Conversation Support System

Abstract

To provide a conversation support system which improves lowering of clearness of conversation voice even when a plurality on persons opposed to each other through a panel have conversations, and hence promotes smooth exchange of information among the plurality of persons.SOLUTION: A speaker module SPA is provided at an opening portion 12 of a screen panel 10. A microphone module Ma is provided so as to be symmetric relative to a plurality of persons opposed to each other through the screen panel, and the screen panel serves as a baffle plate of a speaker. The speaker module outputs voice of a first phase to a first side of the screen panel and outputs voice of a second phase reverse to the first phase to a second side at the opposite side of the panel. The voice of the first phase and the voice of the second phase are added and cancelled. Thus, the microphone takes voice signals from at least one side or both sides of the plurality of persons opposed to each other to output them from the speaker.SELECTED DRAWING: Figure 2

Description

[Technical Field]

[0001] The present invention relates to a support system for smoothly promoting conversation among a plurality of people separated by panels. [Background technology]

[0002] A known example of this type of system is the face-to-face communication device described in Japanese Patent Laid-Open Publication No. 2001-24803. This device is for providing a face-to-face communication device and method that allows smooth and error-free communication and work in two-way communication at ticket booths, ticket offices, etc., and is characterized in that, in a face-to-face communication device for communicating face-to-face through a wall that separates the inside and outside of a room, a space is provided in the wall, consisting of a transparent plate with sound holes and a transparent plate that blocks sound, and a speaker is installed in this space, or an acoustic duct is installed in the space and a speaker is installed in the acoustic duct. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] Japanese Patent Application Laid-Open No. 2001-24803 Summary of the Invention [Problem to be solved by the invention]

[0004] Due to the spread of infection caused by the new coronavirus, there are increasing cases of acrylic panels being installed at customer service counters in banks and public institutions to prevent the spread of droplets, and transparent panels being installed to separate the driver's seat from the rear passenger seats in taxis for security purposes.

[0005] While these simple measures achieve their intended purpose of preventing the spread of infectious diseases and crime, people are increasingly aware of the negative impact of reduced conversation intelligibility when conversing through a panel. Nowadays, as wearing masks has become mandatory, conversations through masks exacerbate this negative impact. This issue is not addressed in the prior art.

[0006] In order to solve the above problems, the present invention aims to provide a system that improves the decline in clarity of conversational voices even when multiple people facing each other converse through a panel, thereby supporting conversation so that information exchange between multiple people can proceed smoothly. [Means for solving the problem]

[0007] A first invention for achieving the above object is a conversation support system comprising: a panel that has a microphone and a speaker and separates a plurality of people; and a control circuit that processes sound captured by the microphone and outputs sound from the speaker, the system supporting conversation between a plurality of people facing each other via the panel, the panel having an opening in which the speaker is installed, the microphones being provided and fixed to the panel so as to be symmetrical with respect to the plurality of people facing each other via the panel, the panel serving as a baffle plate for the speaker, the speaker being provided with a baffle plate for the panel. The system is characterized in that audio of a first phase is output to a first side of a panel, and audio of a second phase, which is in the opposite phase to the first phase, is output to a second side of the panel opposite the first side, and the audio of the first phase and the audio of the second phase are preferably added together and cancelled out at the microphone position, and the microphone picks up audio signals from at least one side or both sides of the multiple people facing each other, preferably only this audio signal, and more preferably exclusively (avoiding audio from the speaker and prioritizing the speaker's voice), and outputs the audio from the speaker.

[0008] Furthermore, a second invention for achieving the above object is a conversation support system that supports conversation between a plurality of people facing each other via the panel, the system comprising: a panel that has a microphone and a speaker and separates a plurality of people; and a control circuit that processes sound captured by the microphone and outputs the sound from the speaker, the system comprising a plurality of speaker modules including a first speaker module and a second speaker module, the first speaker module being provided on a first side of the panel, a speaker module is provided on a second side of the panel, the first speaker module and the second speaker module face each other symmetrically across the panel, the control device causes the sound captured by the microphone to be output to the first side via the first speaker module and to the second side via the second speaker module, and the control device further causes the sound from the first speaker module and the sound from the second speaker module to be output in phase or in opposite phase to each other. [Effects of the Invention]

[0009] As described above, the present invention can improve the reduction in volume and clarity of conversational voices even when multiple people facing each other converse via a panel, thereby supporting conversations so that information exchange between multiple people can proceed smoothly. [Brief explanation of the drawings]

