JP2010206365A - Interaction device - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To detect a distance to an operator without requiring an additional dedicated sensor or microphone, and to perform interaction with appropriate gain. <P>SOLUTION: A reception terminal 20 includes a microphone 207 for inputting voice, and a speaker 208 for outputting voice, acquires corresponding noise information by noise input through the microphone 207, outputs pseudo noise generated based on the acquired noise information through the speaker 208, acquires corresponding reflected voice information by voice reflected on an object of the pseudo noise input through the microphone 207, estimates that the object is a visitor M by carrying out a predetermined calculation process based on the acquired reflected voice information to detect the distance to the visitor M, and adjusts the gain of the microphone 207 based on the detection result. <P>COPYRIGHT: (C)2010,JPO&INPIT

Description

  The present invention relates to an interactive device that can be operated by an operator using a voice interactive method.

  For example, an interactive device that can be operated by an operator in an interactive manner, such as a reception device that performs a reception work for a visitor to a building, has been known. In such an interactive apparatus, the content of the operator's utterance is input by a voice input means such as a microphone, and is amplified by an appropriate gain and then recognized. At this time, if the gain is too small with respect to the utterance voice level, erroneous recognition occurs, and if the gain is too large with respect to the utterance voice level, recognition becomes impossible due to sound cracking. Generally, in the dialogue processing, as a general rule (unless specifically requested), the operator utters the same content only once. Therefore, in the above case, the utterance content is not recognized. In order to prevent such recognition omission, it is necessary to optimize the gain. Generally, when the distance from the device to the operator is long, the speech level input to the sound input means is small (in other words, the gain needs to be increased), and when the distance from the device to the operator is short, the sound input means The input speech level is high (in other words, the gain needs to be reduced). Therefore, in order to optimize the gain, it is preferable to accurately detect the distance from the apparatus to the operator in a non-contact manner and perform the dialogue process with an appropriate gain corresponding to the distance.

  For such non-contact distance detection, for example, the prior art described in Patent Document 1 is known. In this prior art, an ultrasonic pulse is generated and output to an object (object), and a reflected wave (echo pulse) on the detection object is detected. Then, by calculating the transmission time of the ultrasonic pulse, the distance to the object is detected based on the transmission time.

JP 2005-351897 A

  However, when the distance detection method using ultrasonic waves as in the above-described conventional technique is applied to the interactive device, there is a problem that it becomes necessary to newly provide a sensor or microphone dedicated to distance detection.

  An object of the present invention is to provide an interactive apparatus capable of detecting a distance to an operator and performing interactive processing with an appropriate gain without newly providing a dedicated sensor or microphone.

  In order to achieve the above object, the first invention is an interactive apparatus that can be operated by an operator in an interactive manner, and includes an audio input means for inputting audio, and an audio output means for outputting audio. Source sound acquisition means for acquiring source sound information including a corresponding amplitude or frequency from a source sound that is input via the voice input means and serves as a generation source of detection sound for distance detection, and the voice input means includes the A detection sound output means for outputting the detection sound generated based on the source sound information acquired by the source sound acquisition means within the predetermined time after the sound is input via the sound output means; and the sound input Based on the reflected sound information acquired by the reflected sound acquisition means and the reflected sound acquisition means for acquiring the corresponding reflected sound information by the reflected sound of the detected sound from the object input through the means. A distance detection means for performing a predetermined calculation process, detecting that the object is the operator and detecting a distance to the operator, and the voice input means based on the detection result of the distance detection means And a sensitivity adjusting means for adjusting the gain.

  In the interactive apparatus according to the first aspect of the present application, the distance to the operator is detected using the detection sound for distance detection. In other words, the source sound that is the generation source when the detected sound is generated is input via the sound input unit, and the corresponding source sound information is acquired by the source sound acquisition unit. Then, based on this source sound information, the detection sound output means outputs the detection sound via the sound output means. When the output detection sound propagates toward the object and is reflected by the object, the reflected sound is input via the sound input means, and the corresponding reflected sound information is acquired by the reflected sound acquisition means. Since the time from when the detection sound is emitted until the reflected sound returns is proportional to the distance from the device to the target object, the distance detection means determines that the target object is an operator based on the reflected sound information. The distance to the operator is detected assuming that there is. Based on the detected distance, the sensitivity adjustment unit adjusts the gain of the voice input unit. As a result, when the distance to the operator is relatively close, the spoken voice of the operator during the dialogue is input at a relatively high level, so the gain of the voice input means is lowered and the distance to the operator is compared. When the user is far away, the speech voice of the operator at the time of dialogue is input at a relatively small level. Therefore, it is possible to perform dialogue processing at an appropriate signal level by increasing the gain of the voice input means. As a result, it is possible to perform reliable interactive processing without recognition failure.

  As described above, in the first invention of the present application, the distance to the operator can be detected using the sound input / output via the voice input means and the voice output means. That is, by utilizing voice input means (such as a microphone) and voice output means (such as a speaker) that are originally provided for interactive processing, other distance detection sensors and dedicated microphones are newly provided. Distance detection can be performed without fail, and reliable dialogue processing can be performed with an appropriate gain based on the distance detection.

  According to a second aspect of the present invention, in the first aspect, the source sound acquisition unit is configured to generate the source sound based on the utterance voice as the source sound that is uttered by the operator in the dialogue processing of the dialogue apparatus and input by the voice input unit. Utterance voice information as information is acquired, and the detection sound output means outputs the detection sound generated based on the utterance voice information acquired by the source sound acquisition means via the voice output means. And

  When the detection sound is generated based on the source sound, if the level of the original source sound is too small, the level of the output detection sound is also small, and it is difficult to detect the reflected sound. Usually, an operator who intends to perform an operation in an interactive manner intends to recognize his / her speech as reliably as possible, and speaks slowly and at a high volume. Therefore, when detecting the distance, the detection sound (pseudo utterance sound) is generated and used based on the utterance voice of such an operator, so that accurate and reliable distance detection can be performed. In addition, by using the utterance voice that the operator himself utters, the effect that the operator is relatively less aware of the detection using sound, and the gain of the operator by predicting the position of the operator in advance. Can be adjusted (compared to the case where adjustment is performed by auto gain control).

  According to a third aspect of the present invention, in the first aspect, the source sound acquisition means is a device sound as the source sound that is output from the voice output means and input by the voice input means in an interactive process with an operator, or the device The apparatus sound information as the source sound information is acquired from the reflected sound of the sound, and the detection sound output means uses the detection sound generated based on the apparatus sound information acquired by the source sound acquisition means as the sound. It outputs through an output means.

