparatus acoustic 10, whereas with the acoustic apparatus of the state of the art this component corresponding to the ambient noise is not attenuated.

It is thus conceivable that the acoustic device 10 according to the invention can further reduce the noise in the electrical signal corresponding to the voice of the user, to improve the quality of the signal delivered by the acoustic device.

FIG. 5 illustrates a first variant of this first embodiment for which the elements similar to those which have been previously described are identified by identical references, and are not described again.

According to this first variant embodiment, the noise reduction device 20 comprises means 70 for digital noise processing, and does not include the third microphone or the first determination means.

The digital noise processing means 70 are adapted to receive the first electrical signal E1 and to output the first corrected intermediate signal Int1, the first corrected intermediate signal Int1 then having a signal-to-noise ratio higher than the signal-to-noise ratio of the first signal E1.

The digital noise processing means 70 are for example made in the form of digital noise processing software capable of being stored in the memory 47B of the processing unit.

In the embodiment of FIG. 6, the digital noise processing means 70 are suitable for implementing a noise reduction algorithm based on Wiener filtering and on extended spectral subtraction. The digital processing means 70 comprise an input 70A, an output 70B, a Wiener filter 71 connected to the input 70A, and an adder 72 connected to the output of the Wiener filter 71 via a first amplifier 73. The digital processing means 70 comprise a first inverter 74 connected at the output of the adder 72 and a second amplifier 75 connecting the output of the first inverter 74 to the adder 72. The digital processing means 70 comprise a first adder 76 connected to the input 70A and at the output of the adder 72, and a second inverter 77 connected between the output of the first adder 76 and the Wiener filter 71. The digital processing means 70 comprise a second summer 78 connected to the input 70A and at the output of the Wiener filter 71 via a third amplifier 79, the output of the second summer 78 being connected to the output 70B of the digital processing means 70.

The digital processing means 70 are adapted to receive at the input 70A at time p the module of a noise-containing speech signal, the noisy speech signal at the instant p being denoted by X _p . The Wiener filter 71 has a transfer function H _p and is capable of outputting the noise spectrum module estimated at the instant p, the spectrum noise estimated at time p being noted N _p . The first amplifier 73 has a gain equal to 1 -λ, and the second amplifier 75 has a gain equal to λ, where λ is a coefficient between 0 and 1. The signal delivered at the output of the adder 72 represents an average of the spectral estimates of the noise, denoted by N _p , and verifies the following equation

The first inverter 74 makes it possible to obtain at output the noise spectrum module estimated at l, from the noise spectrum module estimated at the instant p

The signal delivered at the output of the first adder 76 corresponds to the modulus of an average of the estimates of the speech at the instant p, the mean of the estimates of the speech at the instant p being noted S _p and verifying the following equation: r¾ = | - ^ (5)

The second inverter 77 makes it possible to obtain at the output the modulus of the average of the estimates of the speech at the previous instant p-1 from the modulus of the average

speech estimates at the instant p received at the input.

The third amplifier 79 has a gain equal to K, and the signal delivered at the output of the second adder 78 corresponds to the modulus of the estimated speech signal at the instant p, the estimated speech signal at the instant p being denoted S _p and checking the following equation:

(6)

The module of the speech signal estimated at time p is delivered at the output 70B of the digital noise processing means 70, and corresponds to the first corrected intermediate signal Int1 which has a signal-to-noise ratio higher than the signal-to-noise ratio of the first signal. signal E1 received at the input of digital noise processing means 70.

In the embodiment of FIG. 5, the corrected electrical signal Se is equal to the subtraction of the second signal E2 from the first corrected intermediate signal Int1, and then verifies the equation (2) described above. According to equation (2), the corrected electrical signal Se therefore essentially contains the only component corresponding to the voice of the user, since the digital noise processing makes it possible to greatly reduce the component corresponding to the ambient noise. , then that the subtraction of the second electrical signal E2 makes it possible to eliminate the osteophonic component of the sound signal.