[0010] [Figure 1] FIG. 2 is a front view of the partition panel (panel). [Figure 2] FIG. 2 is a plan view of the panel, showing the arrangement of two people facing each other through the panel. [Figure 3] FIG. 2 is a front view of the speaker module viewed from above and below. [Figure 4] FIG. 2 is a side view of the speaker module. [Figure 5] FIG. 2 is a perspective view of a speaker module. [Figure 6A] FIG. 1 is a block diagram of a speaker unit in which two speaker modules are connected in parallel and face each other. [Figure 6B] FIG. 1 is a block diagram of a speaker unit in which two speaker modules are connected in series facing each other. [Figure 7A] FIG. 1 is a block diagram of a speaker unit in which two speaker modules are connected in parallel with their backs facing each other. [Figure 7B] FIG. 1 is a block diagram of a speaker unit in which two speaker modules are connected in series with their backs facing each other. [Figure 8] FIG. 3 is a block diagram of a conversation support system according to another embodiment different from that of FIG. 2. [Figure 9] FIG. 10 is a block diagram of a conversation support system according to yet another embodiment. [Figure 10A] FIG. 10 is a front view of a panel in a conversation support system according to yet another embodiment. [Figure 10B] FIG. 2 is a schematic diagram showing the cross-sectional structure of this panel. [Figure 10C] FIG. 2 is another schematic diagram showing the cross-sectional structure of this panel. [Figure 11A] FIG. 10 is a front view of a panel in a conversation support system according to yet another embodiment. [Figure 11B] FIG. 2 is a schematic diagram showing the cross-sectional structure of this panel. [Figure 12A] 1A and 1B are front views of a panel illustrating an embodiment of the arrangement and number of microphone modules on the panel. [Figure 12B] 10A and 10B are front views of the panel, illustrating other embodiments relating to the arrangement and number of microphone modules on the panel. [Figure 13A] 12B and 12C, and is a front view of a panel, showing an embodiment relating to the arrangement and number of microphone modules on the panel, which is different from that shown in FIGS. 12A and 12B. FIG. [Figure 13B] FIG. 12B is a plan view of a conversation support system including the panel of FIG. 12A. [Figure 14A]FIG. 13B is a front view of a panel according to another embodiment different from that shown in FIG. 13A. [Figure 14B] FIG. 13B is a plan view of a conversation support system including the panel of FIG. 13A. [Figure 15A] FIG. 10 is a front view of a panel in an embodiment in which speakers output in-phase audio to each side of the panel, and microphones output out-of-phase signals based on the audio from each side of the panel. [Figure 15B] FIG. 15B is a plan view of a conversation support system including the panel of FIG. 15A. [Figure 16A] FIG. 15B is a front view of a panel according to an embodiment different from those of FIGS. 15A and 15B. [Figure 16B] FIG. 16B is a plan view of a conversation support system including the panel of FIG. 16A. [Figure 17A] FIG. 15B is a front view of a panel according to another embodiment different from the embodiment according to FIGS. 15A and 15B. [Figure 17B] FIG. 17B is a plan view of the panel of FIG. 17A. [Figure 18A] FIG. 10 is a front view of a panel according to an example arrangement of multiple microphone modules. [Figure 18B] FIG. 18B is a plan view of the panel of FIG. 18A. [Figure 19A] FIG. 10 is a front view of a panel according to an example of an arrangement of a plurality of speaker modules. [Figure 19B] FIG. 19B is a plan view of the panel of FIG. 19A. [Figure 20A] FIG. 1 is a front view of a panel of a two-way public address system according to the prior art. [Figure 20B] FIG. 1 is a block diagram of the system, including a plan view of the panel. [Figure 20C] FIG. 10 is a block diagram of another example of the system. [Figure 21] FIG. 10 is a block diagram illustrating another embodiment of the present invention. [Figure 22] FIG. 10 is a block diagram illustrating yet another embodiment of the present invention. DETAILED DESCRIPTION OF THE INVENTION

[0011] With the widespread use of splash-proof acrylic partition panels, a simple public address system has been developed as a conversation support measure for elderly people, in which a microphone placed near the speaker picks up the sound, amplifies it, and then emits the sound toward the listener from a speaker placed on the other side of the panel.

[0012] However, since it is difficult for the elderly speaker to hear what they are saying, if a public address system is installed on the speaker's side, the sound will inevitably be transmitted to the microphone, causing feedback and echoes. In an attempt to avoid this, the volume may not be sufficient for two-way conversation, and the sound quality may deteriorate significantly. In addition, the system must be optimally adjusted and operated each time to suit the conditions under which the partition panel is placed.

[0013] To improve droplet prevention and crime prevention performance, it is necessary to reduce the openings and gaps in the panels, but this also exacerbates speech impediments. In particular, for elderly people, in addition to the partition panel, factors such as wearing a mask and hearing loss due to aging combine to cause a decrease in sound pressure concentrated in the high frequency range, leading to serious speech impediments. In many cases, even wearing a hearing aid does not improve the condition.

[0014] The present invention aims to compensate for and repair the reduction in volume and sound quality caused by the partition panel, the two-way conversation disturbance, echo, howling, etc., and to stably increase the volume of the conversation voice or correct the frequency as necessary, thereby restoring and improving sufficient clarity and intelligibility of the conversation. According to the present invention, even if the size of the opening part of the panel is limited or even if it is completely eliminated, it is possible to avoid the conversation disturbance and improve the droplet prevention effect and crime prevention.

[0015] A partition panel separates multiple people, such as people facing each other, and is provided for purposes such as preventing droplets, crime, and infection, and its use is not particularly limited. The material of the partition panel is also not particularly limited, and may be made of resin such as acrylic, glass, wood, paper, etc. The same applies to the size, shape, light transmittance (transparent / semi-transparent / opaque), and color of the partition panel. Hereinafter, the partition panel will be abbreviated as "panel."

[0016] The term "speaker" in the claims should be understood to encompass a "speaker module" and a "speaker unit" that combines multiple "speaker modules." Furthermore, the term "microphone" should be understood to encompass a "microphone module" and a "microphone unit" that combines multiple "microphone modules."

[0017] The conversation support system according to the present invention includes a panel, a speaker and a microphone applied to the panel, and a control circuit for the speaker and the microphone. Fig. 1 is a front view of the panel, and Fig. 2 is a plan view of the panel, showing the arrangement of two people facing each other through the panel. Reference numeral 10 denotes a panel made of a transparent acrylic plate. This panel is placed as a partition between multiple people facing each other and has a substantially rectangular shape (front view). A cylindrical opening 12 is located approximately midway along the height of the panel and near one end along the width. A speaker module SPA is mounted in the opening 12 so that it is flush with the panel and has no gap with the opening wall. An open area 16 cut out in a substantially rectangular or arched shape is provided below the panel for purposes such as exchanging documents between people facing each other through the panel.

[0018] A front view of the speaker module viewed from above and below is shown in Fig. 3, a side view of the speaker module viewed from the side is shown in Fig. 4, and a perspective view of that view is shown in Fig. 5. In Fig. 3, reference numeral 120 denotes a damper, reference numeral 122 denotes a cone paper (diaphragm), and reference numeral 124 denotes an edge. The speaker module SPA is housed in a cylindrical holder 126, which is then placed in a case 128 made of punched metal (an acoustically transparent material), and the case 128 and its components are fitted into the opening 12 of the panel. The case 128 is not essential. The speaker module SPA does not have an enclosure (case), so-called a caseless type, and the panel 10 itself is used as a large baffle plate.