  When the detection sound is generated based on the source sound, if the level of the original source sound is too small, the level of the output detection sound is also small, and it is difficult to detect the reflected sound. In general, in an interactive apparatus that is operated by an operator in an interactive manner, the user speaks slowly and loudly toward the operator in order to provide easy-to-understand explanations and guidance. Therefore, at the time of distance detection, it is possible to perform highly accurate and reliable distance detection by generating and using a detection sound (pseudo device sound) based on such device utterance (in this case, the detection timing is Preferably at the end of the device utterance). In addition, by using the voice uttered by the device for dialogue, it is relatively difficult for the operator to recognize that the sound is detected using sound, and sound information (frequency that is convenient for distance calculation) Or amplitude) can be used to speed up analysis after voice input.

  In a fourth aspect based on the first aspect, the source sound acquisition means generates ambient sound information as the source sound information from the ambient sound as the source sound generated around the apparatus and input by the audio input means. The detection sound output means acquires and outputs the detection sound generated based on the ambient sound information acquired by the source sound acquisition means via the sound output means.

  When detecting the distance, the detection of the detection sound (pseudo-ambient sound) based on the ambient sound input by the voice input means is used, and the distance detection is realized without the operator realizing that the sound is detected. It can be performed.

  According to a fifth invention, in any one of the first to fourth inventions, when the distance to the operator detected by the distance detection means becomes a predetermined value or less, the gain adjustment by the sensitivity adjustment means is started. It is characterized by having a start control means.

  In the case of an interactive device that operates in an interactive manner, an operator generally utters after relatively approaching the device. Therefore, when the distance from the apparatus to the operator is long, the possibility that the operator starts the operation is low. In the fifth invention of this application, in response to this, the start control means starts adjusting the gain of the voice input means after the distance to the operator is relatively short. Thereby, useless adjustment operation can be avoided and efficient processing can be performed.

  According to a sixth invention, in the fifth invention, when the distance to the operator detected by the distance detecting means is larger than the predetermined value, a setting control means for setting the gain of the voice input means to a predetermined value or more. It is characterized by having.

  As described above, when the distance from the device to the operator is long, it is unlikely that the operator will start an interactive operation. However, depending on the operator (or depending on the situation), there is a possibility that the operator speaks at a relatively long distance. In this case, since the distance from the apparatus to the operator is long, the speech voice of the operator is input from the voice input means at a relatively low level as it is. Therefore, in the sixth invention of the present application, the setting control means sets the gain of the voice input means to a predetermined value or more to increase the signal level. Thereby, the possibility of recognition failure can be reduced even during such an operation from a distance.

  In a seventh aspect based on the sixth aspect, the voice input means by the sensitivity adjustment means when a predetermined period has elapsed after the dialog processing is completed using the voice input means whose gain has been adjusted by the sensitivity adjustment means. And an end control means for ending the gain adjustment.

  When the dialogue processing is performed after the sensitivity adjustment of the voice input means according to the distance to the operator, and the dialogue processing is completed for a while, the operator who has been talking has already moved to another place, and the device There is a high possibility that no one is in the vicinity. Accordingly, in the seventh invention of the present application, the end control means stops the gain adjustment of the voice input means after the predetermined period has elapsed after the end of the dialogue processing. As a result, it is possible to reliably realize a state of waiting for the next operator based on the gain value at that time.

  According to the present invention, it is not necessary to newly provide a dedicated sensor or microphone, and it is possible to detect a distance to the operator and perform a dialogue process with an appropriate gain.

It is a system configuration figure showing the whole visitor reception system composition of one embodiment of the present invention. It is a functional block diagram showing the function structure of the whole system of a visitor reception system. It is a figure showing an example of the display screen in a display part. It is a functional block diagram which shows the functional structure of a reception terminal. It is a functional block diagram showing the functional structure of DB server. It is explanatory drawing explaining the outline | summary of the procedure until it outputs pseudo noise from a speaker. It is the figure explaining the outline | summary of the method of detecting the distance to a visitor, and the content of the gain control when the detected distance is larger than a predetermined value. It is a figure explaining the content of the gain control when the detected distance is below a predetermined value. It is the figure which represented typically the state after the reception process completion by a reception terminal. It is a flowchart showing the control procedure performed by the control circuit part of a reception terminal. It is a flowchart showing the control procedure performed by the control circuit part of a reception terminal. It is explanatory drawing explaining the outline | summary of the procedure until it outputs a pseudo utterance voice from a speaker in the modification which uses a visitor's utterance voice as a source sound. It is a flowchart showing the control procedure performed by the control circuit part of a reception terminal. It is a flowchart showing the control procedure performed by the control circuit part of a reception terminal. It is explanatory drawing explaining the outline | summary of the procedure until it outputs a pseudo guidance audio | voice from a speaker in the modification which uses the guidance audio | voice of a reception terminal as a source sound.

  Hereinafter, an embodiment of the present invention will be described with reference to the drawings. In the present embodiment, the dialogue apparatus of the present invention is applied to a visitor reception system that performs reception work for visitors to buildings, companies, and other buildings, for example.

(A) Basic Configuration of System FIG. 1 is a system configuration diagram showing an overall configuration of a visitor reception system according to the present embodiment.

  In FIG. 1, a visitor reception system 1 is installed near the entrance of a company, for example, and has an acceptance terminal 20 (interactive device) that can be operated by an operator (in this example, a visitor to the company) M in an interactive manner. is doing. The reception terminal 20 is provided with a microphone 207 (voice input means) for inputting voice and a speaker 208 (voice output means) for outputting voice.

The reception terminal 20 performs a dialogue process with the visitor M (in this example, a reception process by a dialogue with the visitor M) and a distance to the visitor M using voices input to and output from the microphone 207 and the speaker 208. Detection is performed, and the gain (described later) of the microphone 207 is adjusted based on the detected distance. In the present embodiment, as a method of detecting the distance from the reception terminal 20 to the visitor M, a detection sound for distance detection (in this example, pseudo noise described later) is output from the speaker 208, and the pseudo noise is visited. The time required until the reflected sound is input to the microphone 207 is measured. Then, the distance to the visitor M is detected from the relationship that the required time is proportional to the distance to the visitor M. That is, if the distance to the visitor M is L and the required time is t,
L = c × t / 2 (Formula 1)
(The details will be described later with reference to FIG. 7). Note that c is the speed of sound (about 340 [m / s], but varies depending on the density and pressure of air as a medium).

The distance to the visitor M can be detected by solving the above (Formula 1). Based on the detected distance, the gain of the microphone 207 is adjusted (set) (details will be described later).
As illustrated in FIG. 1, the reception terminal 20 includes a display unit 210, the microphone 207, and the speaker 208. The display unit 210 is composed of, for example, a liquid crystal display. In this example, the display unit 210 is supported by an arm 211 with respect to a base 212 installed horizontally, and the surface direction is obliquely upward so as to be perpendicular to the line of sight of the visitor M. Facing. The microphone 207 is arranged in a substantially arc shape with the tip thereof facing the visitor M side with respect to the base 212.