In other words, with the acoustic apparatus 10 according to the invention, the component corresponding to the ambient noise is greatly reduced in the signal delivered at the output of the acoustic apparatus 10, whereas with the acoustic apparatus of the state of the technical this component corresponding to the ambient noise is not attenuated.

FIG. 7 illustrates a second variant of this first embodiment for which the elements similar to those which have been previously described are identified by identical references, and are not described again.

According to this second variant embodiment, the noise reduction device 20 comprises the third microphone 54 and the first determination means 56. The noise reduction device 20 does not include digital noise processing means.

According to this second variant embodiment, the acoustic apparatus 10 further comprises first filtering means 80 connected between the noise reduction device 20 and the calculation means 48, the first filtering means 80 being able to reduce the amplitude. signal for frequencies lower than a first predetermined frequency F1. In other words, the first filtering means 80 form a high-pass filter with a cut-off frequency equal to the first predetermined frequency F1.

The first filtering means 80 are for example made in the form of a first filtering software capable of being stored in the memory 47B of the processing unit. The first predetermined frequency F1 has a value, for example between 800 Hz and 1600 Hz, preferably substantially equal to 1200 Hz.

In addition, according to the second variant embodiment, the acoustic apparatus 10 comprises second filtering means 82 connected between the bone mechanical excitation transducer of the second microphone 14 and the calculating means 48, the second filtering means 82 being able to reduce the amplitude of the signal for frequencies higher than a second predetermined frequency F2. In other words, the second filtering means 82 form a low-pass filter with a cut-off frequency equal to the second predetermined frequency F2.

The second filtering means 82 are for example made in the form of a second filtering software capable of being stored in the memory 47B of the control unit. treatment. The second predetermined frequency F2presents a value, for example between 400 Hz and 1200 Hz, preferably substantially equal to 800 Hz.

In the embodiment of FIG. 7, the corrected electrical signal Se is then equal to the addition of a low frequency part of the second signal E2 and of a high frequency part of the first corrected intermediate signal Int1, and then verifies the following equation:

Sc = Int1 HF + E2 _BF (7)

The corrected electrical signal Is thus verified the following equation:

Sc = (E1-E3) HF + _E2bf (8)

As a variant, the corrected electrical signal Se is then equal to the subtraction of a low frequency part of the second signal E2 and of a high frequency part of the first corrected intermediate signal Int1, and then verifies the following equation:

Sc = Int1 HF - E2 _BF (9)

According to this variant, the corrected electrical signal Se then satisfies the following equation:

Sc = (E1 - E3) HF - E2 _B F (10)

In addition, a filter, not shown, is used to phase the signals E1, E2 and E3 before subtracting the high frequency part of the signal E3 from the high frequency part of the signal E1, and then adding the result of this subtraction from the low frequency part of the signal E2 according to equation (8).

Subsequently, the term "low frequency part", the signal frequencies below the second frequency F2, that is to say those that have not been attenuated by the second filtering means 82. Similarly , the term "high frequency part", the signal frequencies greater than the first frequency F1, that is to say those that have not been attenuated by the first filtering means 80.

From the foregoing, as well as equation (8), the corrected electrical signal Se thus essentially contains the high frequency portion of the component corresponding to the user's voice and the low frequency portion of the osteophonic component of the signal. sound, since the subtraction of the third electrical signal E3 eliminates the component corresponding to the ambient noise from the first signal E1. In addition, the signal from the first filtering means 80 comprises a reduced osteophonic component, since the osteophonic component of the sound signal lies mainly in the low frequency domain.

According to this second variant embodiment, the quality of the signal delivered by the acoustic apparatus 10 according to the invention is further improved since the low frequency part of the osteophonic component offers better intelligibility of the signal. sound from the vocal cords as the low frequency part of the component corresponding to the voice, especially in case of whispering of the user. Those skilled in the art will also observe that when the user is speaking normally, the low frequency part of the osteophonic component is substantially identical to the low frequency part of the component corresponding to the voice.