[0019] Meanwhile, the microphone is installed, fixed, positioned, connected, or attached to the boundary of the panel 10. The boundary is, for example, the outline or ridge of the panel 10 itself. The boundary includes a plane as an imaginary outline that virtually expands, extends, or extends the panel 10, and a lower opening plane. By placing the microphone along the boundary of the panel 10, the microphone can be positioned evenly or symmetrically with respect to a person facing through the panel.

[0020] In FIG. 1 , microphone modules Ma and Ma′ are fixed, for example, by affixing, to the center of the upper end of the approximately arch-shaped opening 16 of the panel or the center of the upper end of the panel 10. The type of microphone module need not be limited, but a MEMS type (so-called silicon microphone) is preferable. As will be described later, microphone modules may be provided on both sides of the panel, and their outputs may be added or subtracted for use. Furthermore, by providing microphone module Ma at the upper end of the opening 16, the microphone module can be made inconspicuous to a person facing the panel 10. A microphone fixed in this manner operates as a substantially omnidirectional microphone.

[0021] As shown in Figure 2, microphone modules Ma (Ma') mounted on the boundary of panel 10 pick up the voices of speakers on both sides of panel 10, and through control circuit (amplifier, power amplifier, etc.) 40, the vibration of the diaphragm of speaker module SPA mounted in opening 12 of the panel causes the voices to be emitted to both sides of panel 10 in opposite phase (negative phase) and at the same level, so that speaker 30A and listener 30B hear the sound equally.

[0022] At this time, the sound that passes through the speaker module on both sides of the panel becomes H2(ω) ≒ -H1(ω), i.e., H2(ω) + H1(ω) ≒ 0, and since they are canceled out to approximately zero at the microphone module position, feedback and echo are minimized. As a result, the conversation support system can stably secure the necessary amplification gain and provide sufficient volume for conversation, always and in both directions.

[0023] The panel 10 functions as a large baffle board for small speakers, providing good sound quality across the entire frequency range, including low frequencies. If it is desirable or necessary to block vibrations propagating through the panel and structure-borne sound from the speaker module, the speaker module may be mounted on the panel or supported in a vibration-isolating manner via vibration-isolating material. The same applies to microphones.

[0024] The control circuit 40 comprises a preamplifier / microphone amplifier 42 and a power amplifier 44. Furthermore, digital signal processing circuits / elements (DSP; Digital Signal Processor) such as an equalizer, digital filter, and delay circuit can be inserted between these to correct / emphasize specific frequency bands to further improve voice quality, clarity, and intelligibility, or to apply time delays or speech speed conversion to keep the sound image from straying from the speaker, thereby providing an even more comfortable conversation environment.

[0025] The operation of the conversation support system will be explained in more detail with reference to Figure 2. Speaker 30A speaks with an average spectrum S0(ω), and the voice of source speaker 30A is picked up by a microphone module (a silicon microphone: MEMS Mic, approximately 3 mm x 5 mm x 1 mm thick). The voice is then amplified by a microphone amplifier 42 and a power amplifier 44, which is preceded by a minimum number of frequency filters, and the voice is emitted from speaker module SPA to speaker 30B.

[0026] The response R1(ω) at the ear position of speaker 30B of this loudspeaker system is calculated using the following equation: M(ω), S1(ω), H1(ω), and H2(ω) are the microphone response, the transfer function (including the speaker module response) from the front side of speaker module SPA (speaker 30A side) to speaker 30B and the microphone module position, and the transfer function from the back side of the speaker module (speaker 30B side) to the microphone module position, respectively.

[0027] The response R1(ω) to speaker 30B is the sum of a geometric series with the first term a≡q S0(ω)M(ω)S1(ω) and the common ratio r≡qM(ω)[H1(ω)+H2(ω)], where R1(ω)=a(1-r n ) / (1-r), and n=∞→a / (1-r), which gives us the following (see Note 1 below).

[0028] R1(ω)=q S0(ω)M(ω)S1(ω)+q 2 S0(ω)M 2 (ω)S1(ω)[H1(ω)+H2(ω)] +q 3 S0(ω)M(ω)[H1(ω)+H2(ω)]{qM(ω)[H1(ω)+H2(ω)]} +q 4 S0(ω)M(ω)[H1(ω)+H2(ω)]{[qM(ω)[H1(ω)+H2(ω)]}{[qM(ω)[H1(ω)+H2(ω)]} +······ =Σq S0(ω)M(ω)S1(ω)·{q M(ω)[H1(ω)+H2(ω)]}N-} =q S0(ω)M(ω)S1(ω)<1-{q M(ω)[H1(ω)+H2(ω)] N}> / {1-[q M(ω)[H1(ω)+H2(ω)]}····I

[0029] If the absolute value of the common ratio qM(ω) [H1(ω) + H2(ω)] is 1 or less, the Nth power term in equation I disappears, R1(ω)=q S0(ω)M(ω)S1(ω) / {1-[qM(ω)[H1(ω)+H2(ω)]}·····II Here, G(ω)≡R1(ω) / S0(ω)=q M(ω)S1(ω) / {1-[qM(ω)[H1(ω)+H2(ω)]} is called the loop gain (LG), and qM(ω)[H1(ω)+H2(ω)] is called the open loop gain (OLG).

[0030] Furthermore, since the speaker sound is radiated to both sides of the plate at the same level and in opposite phase, H1(ω) ≒ -H2(ω), and therefore, R1(ω)≒q S0(ω)M(ω)S1(ω)·········III This becomes:

[0031] To speaker 30A, S2(ω)≒-S1(ω), and R2(ω)≒q S0(ω)M(ω)S2(ω)≒-R1(ω) IV That is, the speakers 30A and 30B hear the same amplified sound at the same level, but in opposite phases.