  Note that the display unit 210 may be configured by a touch panel so that the visitor M can operate the displayed display screen while directly touching the screen.

  FIG. 2 is a functional block diagram showing the functional configuration of the entire system of the visitor reception system 1.

  In FIG. 2, the visitor reception system 1 includes a plurality of (two in this example) provided corresponding to the reception terminal 20, a DB server 10 constituted by a well-known personal computer, and employees of the company. ) An IP telephone 60 and an IP-PBX (Internet Protocol Private Branch Exchange) 50, which is a well-known switching apparatus that performs circuit switching of the plurality of IP telephones 60, all of which are connected via a router 40 Yes.

  The reception terminal 20 includes a terminal main body 20A, the display unit 210, the gain variable amplifier 217, the speaker 208, and the microphone 207 connected to the gain variable amplifier 217, which are connected to the terminal main body 20A. is doing.

  The microphone 207 converts the input sound into sound information and outputs the sound information to the variable gain amplifier 217.

  The variable gain amplifier 217 amplifies the audio information input from the microphone 207 (in this example, amplification in which the gain of the microphone 207 is determined by a control signal from the CPU 201 described later), and outputs the amplified information to the terminal body 20A. The gain of the microphone 207 is the ratio between input and output (output / input), that is, the degree of amplification (= amplification degree) performed by the gain variable amplifier 217 under the control of the CPU 201.

  The speaker 208 uses a voice signal input from the terminal main body 20A as a notification sound (guidance voice) for the visitor M or a detection sound for distance detection (in this example, pseudo noise described later. The sound may be converted into a later-described (1) and (2) modification) and output.

  FIG. 3 is a diagram illustrating an example of a display screen in the display unit 210. On this screen, a virtual person IM that performs a reception work, which is generated by a drawing program described later, is displayed together with an office-like background G when a reception process described later is started. In addition, a sentence B (abbreviated as “***” in the drawing) corresponding to the voice uttered from the speaker 208 is also displayed.

(B) Detailed Function of Reception Terminal FIG. 4 is a functional block diagram showing a functional configuration of the reception terminal 20.

  In FIG. 4, the terminal body 20A of the reception terminal 20 includes a control circuit unit 200, an input / output (I / O) interface 204, a hard disk device (HDD) 205, and a timer 209 that is a time measuring means. Yes.

  The control circuit unit 200 includes a CPU 201, a ROM 202 that stores programs necessary for basic operations of the receiving terminal 20 and setting values for the programs, and a RAM 203 that temporarily stores various data. The CPU 201 controls the overall operation of the reception terminal 20 according to programs stored in the ROM 202 and the HDD 205.

  The I / O interface 204 includes the CPU 201, the hard disk device 205, the timer 209, the display unit 210, the gain variable amplifier 217, the speaker 208, and a network (NW) card 206. It is connected.

  The HDD 205 includes a language model storage area 252 for use in speech recognition, a word dictionary storage area 253, a visitor dictionary storage area 254 used for speech recognition for specifying visitors, and a program storage area 256. A plurality of storage areas are provided.

  In the program storage area 256, for example, a plurality of programs for controlling various operations of the reception terminal 20 are stored. Examples of the stored program include a system program that controls basic operations of the receiving terminal 20, a communication program that controls communication with the DB server 10, a drawing program that generates an image to be displayed on the display unit 210, and the like described above. Voice recognition program for executing voice recognition, DB collation program for accessing and collating database of DB server 10, voice synthesis program, dialogue control program, telephone connection program for connection between IP telephone 60 and IP-PBX 50 There are a distance detection program for controlling the above-described distance detection, a sensitivity adjustment program for controlling the gain adjustment of the microphone 207, and the like.

  Although not shown, the HDD 205 also stores a well-known acoustic model generally used in voice recognition processing, setting values used in various processing, and the like. Although not described in detail, the acoustic model is a statistical model of the acoustic features of speech. For example, for each vowel and consonant, a phoneme corresponding to the acoustic feature (for example, frequency characteristics) It is expressed by.

  The NW card 206 is connected to the router 40 and is an expansion card for enabling data transmission / reception with the DB server 10 or the like.

(C) Detailed Function of DB Server FIG. 5 is a functional block diagram showing a functional configuration of the DB server 10.

  As shown in FIG. 5, the DB server 10 includes a CPU 101, a ROM 102 and a RAM 103 connected to the CPU 101, an input / output (I / O) interface 104 connected to the CPU 101, and an I / O interface 104. A mouse controller 106, a key controller 107, a video controller 108, a communication device 109, and a hard disk device (HDD) 150 are connected to each other.

  The ROM 102 stores various programs including the BIOS for operating the DB server 10. The RAM 103 temporarily stores various data. The CPU 101 governs overall control of the DB server 10 according to programs stored in the ROM 102 and an HDD 150 described later.

  A mouse 116, a keyboard 117, and a display 118 are connected to the mouse controller 106, the key controller 107, and the video controller 108, respectively. The communication device 109 is connected to the router 40 and can exchange data with an external device such as the reception terminal 20.

  The HDD 150 has a plurality of storages including a visitor reservation database (DB) storage area 151 for storing visit information related to the scheduled visitor M, an employee database (DB) storage area 155 for storing employee information, and a program storage area 156. Has an area.

  The program storage area 156 stores various programs for causing the DB server 10 to execute various processes such as a system program and a communication program. For example, those programs stored in a CD-ROM are installed via a CD-ROM drive (not shown) and stored in the program storage area 156. Alternatively, a program downloaded from outside the system via an appropriate network may be stored.

(D) Flow of gain adjustment The greatest feature of the present embodiment configured as described above is that pseudo noise for distance detection is generated based on noise information acquired through the microphone 207 and output through the speaker 208. And detecting the distance to the visitor M based on the reflected sound information of the pseudo noise acquired via the microphone 207 and adjusting the gain of the microphone 207 based on the distance detection result. Hereinafter, the details will be described in order.

  FIG. 6 is an explanatory diagram for explaining an outline of the procedure until the pseudo noise is output from the speaker 208.

  FIG. 6A schematically shows a procedure for generating pseudo-noise information from noise input to the microphone 207. As shown in FIG. 6A, noise (ambient sound as a source sound that is a source of pseudo noise around the reception terminal 20 (in this example, a door 30 installed at a predetermined location in the company) This noise propagates and is input to the microphone 207. Thereby, noise information (ambient sound information) as source sound information including the amplitude (or frequency, which may be the same) corresponding to the input noise is acquired, and pseudo noise information is generated based on the noise information. . At this time, the pseudo-noise information may be generated so that the sampling frequency is 16 [kHz] or more. In that case, compatibility with a commercially available acoustic model can be ensured, so that convenience and applicability are further increased.