In other words, with the acoustic apparatus 10 according to this second variant embodiment, the component corresponding to the ambient noise is eliminated from the signal delivered at the output of the acoustic apparatus 10, while offering a better restitution of the signal in the event of a whispering of the sound. 'user.

FIG. 8 illustrates a third variant of this first embodiment for which the elements similar to those which have been previously described are marked with identical references, and are not described again.

According to this third variant embodiment, the noise reduction device 20 comprises the digital noise processing means 70, and has neither the third microphone nor the first determination means.

According to this third variant embodiment, the acoustic apparatus 10 furthermore comprises the first filtering means 80 connected between the noise reduction device 20 and the calculation means 48.

In addition, according to the third variant embodiment, the acoustic device 10 comprises the second filtering means 82 connected between the bone mechanical excitation transducer of the second microphone 14 and the calculation means 48.

In the embodiment of FIG. 8, the corrected electrical signal Se is then equal to the addition of the low frequency part of the second signal E2 and of the high frequency part of the first corrected intermediate signal Int1, and then verifies the equation (7) previously described.

According to equation (7) and the above, the corrected electrical signal Se therefore essentially contains the high frequency part of the component corresponding to the voice of the user and the low frequency part of the osteophonic component of the sound signal, since the digital noise processing makes it possible to strongly reduce the component corresponding to the ambient noise in the first signal E1.

The advantages of this third embodiment are similar to those of the second embodiment described above.

FIG. 9 illustrates a fourth variant of this first embodiment for which the elements similar to those which have been previously described are identified by identical references, and are not described again. According to this fourth variant embodiment, the noise reduction device 20 comprises the third microphone 54 and the first determination means 56. The noise reduction device 20 does not include digital noise processing means.

The acoustic apparatus 10 comprises the first filtering means 80 connected between the noise reduction device 20 and the calculation means 48. In addition, the acoustic apparatus 10 comprises the second filtering means 82 connected between the mechanical excitation transducer. bone of the second microphone 14 and the calculation means 48.

According to this fourth variant embodiment, the noise reduction device 20 further comprises second means 90 for determining a second corrected intermediate signal Int2 as a function of the first corrected intermediate signal Int1 and the second signal E2, the second determination means 90 being connected at the output of the first determination means 56.

The second determination means 90 are, for example, made in the form of a second determination software that can be stored in the memory 47B of the processing unit.

In the embodiment of FIG. 9, the corrected second intermediate signal Int2 is equal to the subtraction of the second signal E2 from the first corrected intermediate signal Int1, and then verifies the following equation:

Int2 = lnt1 - E2 (1 1)

The second corrected intermediate signal Int2 thus satisfies the following equation: lnt2 = E1 - E3 - E2 (12)

In addition, a filter, not shown, is used to phase the signals E1, E2 and E3 before subtracting the signals E2 and E3 to the signal E1 according to equation (12).

According to equation (3), the second corrected intermediate signal Int2 thus essentially contains the only component corresponding to the voice of the user, since the subtraction of the third electrical signal E3 makes it possible to eliminate the component corresponding to the noise ambient, and that the subtraction of the second electrical signal E2 eliminates the osteophonic component of the sound signal.

In the embodiment of FIG. 9, the corrected electrical signal Se is then equal to the addition of a low frequency part of the second signal E2 and of a high frequency part of the second corrected intermediate signal Int2, and then verifies the following equation:

Se = lnt2 _HF + E2 _BF (13) The corrected electrical signal Is thus verified the following equation:

Se = (E1-E3-E2) HF + E2 _BF (14)

In addition and in a similar way, a filter, not shown, is used to phase the signals E1, E2 and E3 before subtracting the high frequency part of the signals E2 and E3 from the high frequency part of the signal E1, then of add the result of this subtraction to the low frequency part of the signal E2 according to equation (1 1).