[0032] (Note 1) 1st time: q S0(ω)M(ω)S1(ω) directly reaches speaker 30B Second time: q S0(ω)M(ω) returns to Mic via [H1(ω)+H2(ω)] and reaches speaker 30B via M(ω)qS1(ω) → qS0(ω)M(ω)·[H1(ω)+H2(ω)]M(ω)qS1(ω) =q 2 S0(ω)M 2 (ω)[H1(ω)+H2(ω)] S1(ω) 3rd time:q 2 S0(ω)M2 (ω)[H1(ω)+H2(ω)] returns to the microphone via [H1(ω)+H2(ω)] and reaches speaker 30B via M(ω)qS1(ω) as above. →q 2 S0(ω)M 2 (ω)[H1(ω)+H2(ω)]·[H1(ω)+H2(ω)]M(ω)qS1(ω) =q 3 S0(ω)M 3 (ω)[H1(ω)+H2(ω)] 2 S1(ω)

[0033] In other words, if we can control the system so that H1(ω) ≒ -H2(ω) in equation II, then the denominator of R1(ω) in equation III will be 1, By increasing q, R1(ω) ≒ q S0(ω)M(ω)S1(ω), it is possible to ensure the necessary stable amplification volume without feedback or echo (i.e., without the Nth power term in Equation I diverging).

[0034] In reality, the sound levels emitted from the front and back of the speaker module are almost equal in volume but in opposite phase, so The condition H1(ω) ≒ -H2(ω) is met, and the echo sound from the speaker module is canceled out and reduced to almost zero on the ridge of the panel. Therefore, it can be said that installing a single microphone module on the ridge of panel 10 is sufficient. In reality, because reflected sound and reverberation follow the direct sound in a room, it is difficult to completely satisfy the condition H1(ω) = -H2(ω). However, this condition is almost met for direct sound and early reflected sound, which are normally at their maximum levels, so it is possible to ensure a practically sufficient sound reinforcement gain q.

[0035] Since speaker 30A also receives equal-level, opposite-phase voice S2(ω)≒-S1(ω), speaker 30A can constantly monitor his / her own voice, i.e., can confirm the volume at which his / her own voice is reaching speaker 30B. In this way, a feature of the conversation support system is that voices are always provided to both sides of the board at equal levels, albeit in opposite phase.

[0036] Of course, there is no problem in the system even if the speakers 30A and 30B speak at the same time. R2(ω) ≒ q - R1(ω), that is, |R2(ω)| ≒ |R1(ω)|. To further fulfill the condition H1(ω) ≒ -H2(ω), as will be described later, instead of a single speaker module, two speaker modules may be arranged facing each other and / or facing each other back to back, and the two speaker modules may be driven with opposite-phase signals.

[0037] The installation positions and number of microphone modules may be changed as appropriate with respect to the panel 10. For example, they may be located on the edge (boundary) of either the left or right side of the opening 16, the upper edge of the panel 10, or the edge of either the left or right side of the panel. Furthermore, as already mentioned, microphone modules may be provided at positions on an imaginary extension plane of the boundary of the panel 10. In this case, the microphone modules may be connected to a support member that supports them on the panel, and the microphone modules may be fixed to the panel via the support member.

[0038] Furthermore, the number of microphone modules placed on the panel boundary is not limited to one, and multiple microphone modules may be placed symmetrically with respect to the panel 10. Multiple microphone modules may be arranged adjacent to each other in a line on the panel. This increases the directivity in the front and rear directions of the panel 10, or in both front and rear directions, thereby increasing the direct sound ratio of the conversational voice and improving the clarity of the conversation. In other words, this increases the sound collection efficiency for speakers 30A and 30B on both sides of the panel, resulting in the same effect as increasing the direct sounds to the speakers, i.e., the direct sounds S1(ω) and S2(ω) described above in II and III, and allows for a stable increase in the amplification gain of the entire system without echo or feedback.

[0039] When there are multiple microphone modules, the control circuit 40 adds the audio signals from the multiple microphone modules as in-phase (in-phase addition), amplifies them with a power amplifier, and then outputs them to the speaker. When the speaker is configured with a speaker module mounted on each of multiple openings, the control circuit 40 synchronizes the amplified audio signals and outputs them to each speaker module.

[0040] The above-described embodiment has been described as an example in which one speaker module is mounted in the opening 12 of the panel 10. In order to perfectly match the audio radiated in both directions, i.e., to more perfectly match H1(ω) ≈ -H2(ω), two speaker modules can be mounted as a pair in the opening 12 of the panel 10. The speaker unit configuration in which two speaker modules are paired includes a type in which the fronts of the two speaker modules face each other and a type in which the backs of the two speaker modules face each other. The speaker unit is mounted within the opening 12 of the panel 10. The control circuit 40 drives the two speaker modules to output audio in opposite phases and at the same level.

[0041] Figure 6A is a block diagram of a speaker unit in which two speaker modules SPA, SPB are arranged facing each other and the negative terminal of the first speaker module SPA is connected to the negative terminal of the second speaker module SPB, thereby connecting the two speaker modules SPA, SPB in parallel to an amplifier 44. Figure 6B is a block diagram of a speaker unit in which two speaker modules SPA, SPB are arranged facing each other and the positive terminal and negative terminal of the first speaker module SPA and the second speaker module SPB are connected to each other, thereby connecting the two speaker modules in series to an amplifier 44.

[0042] Figure 7A is a block diagram of a speaker unit in which two speaker modules SPA, SPB are arranged back to back facing each other and the negative terminal of the first speaker module SPA is connected to the negative terminal of the second speaker module SPB, thereby connecting the two speaker modules SPA, SPB in parallel to an amplifier 44. Figure 7B is a block diagram of a speaker unit in which two speaker modules SPA, SPB are arranged back to back facing each other and the positive terminal and negative terminal of the first speaker module SPA and the second speaker module SPB are connected to each other, thereby connecting the two speaker modules SPA, SPB in series to an amplifier 44.