  FIG. 6B schematically shows a state in which pseudo noise is output from the speaker 208. As shown in FIG. 6B, when pseudo noise information is generated as shown in FIG. 6A, pseudo noise for distance detection is output from the speaker 208 based on the pseudo noise information. The pseudo noise is output from the speaker 208 within a predetermined time (for example, 1 [msec]) after the noise that is the generation source of the pseudo noise is input to the microphone 207.

  Also, the timer 209 (see FIG. 4) is started almost simultaneously with the pseudo-noise output from the speaker 208. As a result, the time required for the pseudo-noise to be reflected from the object after the pseudo-noise is output from the speaker 208 and the reflected sound to be input to the microphone 207 (hereinafter simply referred to as “required time”) is measured. (Measurement) is started. In the present embodiment, the object that is the reflection source of the reflected sound of the pseudo noise is considered to be the visitor M (estimation), and the distance to the visitor M is detected as described later.

  FIG. 7 is a diagram for explaining the outline of the method for detecting the distance to the visitor M and the content of the gain control according to the detected distance.

  In FIG. 7, the pseudo noise output from the speaker 208 as shown in FIG. 6B is propagated to a predetermined distance range (a distance range in which propagation is possible, depending on power). At this time, if the visitor M exists within the range, the pseudo noise is reflected by the visitor M, and the reflected sound propagates and is input to the microphone 207. Thereby, the corresponding reflected sound information is acquired. When the reflected sound is input to the microphone 207, the measurement of the required time performed by the timer 209 ends. That is, the measured value of the timer 209 at this time is the required time.

  Further, in this example, the acquisition of the reflected sound information is not started until a predetermined minimum sound wave receiving time elapses after the time measurement by the timer 209 is started. The minimum sound wave reception time is the time until the pseudo noise output from the speaker 208 is directly input to the microphone 207 without being reflected by the visitor M (= the so-called pseudo noise around the speaker 207 from the speaker 208) Time). For example, if the distance between the speaker 208 and the microphone 207 is 30 [cm], the minimum sound wave receiving time is 1.73 [msec]. The reflected sound of pseudo noise is not input to the microphone 207 until the measurement time of the timer 209 has passed the minimum sound wave reception time. Therefore, by waiting without starting the acquisition of the reflected sound information until the minimum sound wave reception time elapses, unnecessary sound (the above-described surrounding pseudo noise) input to the microphone 207 can be excluded.

  Furthermore, in this example, when a predetermined maximum sound wave reception time elapses after the timer 209 starts measuring, the acquisition of the reflected sound information is terminated and the acquisition of noise information is started again. It has become. The maximum sound wave reception time is the reflection of pseudo noise output from the speaker 208 from a predetermined distance (for example, the maximum distance at which reception processing by the reception terminal 20 can be performed), and the reflected sound is reflected from the microphone. This is the time required until it is input to 207. For example, when the maximum distance is 100 [cm], the maximum sound wave receiving time is 5.77 [msec]. After the maximum sound wave receiving time has elapsed, the reflected sound of the pseudo noise input to the microphone 207 is an object (a visitor M) that exists at a position farther than the predetermined distance (farther to accept). (Not limited). When the measurement time of the timer 209 exceeds the maximum sound wave receiving time, unnecessary reflected sound can be excluded by terminating the acquisition of reflected sound information.

  Further, since the pseudo noise and the reflected sound are both sound waves, they propagate between the reception terminal 20 and the visitor M at the speed of sound. The required time is a round-trip propagation time in which the pseudo noise and its reflected sound, that is, a sound wave, reciprocate between the reception terminal 20 and the visitor M (more specifically, a pseudo-range between the speaker 208 and the visitor M). The total time of the propagation time of the noise and the propagation time of the reflected sound between the visitor M and the microphone 207). That is, the product of the sound speed and half of the required time (= corresponding to the one-way propagation time) is the distance from the reception terminal 20 to the visitor M. For this reason, the distance from the reception terminal 20 to the visitor M can be detected (calculated) by solving the above (Formula 1) (see FIG. 1).

For example, if the sound speed is 346.5 [m / s] and the measured value of the timer 209 (= the above required time) is 2.0 [msec], the distance L to the visitor M is
L = 346.5 × 2.0 × 10 ⁻³ /2=346.5×10 ⁻³ [m] ≈35 [cm]
It becomes.

  Then, as shown in the lower part of FIG. 7, when the distance L from the reception terminal 20 to the visitor M detected by the above method is larger than a predetermined value L0 (for example, 80 [cm]) (L> L0) In this case, the gain of the microphone 207 (= the degree of amplification in the gain variable amplifier 217) is set to a predetermined value or more (in this example, fixed).

  On the other hand, the content of gain control when the detected distance L is equal to or less than the predetermined value L0 will be described with reference to FIG. FIG. 8 corresponds to FIG.

  In FIG. 8, in this example, the visitor M is slightly closer to the reception terminal 20 than in the state of FIG. 7, whereby the distance L from the reception terminal 20 to the visitor M becomes the predetermined value L0 or less. Yes. Then, in response to the detection of L ≦ L0, the adjustment of the gain of the microphone 207 is started, and the gain value is controlled in accordance with the detected distance.

  In the reception terminal 20 of the present embodiment, for example, the reception process is started at almost the same timing as the adjustment of the gain of the microphone 207 is started as described above (that is, when L ≦ L0 is satisfied) ( Alternatively, the reception process may be started by performing a predetermined start operation (refer to modifications (1) and (2) described later). The reception process is performed in an interactive manner while displaying a predetermined display screen on the display unit 210.

  That is, in the reception process, a predetermined guidance voice (for example, “Welcome. How is it?”) Is output from the speaker 208, and a predetermined display screen is displayed on the display unit 210 in conjunction with this (for example, , See FIG. 3 described above). When the visitor M speaks to the reception terminal 20 in accordance with the guidance voice and display, the voice of the corresponding visitor M is input by the microphone 207.

  At this time, information on the speech voice of the visitor M input to the microphone 207 is amplified by the variable gain amplifier 217 according to the gain (amplification degree) adjusted based on the detected distance. The amplified voice information is input to the terminal body 20A (see FIG. 2 and the like), and voice recognition is performed using the visitor dictionary stored in the visitor dictionary storage area 254 (for details, see FIG. 11). Will be described later). Further, the distance detection of the visitor M (that is, the visitor M who is operating the reception terminal 20) is also performed during the period in which the visitor M operates the reception terminal 20 (operation by speech) (see FIGS. 6 to 8). The gain is adjusted as needed based on the result of the detection distance.