From the foregoing, as well as equation (14), the corrected electrical signal Se thus contains only the high frequency portion of the component corresponding to the voice of the user and the low frequency portion of the osteophonic component of the signal. sound, since the subtraction of the third electrical signal E3 makes it possible to eliminate the component corresponding to the ambient noise from the first signal E1, and the subtraction of the second electrical signal E2 makes it possible to eliminate the osteophonic component of the sound signal from the first signal E1; .

According to this fourth variant embodiment, the quality of the signal delivered by the acoustic apparatus 10 according to the invention is further improved since the high frequency part of the delivered signal comprises only the component corresponding to the voice of the user.

In other words, with the acoustic apparatus 10 according to this fourth variant embodiment, the component corresponding to the ambient noise is eliminated from the signal delivered at the output of the acoustic apparatus 10, while offering a better restitution of the signal in case of whispering of the sound. user and better intelligibility of the signal delivered in high frequencies.

FIG. 10 illustrates a fifth variant of this first embodiment for which the elements similar to those which have been previously described are marked with identical references, and are not described again.

According to this fifth variant embodiment, the noise reduction device 20 comprises the digital noise processing means 70, and does not include either the third microphone or the first determination means.

According to this fifth variant embodiment, the noise reduction device 20 further comprises the second means 90 for determining a second signal corrected intermediate Int2 as a function of the first corrected intermediate signal Int1 and the second signal E2, the second determination means 90 being connected at the output of the digital noise processing means 70.

In the embodiment of FIG. 10, the corrected electrical signal Se is then equal to the addition of the low frequency part of the second signal E2 and of the high frequency part of the second corrected intermediate signal Int2, and then verifies the equation (13) previously described.

According to equation (13) and the above, the corrected electrical signal Se thus contains the high frequency part of the component corresponding to the voice of the user and the low frequency part of the osteophonic component of the sound signal, being Since the digital noise processing makes it possible to strongly reduce the component corresponding to the ambient noise in the first signal E1 and the subtraction of the second electrical signal E2 makes it possible to eliminate the osteophonic component of the sound signal from the first signal E1.

The advantages of this fifth embodiment are similar to those of the fourth embodiment described above.

FIG. 11 illustrates a sixth variant of this first embodiment for which the elements similar to those which have been previously described are identified by identical references, and are not described again.

According to this sixth variant embodiment, the acoustic apparatus 10 further comprises switching means 95 able to switch between a first configuration in which the electrical signal S adapted to be delivered at the output of the acoustic apparatus 10 is a function of the electrical signal corrected Se comes from the calculation means and a second configuration in which the electrical signal S adapted to be delivered at the output of the acoustic device 10 is only function of the second electrical signal E2.

The switching means 95 are, for example, connected at the output, on the one hand, calculating means 48, and secondly, second filtering means 82, as shown in FIG.

In addition, the switching means 95 are controllable means according to a control law, and the control law depends on a first signal-to-noise ratio SNRi which is a function of the first electrical signal E1 and the second electrical signal E2.

The first signal-to-noise ratio SNRi checks, for example, the following equation: SNR, (15)

where Ern and Εη ₂ are the respective estimated energies of the first and second electrical signals E1 and E2.

Estimated energies En _! and In ₂ are calculated according to an average, such as an arithmetic mean, of the values of the first E1, and respectively second E2, electrical signals during a given time period.

The control law is such that the switching means 95 switch in the first configuration when the first signal to noise ratio SNF ^ has a value greater than a first threshold. The first threshold is for example equal to 15 dB.

In addition, the control law also depends on a second signal to noise ratio SNR ₂ which is a function of the first corrected intermediate signal Int1 and the second electrical signal E2.

The second signal-to-noise ratio SNR ₂ checks, for example, the following equation:

where En ₂ and En ₃ are the respective estimated energies of the second electrical signal E2 and the first corrected intermediate signal Int1.

A gap Δ is also defined as follows:

A = SNR ₂ - SNR ₁ (17)

The control law is then such that the switching means 95 also switch in the first configuration when the first signal-to-noise ratio SNRi has a value less than or equal to the first threshold and the difference Δ has a value greater than a second threshold . The second threshold is for example equal to 5 dB.