[0043] Figure 8 is a block diagram of a conversation support system according to another embodiment different from that of Figure 2. A control circuit 40 synchronously adds (45) the audio signals from three microphone modules Ma-1, Ma-2, and Ma-3 that are installed in a line along the ridge of the panel, and then outputs the audio signals amplified by a microphone amplifier 42 and a power amplifier 44 to each speaker module of a speaker unit SPU1, which is made up of a pair of speaker modules connected facing each other, and drives the speaker modules to output audio at the same level but in opposite phases.

[0044] For speaker unit SPU2, in which two speaker modules are connected back-to-back, control circuit 40' uses two power amplifiers 44 to drive the pair of speaker modules in opposite phases, as shown on the left side of Figure 8, and connects an inverting amplifier 43 in front of one of them to form audio signals of opposite phases before the speaker units, and outputs one of the signals of opposite phases and equal levels to one speaker module, and the other signal to the other speaker module. Either method may be used to output opposite-phase audio from the speaker modules.

[0045] 9 is a block diagram of a conversation support system according to yet another embodiment. For speaker unit SPU1, which has two speaker modules connected back-to-back, control circuit 40 generates opposite-phase signals by placing audio transformers 47 in front of multiple amplifiers 44. Furthermore, for speaker unit SPU2, which has two speaker modules connected back-to-back, control circuit 40' electronically generates opposite-phase signals using multiple operational amplifiers (OP amplifiers) 58 and the like.

[0046] 10A, 10B, and 10C are block diagrams of a conversation support system according to yet another embodiment. FIG. 10A is a front view of panel 10, and FIGS. 10B and 10C are schematic diagrams showing the cross-sectional structure of panel 10 at different height positions in the front view. SPA and SPB are single speaker modules mounted in openings in panel 10, and SPU1 to SPU4 are the aforementioned speaker units mounted in circular openings in the panel. To accurately align the sound image on the center line of panel 10, in other words, to accurately localize the sound image directly in front of the panel, the speaker modules SPA and SPB in FIG. 10A may be positioned on the center line of panel 10 at a certain distance from microphone Ma to avoid feedback and echoes.

[0047] 10B and 10C, multiple speaker units SPU1 to SPU4 may be mounted on the panel 10 so as to be symmetrical on the left and right sides of the panel. The number of speaker modules and / or speaker units mounted on the panel may be one or more, and their positions are not particularly limited as long as they are symmetrical and even with respect to the speakers 30A and 30B.

[0048] 11A and 11B, the speaker module SPA may be placed in a small case (enclosure) 100 to form a speaker module with a housing, and multiple speaker modules with a housing may be applied to both sides of the panel 10 so as to face each other via the panel 10. The speaker modules with a housing are not mounted in openings in the panel 10, but are fixed to the surface of the panel by, for example, adhering. The multiple speaker modules with a housing may be arranged so as to be aligned and plane-symmetrical while facing each other on the front and back sides of the panel. When a pair of speaker modules with a housing is applied to a panel, they may be symmetrical above, below, left, and right of the panel.

[0049] This embodiment eliminates the need for openings in the panel to mount the speaker module, and therefore the speaker can be made into a separate unit along with the microphone and control circuit, making it possible to commercialize the product as a system that can be retrofitted to an existing panel. The speaker module with enclosure and microphone module Ma do not need to be directly connected to the panel, as long as they are evenly and symmetrically distributed across the panel for multiple people facing each other.

[0050] Regarding the arrangement and number of microphone modules on the panel, as shown in FIG. 12A , microphones may be installed at one or more predetermined positions along the boundary of the panel 10, as long as the microphone modules are positioned so that they can evenly capture the reflected sound from the speakers on both the front and rear sides of the panel 10. Ma, Mb, Mc, Md, Me, and Mf indicate microphones arranged on the actual boundary of the panel (ridge line: indicated by dashed dotted line 200), and Ma' and Mf' indicate microphone modules arranged on the imaginary extended boundary (extended surface) of the panel. Mf' is placed on the table surface directly below the opening 16. The arrangement and number of microphone modules on the panel may include both of these arrangements.

[0051] Next, an embodiment relating to the arrangement and number of microphone modules on the panel, which differs from that shown in Fig. 12A, will be described with reference to Fig. 12B. As surrounded by the dashed line, multiple microphone modules are arranged at equal intervals as small as possible along the ridge line of the panel to form a microphone module unit, and control circuit 40 adds the signals from each module while keeping them in phase, resulting in a so-called mixed output.

[0052] MaU is a microphone unit in which microphone modules Ma-1, Ma-2, and Ma-3 are arranged in a line along the top edge of panel 10, and MaU' is a microphone unit in which microphone modules Ma'-1, Ma'-2, and Ma'-3 are arranged in a line along an imaginary extension line of panel 10, similar to Ma' in Figure 12A.

[0053] Furthermore, MaU1 is a microphone module unit in which microphone modules Maf-1, Maf-2, and Maf-3 are arranged in a line along the upper edge of the opening 16 of the panel 10, and MaU1' is a microphone unit in which microphone modules Maf'-1, Maf'-2, and Maf'-3 are arranged in a line along the ground surface facing the upper edge of the opening 16 of the panel 10.

[0054] This microphone unit (line microphone) is the most acoustically efficient and practical configuration, and can increase the directivity of sound in the front direction on both sides of the panel. This increases the sound pickup efficiency for speakers 30A and 30B on both sides of the panel, resulting in the same effect as increasing the direct sounds to the speakers, i.e., the direct sounds S1(ω) and S2(ω) in equations II and III described above, and can stably increase the sound amplification gain of the entire system without echo or feedback.

[0055] Next, regarding the arrangement and number of microphone modules on the panel, embodiments different from those shown in FIGS. 12A and 12B will be described with reference to FIGS. 13A and 13B. FIG. 13A is a front view of the panel, and FIG. 13B is a plan view of a conversation support system including the panel. In the previously described embodiments, microphone modules were placed on the actual boundary of the panel or on a virtual extension (extension plane) of the panel. However, in this embodiment, one or more microphone modules are arranged symmetrically with respect to each other across the panel, back-to-back facing each other (facing forward from the panel), and connected to both sides of the panel. The microphone amplifier 44 of the control circuit 40 adds the outputs of the microphone modules Ma and Ma' on both sides of the panel while maintaining their in-phase signals, and the power amplifier 44 drives the multiple speaker modules of the speaker unit SPU in opposite phases, i.e., outputs audio signals of opposite phases to each of the multiple speaker modules.