  FIGS. 9A and 9B are diagrams schematically illustrating a state after the reception process by the reception terminal 20 is completed. Immediately after the reception process by the reception terminal 20 is completed, the visitor M is present in the vicinity of the reception terminal 20 that is still performing the dialogue operation (the state shown in FIG. 9A). When a predetermined period (for example, 10 seconds) elapses after the reception process ends, the visitor M present in the vicinity of the reception terminal 20 waits in the waiting room according to the reception result in the reception process (for example, instructed by the person in charge). The reception terminal 20 moves away to another place (for example, a waiting room or a room where a person in charge is present), and no one is in the vicinity of the reception terminal 20 (see FIG. 9 (b)).

  In the present embodiment, the state shown in FIG. 9B, that is, when a predetermined period has elapsed after the reception process is completed, the gain of the microphone 207 is set to the currently set gain (after the reception process is completed, In this example, the gain is fixedly set to (gain immediately before a predetermined period elapses), and the gain adjustment performed during the reception process is terminated. Then, with the gain fixed in this manner, the process returns to the state shown in FIG. 6A and waits for the next visitor M to come.

(E) Control Procedure FIG. 10 is a flowchart showing a control procedure executed by the control circuit unit 200 of the receiving terminal 20 in order to realize the contents described above. Note that the processing shown in this flow is a program group for visitor reception processing stored in the program storage area 256 of the HDD 205 (the aforementioned system program, drawing program, voice recognition program, dialogue control program, distance detection program, sensitivity adjustment). The program is executed by the CPU 201 in accordance with a program or the like.

  In FIG. 10, for example, this flow is started when the receiving terminal 20 is turned on (“START” position).

  First, in step S10, a predetermined initialization process is executed (at this time, the adjustment start flag Fs indicating that the adjustment of the gain of the microphone 207 is started is also initialized to Fs = 0).

  In step S20, noise that is a source of generation of pseudo noise for distance detection is input via the microphone 207, the gain variable amplifier 217, and the I / O interface 204, and noise information is acquired (source sound acquisition means). As a function).

  Thereafter, in step S30, a predetermined process is performed on the noise information acquired in step S20 to generate pseudo noise information.

  In step S40, the generated pseudo noise information is output to the speaker 208 via the I / O interface 204, and the pseudo noise is output (function as detection sound output means).

  Thereafter, in step S50, a control signal is output to the timer 209 via the I / O interface 204 to start the timer 209. Thereby, the pseudo noise output in step S40 is reflected on the object (visitor M), and measurement (time measurement) is started until the reflected sound is input to the microphone 207 in step S70 described later. Is done.

  Then, the process proceeds to step S60, and based on the measurement time of the timer 209, it is determined whether or not the measurement time has passed the minimum sound wave reception time. The determination is not satisfied until the minimum sound wave receiving time elapses, and the loop stands by. When the minimum sound wave receiving time elapses, the determination is satisfied, and the process proceeds to step S70.

  In step S <b> 70, it is determined whether or not the reflected sound (on the object) of the pseudo noise is input via the microphone 207, the gain variable amplifier 217, and the I / O interface 204. This determination may be made by a known method such as comparing the power spectrum of the pseudo-noise information with the information of the sound input via the microphone 207, the gain variable amplifier 217, and the I / O interface 204. When the reflected sound of the pseudo noise is not input, the determination is not satisfied and the routine goes to Step S80.

  In step S80, based on the measurement time of the timer 209, it is determined whether or not the measurement time has exceeded the aforementioned maximum sound wave reception time. If the maximum sound wave receiving time has not elapsed, the determination is not satisfied and the routine returns to step S70 and the same procedure is repeated. If the maximum sound wave receiving time has elapsed, the determination is satisfied, the process returns to step S20, and the same procedure is repeated.

  On the other hand, when the reflected sound of the pseudo noise is input in step S70, the determination in step S70 is satisfied and the process proceeds to step S90.

  In step S90, the reflected sound information including the corresponding amplitude or frequency is acquired from the reflected sound of the pseudo noise input via the microphone 207, the gain variable amplifier 217, and the I / O interface 204 in step S70. .

  After that, in step S100, based on the reflected sound information acquired in step S90 and the measurement time at this time of the timer 209 started in step S60, a predetermined calculation process (in this example, the above-described calculation process is performed). The method using the above formula 1 described with reference to FIG. 1 and FIG. 7 is performed, and it is estimated that the target object that is the reflection source of the reflected sound is the visitor M, and the distance L to the visitor M is determined. Detect (function as distance detection means).

  In step S110, based on the distance detection result in step S100, it is determined whether or not the detected distance L to the visitor M is equal to or less than the predetermined value L0. If L> L0, the determination is not satisfied and the routine goes to Step S120.

  In step S120, the gain of the microphone 207 is fixedly set to a predetermined value or more (function as a setting control unit). For example, the maximum value in a controllable range as the gain of the microphone 207 may be used. As a result, the audio information input via the microphone 207 is amplified by the gain variable amplifier 217 with the set gain value. Then, it returns to said step S20 and repeats the same procedure.

  On the other hand, if L ≦ L0 in step S110, the determination is satisfied, and the routine goes to step S130.

  In step S130, an appropriate (best) microphone 207 gain is calculated based on the distance detection result in step S100.

  In step S140, the gain of the microphone 207 is adjusted and set based on the calculation result in step S130 (function as sensitivity adjustment means). As a result, the audio information input via the microphone 207 is amplified by the gain variable amplifier 217 with the set gain.

  Thereafter, in step S150, the adjustment start flag Fs is set to Fs = 1 indicating gain adjustment start, and the process proceeds to step S160.

  In step S160, it is determined whether or not a reception end flag Ft indicating that the reception process has ended is Ft = 1. If Ft = 0 remains (= the reception process has not ended), the determination is not satisfied and the routine returns to step S20 and the same procedure is repeated. When Ft = 1 (= when the acceptance process is completed), the determination is satisfied and the routine goes to Step S180.

  In step S180, after the reception process is completed using the microphone 207 whose gain has been adjusted in step S140, in other words, in step S360 of FIG. 11 described later, the reception end flag Ft represents the end of the reception process Ft = After becoming 1, it is determined whether or not a predetermined period has elapsed. After Ft = 1, the determination is not satisfied until the predetermined period elapses, and the loop waits. When the predetermined period elapses, the determination is satisfied, and the process proceeds to step S190.

  In step S190, the gain of the microphone 207 is fixed to the gain set in the previous step S140 (that is, the gain value at this time), and the adjustment of the gain in the step S140 is ended (end control means). As a function). Thereafter, the process proceeds to step S195.