In the other cases, that is to say when the first signal-to-noise ratio SNRi has a value less than or equal to the first threshold and the difference Δ has a value less than or equal to the second threshold, the control law is such that the switching means 95 switch in the second configuration. The electrical signal S is then only dependent on the second electrical signal E2.

More generally, the switching means 95 are suitable for outputting an improved corrected signal Sca satisfying the following equation:

Sca = A x E2 _BF + B x Sc (18)

where A and B are weighting coefficients whose value depends on the first and second SNR ₁ signal-to-noise ratios _; SNR ₂

FIG. 12 illustrates a sixth variant of this first embodiment for which elements similar to those of the fifth variant which have been described previously, are marked with identical references, and are not described again.

According to this sixth variant embodiment, the noise reduction device 20 comprises the digital noise processing means 70 in place of the third microphone and the first means for determining the fifth variant.

Figure 13 illustrates a second embodiment for which the elements similar to those of the first embodiment described above, are identified by identical references, and are not described again.

According to the second embodiment, the second microphone 14 is not disposed in the protective housing 18, but is arranged in an additional box 100, the additional box 100 being connected to one of the two acoustic modules 22 by two arms The bone mechanical excitation transducer 44 of the second microphone is then disposed in the additional housing 100.

The additional box 100 is preferably intended to be applied in contact with the right side of the skull of the user, and the additional box 100 is then preferably connected to the acoustic module 22 right.

The second microphone 14 is then identical to the contact microphone described in document FR 2 945 905 A1. The arrangement of the second microphone 14 inside the additional box 100 is similar to the arrangement of the contact microphone in its protective case as described in the document FR 2 945 905 A1.

The operation of this second embodiment is identical to that of the first embodiment described above, and is not described again.

The advantages of this second embodiment are identical to those of the first embodiment described above, and are not described again.

FIG. 14 illustrates a third embodiment for which elements similar to those of the first embodiment described above, are marked with identical references, and are not described again.

According to the third embodiment, the first microphone 12 and the second microphone 14 are not arranged in the protective case 18, but are arranged in an additional box 200, the additional box 200 being connected to one of the two acoustic modules 22 by two connecting arms 202.

The main electroacoustic transducer 32 and the bone mechanical excitation transducer 44 are then each disposed in the additional housing 200.

The additional box 200 is preferably intended to be applied in contact with the right side of the skull of the user, and the additional box 200 is then preferably connected to the acoustic module 22 right, as shown in Figure 14. The second microphone 14 is identical to the contact microphone described in document FR 2 945 905 A1. The arrangement of the second microphone 14 inside the additional box 200 is similar to the arrangement of the contact microphone in its protective case as described in the document FR 2 945 905 A1.

The operation of this third embodiment is identical to that of the first embodiment described above, and is not described again.

The advantages of this third embodiment are identical to those of the first embodiment described above, and are not described again.

Whatever the embodiment and / or whatever the embodiment variant of the first embodiment, the upper arch 24, for example in the form of a strap, allows, because of its small thickness and its flexible material, the port a heavy helmet without discomfort on the top of the head. The rear arch 26, being positioned under the bone of the rock behind the head, also allows the wearing of all heavy combat helmets.

The mechanical support of the second microphone 14 and the two modules 22 in contact with the skull is provided by the rear bow 26, while the upper bow 24 has a holding role in position on the top of the head.

The acoustic device 10 is also suitable for use with a biker helmet, a helmet for a motor vehicle driver, a helmet for an armored vehicle, a fireman's helmet, a helmet for a security guard, a construction helmet, or even a helmet for an aircraft pilot.

Alternatively, the acoustic device 10 is a headphone for operator.

It is thus conceivable that the acoustic device 10 according to the invention can further reduce the noise in the electrical signal corresponding to the voice of the user, in order to improve the quality of the signal delivered.