[0056] As shown in Figures 13A and 13B, multiple speaker units may be applied to a panel symmetrically on the left and right sides of the panel. The microphone arrangement in Figures 13A and 13B is called a surface pressure microphone, which exhibits approximately hemispherical directivity in the front direction of each panel surface and outputs a signal corresponding to the surface sound pressure of the panel surface, which is twice the incoming sound pressure across the entire frequency band.

[0057] 14A and 14B show an embodiment in which microphone units MaU and MaU', each of which has multiple microphone modules arranged in a line, are arranged on both sides of the panel so that the microphone units are plane-symmetrical to each other. The control circuit 40 performs in-phase addition on the signals from the multiple microphone modules arranged in a line for each pair of microphone units and then amplifies them (45, 42). The control circuit 40 then further in-phase adds the in-phase added outputs of the two microphone units (OP amplifier 49) and outputs the amplified (power amplifier 44) outputs to housing-mounted speaker modules SPA and SPB, which are arranged on both sides of the panel and driven in opposite phases to each other.

[0058] In the previously described embodiments, the control circuit outputs opposite-phase audio from the speaker to each side of the panel, and the microphone outputs an in-phase signal based on the audio from each side of the panel. In contrast, the following embodiments have the opposite relationship, i.e., the speaker outputs in-phase audio from each side of the panel, and the microphone outputs an opposite-phase signal based on the audio from each side of the panel, and can achieve the same effects as the previously described embodiments.

[0059] First, of the multiple embodiments, the simplest embodiment is shown in Figure 15A (front view) and Figure 15B (plan view). Two housing-attached speaker modules SPAB1 and SPAB2 face each other across the panel 10, and are attached to the panel 10 so that the two modules are plane-symmetrical to each other. As mentioned above, when applying speaker modules to a panel, it is preferable in terms of sound quality for the speaker modules to be fixed in contact with the panel. Therefore, the housing of the housing-attached speaker module can be fixed directly to the panel, or the housing-attached speaker module can be attached to the panel via vibration-damping material such as a rubber plate or sponge material to prevent structure-borne sound (vibration) from being transmitted from the speaker to the microphone via the panel.

[0060] To achieve an anti-phase, equal level on the microphone side, a pair of microphone modules Ma-1 and Ma-2 are fixed to each side of the panel so that they are plane-symmetrical to each other, and the control circuit 40 performs anti-phase addition (subtraction) of the outputs received by each microphone module. Specifically, as shown in FIG. 15B, a subtraction circuit such as an OP amplifier (operational amplifier IC) 49 performs anti-phase addition of the outputs from each microphone module and outputs the result. The microphone modules may be attached directly to the panel 10, or may be supported on the panel via vibration-isolating material. Since each microphone is attached in approximate contact with the panel, it operates as a sound pressure microphone (outputting twice the incoming sound pressure).

[0061] This embodiment operates as follows: Two microphone modules (silicon Mic / MEMS Mic, approximately 3 mm x 5 mm x 1 mm thick) Ma-1 and Ma-2 are placed on the front and back of the panel 10 in positions that are symmetrical to each other in relation to a speaker 30A speaking with an average spectrum S0(ω), and each of these receives the voice S0(ω) of the speaker on its respective side and the reflected sound H(ω) from two speaker modules SPAB1 and SPAB2 placed in positions that are symmetrical to each other on both sides of the panel.

[0062] The two speaker modules are driven in phase by a power amplifier 44 preceded by a microphone amplifier and a minimal frequency correction circuit (filter), and the microphone modules Ma-1 and Ma-2 placed on each side of the panel output their subtracted (inverse phase added) signals, i.e., the difference signal between the outputs of the two microphone modules, via an OP amplifier 49. In other words, the voices emitted by speakers 30A and 30B are radiated to both sides of the panel in real time as needed, and simultaneously reach the two microphone modules at equal levels, where they are picked up together with the speakers' voices.

[0063] The response R1(ω) at the ear position of speaker 30B of this loudspeaker system is calculated as follows: where M1(ω), M2(ω), S(ω), and H(ω) are the responses of microphone modules Ma-1 and Ma-2, respectively, and the transfer functions from both speaker modules to the speakers and microphone modules on each side of the panel (including the responses of the speaker modules). The response R1(ω) to speaker 30B is the sum of a geometric series with the first term a≡qS0(ω)M1(ω)S(ω) and the common ratio r≡qH(ω)[M1(ω)-M2(ω)], and is given by R1(ω)=a(1-r n ) / (1-r); n=∞ → a / (1-r), which can be shown as follows.

[0064] R1(ω)=q S0(ω)M1(ω)S(ω) +q S0(ω)M1(ω)·q H(ω)[M1(ω)-M2(ω)]·S(ω) +q S0(ω)M1(ω)·q H(ω)[M1(ω)-M2(ω)]·q H(ω)[M1(ω)-M2(ω)]·S(ω) +------ =q S0(ω)M1(ω)S(ω) +q S0(ω)M1(ω)S(ω) · qH(ω)[M1(ω)-M2(ω)] +q S0(ω)M1(ω)S(ω)·q 2 {H(ω)[M1(ω)-M2(ω)]} 2 +------ =q S0(ω)M1(ω)S(ω)·Σ{q H(ω)[M1(ω)-M2(ω)]} N-1 =q S0(ω)M1(ω)S(ω)<1-{q H(ω)[M1(ω)-M2(ω)]} N > / {1-q H(ω)[M1(ω)-M2(ω)]}·····IV-1

[0065] This formula is a rewrite of the M (microphone) and S (speaker) in the formula I mentioned above, and whether "reverse phase, equal level" is applied to (1) the speaker side or (2) the microphone side, the effect of being able to avoid echo and howling is the same.