  In step S195, the adjustment start flag Fs is set to Fs = 0. Then, this flow ends. This flow is repeatedly executed at a predetermined time interval (for example, every 2 seconds) while the receiving terminal 20 is powered on or until a predetermined end operation is performed, for example.

  In the above, step S70 and step S90 function as reflected sound acquisition means described in each claim, and step S110 functions as start control means.

  FIG. 11 is a flowchart showing a control procedure executed by the control circuit unit 200 of the reception terminal 20 in parallel with the flow of FIG. The flow in FIG. 10 relates to the adjustment of the gain of the microphone 207 based on the distance to the visitor M, whereas the flow in FIG. 11 relates to the reception process. Note that these two flows in FIGS. 10 and 11 are performed by the CPU 201 in accordance with the above-described program for visitor reception processing by a known method similar to “multitask processing” often performed by a computer OS or the like, for example. It is designed to be processed in parallel.

  In FIG. 11, for example, this flow is started by turning on the power of the receiving terminal 20 (“START” position).

  First, in step S200, the acceptance end flag Ft is initialized to Ft = 0.

  Thereafter, in step S210, it is determined whether or not the adjustment start flag Fs is Fs = 1 indicating the start of gain adjustment. If Fs = 0 remains (= gain adjustment has not been started), the determination is not satisfied and the loop waits. When Fs = 1 (= when gain adjustment is started), the determination is satisfied, and the routine goes to Step S220.

  In step S220, the visitor dictionary stored in the visitor dictionary storage area 254 of the HDD 205 is acquired while referring to the visitor reservation database 1510 of the DB server 10 in order to recognize the speech voice of the visitor M, and the dictionary is updated. I do. That is, as described above, with reference to the time at which the reception process is started, in other words, the time at which the determination in step S210 is satisfied, within a predetermined time before and after the reservation data in the visitor reservation database 1510 (for example, A dictionary created on the basis of the reservation data with a scheduled visit date and time within one hour or the like) is acquired.

  Then, the process proceeds to step S230, where an audio signal is output to the speaker 208 via the I / O interface 204, and a visit is made, "Welcome. Please enter your name into the microphone." The guidance voice including the dialogue which asks the person's name is output. At this time, a display screen including text having the same content as the dialogue may be displayed on the display unit 210 (the same applies to Step S260, Step S310, Step S330, and Step S340 described later).

  Thereafter, in step S240, information on the voice of the visitor M who has spoken in response to this inquiry is input via the microphone 207, the gain variable amplifier 217, and the I / O interface 204 (the gain variable amplifier at this time). (Amplification in 217 is executed with the gain adjusted in step S140 in FIG. 10), and voice recognition is performed using the visitor dictionary acquired in step S220.

  Then, the process proceeds to step S250, and it is determined whether or not the input speech voice information of the visitor M has been recognized. If the speech cannot be recognized as a language, the determination in step S250 is not satisfied, and the visitor has not been able to recognize the speech, "The speech could not be recognized. Enter your name again into the microphone." The guidance voice including the dialogue to be notified to M is output to the speaker 208, the process returns to step S240, and the same procedure is repeated. Although not shown in FIG. 11, the guidance voice is output a predetermined number of times, and if voice recognition cannot be performed during that time, a corresponding process (for example, a department representative or other Etc.) (the same applies to step S280, step S300, step S330, and step S340 described later). On the other hand, if the speech can be recognized as the language, the determination in step S250 is satisfied, and the process proceeds to step S260.

  In step S260, a voice signal is output to the speaker 208 via the I / O interface 204, and a guidance voice including a dialogue that asks the name of the person in charge "Please input the person in charge name toward the microphone" is output. .

  Thereafter, in step S270, the speech voice information of the visitor M who has spoken in response to this question is input via the microphone 207, the gain variable amplifier 217, and the I / O interface 204 (same as above, the gain at this time). The amplification by the variable amplifier 217 is executed by the gain adjusted in step S140 of FIG. 10), and voice recognition is performed by the same method as in step S240.

  Then, the process proceeds to step S280, where it is determined whether or not the input speech voice information of the visitor M has been recognized. If speech recognition could not be performed as a language, the determination in step S250 was not satisfied, and it was visited that speech recognition could not be performed such as “Sound could not be recognized. The guidance voice including the dialogue to be notified to the person M is output to the speaker 208, the process returns to step S270, and the same procedure is repeated. If the speech can be recognized as the language, the determination is satisfied and the routine goes to Step S290.

  In step S290, the visitor reservation database 1510 of the DB server 10 is accessed, and the visitor name and the person-in-charge name obtained by the voice recognition of the voice information in the above step S240 and step S270 are either one of the visitor reservation database 1510. It is checked whether or not it matches the “visitor name” and “person in charge name” of the reservation data. In addition, it is possible to collate whether or not the matching is within a range having a certain similarity width and an allowable width instead of a perfect match.

  Thereafter, in step S300, it is determined whether or not the collation result in step S290 matches (matches). If the collation result in step S290 does not match (there is no reservation data for the corresponding visitor name and person in charge), the determination is not satisfied, and “Reservation has not been made. The guidance voice including the dialogue notifying the visitor M that the reservation data “please input your name” does not exist is output to the speaker 208, the process returns to step S240, and the same procedure is repeated. If the collation results in step S290 match (there are reservation data for the corresponding visitor name and person in charge), the determination is satisfied and the routine goes to step S310.

  In step S310, an audio signal is output to the speaker 208 via the I / O interface 204, along with the reservation content, “Are you sure you want this content? (Yes” if you want, “No” if wrong). To the microphone) and output a guidance voice that includes a dialogue asking for final confirmation.

  Then, the process proceeds to step S320, where the voice information of the visitor M who has spoken in response to this question is input via the microphone 207, the gain variable amplifier 217, and the I / O interface 204 (same as above, the gain at this time). The amplification by the variable amplifier 217 is executed by the gain adjusted in step S140 of FIG. 10), and voice recognition is performed by the same method as in step S240.

  Thereafter, in step S330, it is determined whether or not the input voice information from the visitor M has been recognized. If the voice cannot be recognized as the language, the judgment in step S330 is not satisfied, and the voice recognition "Could not recognize voice. Enter" Yes "or" No "again into the microphone cannot be recognized. The guidance voice including the dialogue for notifying the visitor M is output to the speaker 208, and the process returns to step S320, and the same procedure is repeated, and if the voice can be recognized as the language, the determination is satisfied and the process returns to step S340. Move.

  In Step S340, it is determined whether or not the information (“Yes” or “No”) acquired by the voice recognition of the voice information in Step S320 is “Yes”. If the answer is “No”, the judgment is not satisfied, and the guidance includes a dialogue to notify the visitor M of the operation re-execution of “Retry the operation from the beginning. Enter your name again into the microphone”. The sound is output to the speaker 208, the process returns to step S230, and the same procedure is repeated. If “yes”, the determination is satisfied and the routine goes to Step S350.