Claims

1 .- Acoustic apparatus (10) comprising:

a first microphone (12), the first microphone (12) comprising a main electroacoustic transducer (32) adapted to receive acoustic sound waves from a sound signal originating from the vocal cords and to transform the acoustic waves into a first electrical signal ( E1),

a second microphone (14), the second microphone (14) comprising a bone mechanical excitation transducer (44) capable of receiving, by bone conduction, vibratory oscillations of said sound signal and of transforming said vibratory oscillations into a second electrical signal (E2) ,

means (48) for calculating a corrected electrical signal (Se) as a function of the first electrical signal (E1) and the second electrical signal (E2), the corrected electrical signal (Se) being able to be delivered at the output of the acoustic apparatus (10),

characterized in that the acoustic apparatus (10) further comprises a noise reduction device (20), the noise reduction device (20) being connected at the output of the main electroacoustic transducer (32) to reduce noise in the first electrical signal (E1) and the calculation means (48) being connected, on the one hand, at the output of the noise reduction device (20), and on the other hand, at the output of the bone mechanical excitation transducer (44). )

in that the noise reduction device (20) comprises a third microphone (54) and first means (56) for determining a first corrected intermediate signal (Int1), the third microphone (54) comprising a secondary electroacoustic transducer (58) adapted to receive acoustic sound waves from an ambient noise and to transform said acoustic waves into a third electrical signal (E3), the first corrected intermediate signal (Int1) being a function of the first electrical signal (E1) and the third electrical signal (E3), and the corrected electrical signal (Se) then being a function of the first corrected intermediate signal (Int1) and the second signal (E2), and

in that the noise reduction device (20) further comprises second means (90) for determining a second corrected intermediate signal (Int2) as a function of the first corrected intermediate signal (Int1) and the second signal (E2). the second determining means (90) being connected at the output of the first determining means (56).

2. Acoustic apparatus (10) comprising:

in that the noise reduction device (20) comprises digital noise processing means (70) adapted to receive the first electrical signal (E1) and to deliver a first corrected intermediate signal (Int1), the first corrected intermediate signal (Int1) having a signal-to-noise ratio higher than the signal-to-noise ratio of the first signal (E1), and

in that the noise reduction device (20) further comprises second means (90) for determining a second corrected intermediate signal (Int2) as a function of the first corrected intermediate signal (Int1) and the second signal (E2). the second determining means (90) being connected at the output of the digital noise processing means (70).

An acoustic apparatus (10) according to claim 1 or 2, wherein the acoustic apparatus (10) further comprises first filtering means (80) connected between the noise reduction device (20) and the means for calculation (48), the first filter means (80) being adapted to decrease the amplitude of the signal for frequencies lower than a first predetermined frequency (F1).

An acoustic apparatus (10) according to any one of the preceding claims, wherein the acoustic apparatus (10) further comprises second filtering means (82) connected between the bone mechanical excitation transducer (44) and the calculating means (48), the second filtering means (82) being adapted to reduce the amplitude of the signal for frequencies higher than a second predetermined frequency (F2).

An acoustic apparatus (10) as claimed in any one of the preceding claims, wherein the acoustic apparatus (10) further comprises switching means (95) adapted to switch between a first configuration in which the electrical signal (S) ) capable of being delivered at the output of the acoustic device (10) is a function of the corrected electrical signal (Se) issued by the calculation means and a second configuration in which the electrical signal (S) capable of being delivered at the output of the acoustic device (10) is only dependent on the second electrical signal (E2).

6. - acoustic device (10) according to claim 5, wherein the switching means (95) are controllable means according to a control law, and the control law depends on a signal-to-noise ratio which is a function of the first electrical signal (E1) and the second electrical signal (E2).

7. - acoustic device (10) according to any one of the preceding claims, wherein the acoustic apparatus (10) further comprises two lateral acoustic modules (22) resting on the lateral flanks of the skull and adapted to transmit a signal sound to the auditory nerve.

8. - Operator head equipment comprising a protective helmet, characterized in that it comprises an acoustic device (10) according to any one of the preceding claims.