[0066] As with the case where the speaker side is in reverse phase, if the absolute value of the common ratio is 1 or less, the N-th power term disappears, R1(ω)=q S0(ω)M1(ω)S(ω) / {1-q H(ω)[M1(ω)-M2(ω)]}... This becomes:

[0067] As in (1), here G(ω)≡R1(ω) / S0(ω)=q M1(ω)S(ω) / {1-q H(ω)[M1(ω)-M2(ω)]} is called the loop gain (LG), and q H(ω)[M1(ω)-M2(ω)] is called the open loop gain (OLG).

[0068] Furthermore, it is intended and configured to utilize the subtraction output of M1(ω) and M2(ω), and if both have the same sign and are at the same level, M1(ω) ≒ M2(ω), as shown in the following equation.

[0069] R1(ω) = q S0(ω)M1(ω)S(ω) ≒ q S0(ω)M2(ω)S(ω) = R2(ω)····VI

[0070] The above Formulas V and VI also have the same form as Formulas II and III if you rewrite M and S. If you can control the system so that M1(ω) ≒ -M2(ω) in Formula V, then R1(ω) will become R1(ω) ≒ q S0(ω)M1(ω)S(ω), as in Formula VI, and by making q larger, there will be no feedback or echo, in other words, the Nth power term in Formula IV-1 will not diverge, and the necessary stable loudspeaker volume can be ensured.

[0071] In reality, the sensitivities of the two microphone modules are roughly equal, and an operational amplifier subtracts signals so that the output is exactly zero, so the condition M1(ω) - M2(ω) ≒ 0 is met, and the response to the speaker follows the VI equation in both directions. In other words, all that is required is for each of the two speaker modules and microphone modules to be positioned so that they are symmetrical relative to the panel. In reality, since there is reflected sound and reverberation following the direct sound in a room, it is difficult to perfectly satisfy the condition M1(ω) = M2(ω), but this condition is generally met for the direct sound and early reflected sound, which are normally at their maximum levels, so it is possible to ensure a sufficient sound reinforcement gain q for practical use.

[0072] Incidentally, as in the case of (1): out-of-phase speaker module side, both speaker modules emit the same sound (in-phase, equal level). Therefore, in a typical environment where the transfer function S(ω) from the speaker to the speaker is approximately equal on both sides of the panel, speaker 30A's own voice also reaches speaker 30A at an equal level, making it possible to constantly monitor the volume at which his or her own voice is reaching listener 30B. Also, as in equation VI, R2(ω) has the same shape as R1(ω), and the same conditions apply when speaker 30B speaks to speaker 30A. Of course, there is no problem if speakers 30A and 30B speak simultaneously.

[0073] Also, in the case of (2) out-of-phase microphones, multiple speaker modules (a pair of speaker modules with housings) can be installed at symmetrical positions on the left and right of the panel to center the sound image in front of the panel, or multiple microphone modules can be arranged in a line at symmetrical positions on both sides of the panel and the in-phase signals added together can be subtracted using an operational amplifier to increase the sound pickup efficiency in front of the panel, i.e., to increase the directional gain in the front direction. Of course, the same effect can be achieved by using an acoustic transformer for the subtraction process. In either case, speaker modules with housings are installed at symmetrical positions on both sides of the panel, back-to-back facing, and driven in-phase.

[0074] 16A, 16B, 17A, and 17B show embodiments different from those shown in FIGS. 15A and 15B. In FIGS. 16A and 16B, multiple housing-mounted speaker modules SPAB1 to SPAB4 are mounted on the front and back of a panel 10 (on both sides of a generally arch-shaped opening 16) in plane-symmetrical positions, facing back to back (with their sound-generating surfaces facing outward, i.e., facing away from the panel surface). Microphone units MaU and MaU', each consisting of three linearly connected microphone modules, are also arranged on both sides of the panel 10, plane-symmetrically facing back to back (with their sound holes facing outward, i.e., facing in opposite directions on the panel surface). A control circuit 40 performs reverse-phase addition (subtraction) of in-phase signals from the microphone units using a downstream operational amplifier or the like, and outputs the difference signal to a power amplifier 44. An operational amplifier 42 performs in-phase addition of the outputs of the microphone modules of the microphone units.

[0075] 17A and 17B, a pair of single microphone modules Ma and Ma' are arranged as microphones, facing each other in plane symmetry across the panel 10. The speaker units SPBA1 to SPAB4 also radiate in-phase audio signals on the front and back of the panel 10, similar to the embodiment of FIGS.

[0076] 18A and 18B show an example of an arrangement of multiple microphone modules. Microphone modules Ma, Mb, Mc, and Md are widely scattered symmetrically with one another on the front side of panel 10. Microphone modules Ma', Mb', Mc', and Md' are scattered on the back side of panel 10 so as to face the microphone modules on the front side.

[0077] 19A and 19B also show one example of speaker module arrangement. Housing-attached speaker modules SPAB1, SPAB2, SPAB3, SPAB4, and SPAB5 are widely scattered symmetrically on the front side of panel 10. Housing-attached speaker modules SPAB1', SPAB2', SPAB3', SPAB4', and SPAB5' are scattered on the back side of panel 10 so as to face the speaker modules on the front side. A pair of microphones Ma and Ma' are bonded to the front and rear surfaces of panel 10, respectively, so as to face each other.

[0078] Next, Figures 20A, 20B, and 20C illustrate a two-way sound amplification system based on conventional technology. Figure 20A shows the front of the panel, Figure 20B is a block diagram of the system including a plan view of the panel, and Figure 20C is a block diagram of another example of the system. A control circuit 40 realizes a system on both sides of the panel in which sounds picked up by microphone modules Ma and Ma' are output from speakers SPAB and SPAB' on the opposite side of the microphone modules. In other words, an "eight-shaped" feedback loop is created: Ma' ⇒ SPAB ⇒ Ma ⇒ SPAB' ⇒ Ma'.... Since there is no clear attenuation element within the loop, feedback ultimately occurs before sufficient sound amplification volume is achieved.