  In step S350, in response to confirming that the legitimate visitor M has visited, a call is made to the IP phone 60 of the corresponding person in charge. Specifically, a notice sentence to the person in charge is created, the text data of the notice sentence is converted into voice data, and the telephone number of the person in charge specified by the reservation data is used to transmit the notice sentence via the IP-PBX 50. Voice data is transmitted to the IP telephone 60 used by the person in charge.

  In step S360, the reception end flag Ft is set to Ft = 1 indicating the end of the reception process, and then this flow ends. This flow is repeatedly executed at a predetermined time interval (for example, every 2 seconds) while the receiving terminal 20 is powered on or until a predetermined end operation is performed, for example.

  As described above, in the reception terminal 20 of the present embodiment, the distance to the visitor M is detected using the detection sound for distance detection (pseudo noise in the above example). That is, when noise that is a generation source at the time of generating pseudo noise is input via the microphone 207, noise information including the corresponding amplitude is acquired (see step S20 in FIG. 10), and pseudo noise based on the noise information is acquired from the speaker 208. (See step S40 in FIG. 10). Then, the distance to the visitor M is detected based on the output reflected sound information of the pseudo noise (see step S100 in FIG. 10).

  Then, based on the detected distance to the visitor M, an appropriate (best) microphone 207 gain corresponding to the distance is calculated and gain adjustment is performed (see step S140 in FIG. 10). That is, when the distance to the visitor M is relatively close, the utterance voice of the visitor M at the time of dialogue is input at a relatively high level, and thus the gain is set low. On the other hand, when the distance to the visitor M is relatively long, the utterance voice of the visitor M at the time of dialogue is input at a relatively small level, and thus the gain is set high. Thereby, the reception process can be performed with an appropriate signal level. As a result, it is possible to perform reliable reception processing with no recognition failure.

  As described above, according to the reception terminal 20 of the present embodiment, the distance to the visitor M can be detected using the sound input / output via the microphone 207 and the speaker 208. That is, by utilizing the microphone 207 and the speaker 208 that are originally provided for the reception process, distance detection can be performed without newly providing a separate sensor or a dedicated microphone for distance detection, Reliable reception processing can be performed with an appropriate gain based thereon.

  In the present embodiment, in particular, noise information is acquired from noise input from the microphone 207, and pseudo noise generated based on the acquired noise information is output via the speaker 208. In this way, at the time of distance detection, by generating and using pseudo noise based on the noise input by the microphone 207, the distance is detected without the visitor M realizing that it is detected using sound. be able to.

  In addition, when an operation is performed interactively at the reception terminal 20, the visitor M generally utters after being relatively close to the reception terminal 20, and the distance from the reception terminal 20 to the visitor M is long. In this case, the possibility that the visitor M starts the operation is low. Accordingly, in response to this, particularly in the present embodiment, the adjustment of the gain is started after the distance L to the visitor M detected in step S100 in FIG. That is, by starting the gain adjustment after the detected distance to the visitor M is relatively short, useless adjustment operation can be avoided and efficient processing can be performed.

  In addition, when the distance from the reception terminal 20 to the visitor M is far, the possibility that the visitor M performs a dialogue operation is low. However, depending on the visitor M (or depending on the situation), the utterance is kept at a relatively long distance. There is also a possibility to do. At this time, since the distance from the reception terminal 20 to the visitor M is long, the uttered voice of the visitor M is input from the microphone 207 at a relatively low level as it is. Accordingly, in the present embodiment, in particular, when the distance L to the visitor M detected in step S100 in FIG. 10 is larger than a predetermined value L0, the gain of the microphone 207 is set to a predetermined value or more (see FIG. 10 step S120), the signal level is increased to some extent. As a result, the possibility of recognition failure can be reduced even during a dialog operation from a distance as described above.

  In addition, the reception process is performed after the gain of the microphone 207 is adjusted in accordance with the distance from the visitor M, and when the reception process is completed, the visitor M who has been in conversation has already moved to another location. However, there is a high possibility that there is no one in the vicinity of the reception terminal 20 (the state shown in FIG. 9B). Correspondingly, in the present embodiment, the gain adjustment is finished when a predetermined period has elapsed after the reception process is completed using the microphone 207 whose gain is adjusted (see step S190 in FIG. 10). . That is, in this case, when a predetermined period elapses after the end of the acceptance process, the gain is fixed to the gain set immediately before, the gain adjustment is terminated, and the adjustment is not performed. Thereby, the waiting state of the next visitor M can be reliably realized by the gain value at this time.

  The present invention is not limited to the above-described embodiment, and various modifications can be made without departing from the spirit and technical idea of the present invention. Hereinafter, such modifications will be described in order.

(1) When a visitor's speech is used as a source sound In the above embodiment, a source sound that is a generation source of a detection sound for distance detection (pseudo noise in the above example) is used around the reception terminal 20. Although the generated noise was used, it is not limited to this. That is, as the source sound, a speech voice uttered by the visitor M in the reception process may be used.

  FIG. 12A and FIG. 12B are explanatory diagrams illustrating an outline of a procedure until the pseudo utterance voice is output from the speaker 208 in the present modification.

  In the reception terminal 20 of the present modification, unlike the above embodiment, for example, the reception process is started when a visitor M appropriately operates an operation unit (not shown). Then, as shown in FIG. 12A, when the visitor M speaks to the reception terminal 20 during the reception process, the utterance voice as the source sound uttered by the visitor M is propagated to the microphone 207. Entered. Thereby, the utterance voice information as the source sound information including the amplitude (or frequency) corresponding to the utterance voice of the input visitor M is acquired, and the pseudo utterance voice information is generated based on the utterance voice information. .

  Then, as shown in FIG. 12B, when the pseudo speech information is generated as described above, the pseudo speech (detected sound) is output from the speaker 208 based on the pseudo speech information. The pseudo utterance voice is output from the speaker 208 within a predetermined time (for example, 1 [msec]) after the utterance voice of the visitor M who is the generation source of the pseudo utterance voice is input to the microphone 207. It has become. The subsequent processing is almost the same as that in FIGS. 7 and 8 of the above-described embodiment. When the reflected sound of the pseudo utterance voice is input, the corresponding reflected sound information is acquired, and the distance to the visitor M is detected. Is called. Then, the gain of the microphone 207 is adjusted according to the detected distance.

  FIG. 13 is a flowchart showing a control procedure executed by the control circuit unit 200 of the reception terminal 20 in this modification, and corresponds to FIG. 10 described above. Portions equivalent to those in FIG. 10 are denoted by the same reference numerals and description thereof is omitted as appropriate.