[0079] As such, simply incorporating two sound reinforcement systems into a system does not solve the problems described above. Figure 20C shows an even more general example, in which unidirectional microphones Ma and Ma' and speakers SPAB and SPAB' are placed in close proximity to the speaker in order to improve the direct sound pickup efficiency and increase the sound pressure at the listener's ears. At first glance, this may seem to increase the sound reinforcement gain, but the closer the microphones and speakers are, the larger the transfer function H(ω) between them becomes. Ultimately, the "figure-eight" feedback loop prevents stable increases in sound reinforcement gain.

[0080] As shown in FIG. 21, by placing a frequency correction circuit (equalizer / EQ) 70 before the power amplifier 44, it is possible to emphasize specific audio bands or cut low frequencies in the audio band that are not related to intelligibility or background noise in the room.

[0081] 22, the echo component (echo residual component) that has leaked from the speaker module SPA to the microphone module Ma is found in advance by measuring the impulse response, and by adding a circuit (echo canceller) 72 that feeds back (74, 76) the same component to the control circuit 40 to cancel it, the loop gain (LG) of the system can be further reduced (toward zero). By using these circuits and technologies in combination, the precision and functionality of industrial products based on the present invention can be improved.

[0082] The above-described embodiment has been described as a system for facilitating conversation between multiple people separated by a panel, but the present invention is not limited to this and the microphone, speaker, and control circuit of the present invention can be applied to any partition that can be called a partition, such as a wall, glass window, door, or screen, which separates multiple people.

[0083] While the above-described embodiments have been described as involving one person facing another via a panel, the present invention can also be applied to situations in which multiple people face each other via a panel. That is, this embodiment of the present invention is an expandable conversation support system that uses one or a pair of speaker modules and one or a pair of microphone modules as a set, and expands it horizontally to accommodate large-scale conferences with multiple people. The opening in the panel can be appropriately configured depending on the position of the microphone module. Furthermore, this embodiment of the present invention is a conversation support system that improves anti-feedback / echo performance, sound quality, sound image localization, etc. by inserting an echo cancellation circuit, frequency correction circuit, delay circuit, etc. between the microphone amplifier and power amplifier using a DSP (Digital Signal Processor). Furthermore, the disclosure of this application encompasses a conversation support method implemented by the control circuit. [Explanation of symbols]

[0084] 10 Screen Panel 12 Opening 16 Open area Ma Microphone Module SPA Speaker Module

Claims

1. A partition panel that is set up between speakers who are talking face-to-face in the same room to separate droplets between the speakers, a microphone and a speaker; a control circuit that processes the sound captured by the microphone and outputs the sound from the speaker; and an opening. Equipped with The speaker is installed in the opening; the microphone is provided on the partition panel so as to be symmetrical with respect to the speaker facing the partition panel, the partition panel is used as a baffle plate of the speaker, and the speaker outputs sound of a first phase to a first side of the partition panel, and outputs sound of a second phase, which is inverse to the first phase, to a second side of the partition panel opposite to the first side; The audio of the first phase and the audio of the second phase are added together and cancelled out, and the microphone captures audio signals from at least one side or both sides of the speakers who are close to each other and facing each other, and outputs the audio signals from the speaker; Further, an open area is provided for communication between the facing speakers. Folding panel.

2. 2. The screen panel of claim 1, wherein the microphones are located along a boundary of the screen panel.

3. The partition panel according to claim 2 , wherein the boundary is a boundary of the partition panel itself and / or a boundary virtually extended from the partition panel.

4. 4. The partition panel according to claim 1, wherein the speaker is composed of a plurality of speaker modules including a first speaker module and a second speaker module, and the first speaker module and the second speaker module are connected to each other facing front to front or facing back to back.

5. The microphone is a plurality of microphone modules connected in a line, the control circuit adds the signals from the plurality of microphone modules in phase and outputs the sum from the speaker; The partition panel according to any one of claims 1 to 4.

6. the microphone comprises a plurality of microphone modules including a first microphone module and a second microphone module, the first microphone module being fixed to the first side and the second microphone module being fixed to the second side; 5. The screen panel according to claim 1, wherein the control circuit adds the signals from the first microphone module and the second microphone module in phase and outputs the result from the speaker.

7. the speaker is composed of a plurality of speaker modules including a first speaker module and a second speaker module, the first speaker module is provided on a first side of the partition panel, the second speaker module is provided on a second side of the partition panel, the first speaker module and the second speaker module face each other symmetrically across the partition panel, the control circuit outputs the sound captured by the microphone to the first side via the first speaker module and to the second side via the second speaker module, and further outputs the sound from the first speaker module and the sound from the second speaker module in phase or in opposite phase to each other; The partition panel according to claim 1.

8. the first speaker module and the second speaker module are speaker modules with housings; The partition panel according to claim 7.

9. the microphone is provided along the boundary of the partition panel; the control circuit outputs the audio from the first speaker module and the audio from the second speaker module in opposite phases to each other; The partition panel according to claim 7 or 8.

10. the microphone is composed of a plurality of microphone modules including a first microphone module and a second microphone module, the first microphone module is provided on a first side of the partition panel, the second microphone module is provided on a second side of the partition panel, and the first microphone module and the second microphone module face each other symmetrically across the partition panel; the control circuit captures the sound from the first microphone module and the sound from the second microphone module in phase or in opposite phase to each other; The partition panel according to claim 7 or 8.

11. The control circuit outputting audio from the first speaker module and audio from the second speaker module in phase with each other; capturing the sound from the first microphone module and the sound from the second microphone module in opposite phases; The partition panel according to claim 10.

12. The control circuit outputting audio from the first speaker module and audio from the second speaker module in opposite phases; capturing the sound from the first microphone module and the sound from the second microphone module without making them out of phase with each other; The partition panel according to claim 10.

Images