  13, first, a predetermined initialization process is executed in step S10 ′ corresponding to step S10 in FIG. 10 (the fact that Fs = 0 is not initialized because the flag Fs is not used is step S10 in FIG. 10). Different). Then, the process proceeds to newly provided step S15, and it is determined whether or not the above-described reception start flag Fm indicating that the reception process has been started is Fm = 1. When Fm = 0 remains (= when the reception process has not been started or ended), the determination is not satisfied and the loop waits. When Fm = 1 (= when the reception process is started), The determination is satisfied, and the routine goes to Step S20 ′ corresponding to Step S20 in FIG.

  In step S20 ′, the utterance voice, which is uttered by the visitor M in the reception process of the reception terminal 20 (see FIG. 14 described later) and becomes the generation source of the pseudo utterance voice for distance detection, is converted into the microphone 207, the gain variable amplifier 217, and Input through the I / O interface 204, and utterance voice information including the corresponding amplitude or frequency is acquired (function as source sound acquisition means).

  Thereafter, in step S30 ′ corresponding to step S30 in FIG. 10, predetermined processing is performed on the utterance voice information acquired in step S20 ′ to generate pseudo utterance voice information.

  Then, the process proceeds to step S40 ′ corresponding to step S40 in FIG. 10, and the generated pseudo speech information is output to the speaker 208 via the I / O interface 204, and the pseudo speech is output from the speaker 208 ( Function as detection sound output means).

  Subsequent steps S50 and S60 are the same as in FIG. 10 described above. The timer 209 is started, the measurement time of the timer 209 waits until the minimum sound wave reception time elapses, and the minimum sound wave reception time elapses. Then, the process proceeds to step S70 ′ corresponding to step S70 in FIG.

  In step S70 ′, it is determined whether or not the reflected sound (on the object) of the pseudo utterance voice is input via the microphone 207, the gain variable amplifier 217, and the I / O interface 204. This determination may be performed by a known method such as comparing the power spectrum of the pseudo speech and the information of the voice input via the microphone 207, the gain variable amplifier 217, and the I / O interface 204. If the reflected sound of the pseudo utterance voice is not input, the determination is not satisfied and the routine goes to Step S80.

  Step S80 is the same as in FIG. 10 described above, and it is determined whether or not the measurement time of the timer 209 has passed the maximum sound wave reception time. Returning to step S70 ', if the maximum sound wave receiving time has elapsed, the process returns to step S20'.

  On the other hand, if the reflected sound of the pseudo utterance voice is input in step S70 ′, the determination in step S70 ′ is satisfied, and the process proceeds to step S90 ′ corresponding to step S90 in FIG.

  In step S90 ′, the corresponding reflected sound information is acquired from the reflected sound of the pseudo utterance speech input via the microphone 207, the variable gain amplifier 217, and the I / O interface 204 in step S70 ′.

  Subsequent steps S100 to S140 are the same as those in FIG. If the gain of the microphone 207 is adjusted in step S140, the process proceeds to step S160 ′ corresponding to step S160 in FIG.

  In step S160 ′, it is determined whether or not the reception start flag Fm has returned to Fm = 0. If Fm = 1 remains (= the accepting process is still being executed), the determination is not satisfied and the routine returns to step S20 ′ and the same procedure is repeated. When Fm = 0 is returned (= when the acceptance process is completed), the determination is satisfied, and the routine goes to Step S180 ′ corresponding to Step S180 in FIG.

  In step S180 ′, after the reception process is completed using the microphone 207 whose gain has been adjusted in step S140, in other words, after the reception start flag Fm becomes Fm = 0 in step S365 of FIG. 14 described later. It is determined whether a predetermined period has elapsed. After Fm = 0, the determination is not satisfied until the predetermined period elapses, and the loop waits. When the predetermined period elapses, the determination is satisfied, and the process proceeds to step S190.

  Step S190 is the same as that of FIG.

  In the above, step S70 'and step S90' function as reflected sound acquisition means described in each claim.

  FIG. 14 is a flowchart showing a control procedure executed by the control circuit unit 200 of the reception terminal 20 in parallel with the flow of FIG. 13, and corresponds to FIG. 11 described above. Portions equivalent to those in FIG. 11 are denoted by the same reference numerals, and description thereof is omitted as appropriate.

  In FIG. 14, first, in step S205 provided in place of step S200 in FIG. 11, the reception start flag Fm indicating that the reception process is started is initialized to Fm = 0.

  Thereafter, in step S212 newly provided, it is determined whether or not an operation for starting the reception process has been performed by the visitor M via an operation unit (not shown). The loop waits until an operation for starting the reception process is performed by the visitor M. When the operation for starting the reception process is performed by the visitor M, the determination is satisfied, and the process proceeds to step S214 newly provided.

  In step S214, the reception start flag Fm is set to Fm = 1 indicating the start of the reception process, and the process proceeds to step S220.

  Subsequent steps S220 to S350 are the same as those in FIG. In step S350, in response to confirming that the legitimate visitor M has visited, if a call is made to the IP phone 60 of the corresponding person in charge, it is provided instead of step S360 in FIG. The process proceeds to step S365.

  In step S365, after the reception start flag Fm is set to Fm = 0, this flow is finished.

  According to this modification, the following effects can be obtained.

  That is, when the detection sound is generated based on the source sound that is the generation source of the detection sound for distance detection, if the level of the source sound is too low, the level of the detection sound output through the speaker 208 is also small and the reflection thereof It may be difficult to detect sound.

  This modification can cope with the above case. That is, normally, the visitor M who wants to perform an operation in an interactive manner at the reception terminal 20 works to intentionally recognize his / her speech as much as possible, and speaks slowly at a loud volume. Therefore, when the distance is detected, the pseudo utterance voice is generated and used based on the utterance voice of the visitor M as described above, which is input by the microphone 207, so that accurate and reliable distance detection can be performed. In addition, by using the uttered voice uttered by the visitor M himself, there is an effect that the visitor M is relatively less aware of the detection using sound.

(2) When the guidance voice of the reception terminal is used as the source sound In the above description, the noise generated around the reception terminal 20 or the voice of the visitor M is used as the source sound. I can't. That is, as the source sound, guidance voice output through the speaker 208 in the reception process may be used.

  FIG. 15 is an explanatory diagram illustrating an outline of a procedure until the pseudo guidance voice is output from the speaker 208 in the present modification.

  In the reception terminal 20 of this modification, as in the modification of (1) above, the reception process is started by appropriately operating an operation unit (not shown) by the visitor M, for example. Then, as shown in FIG. 15A, during the reception process, when the guidance voice (device voice) as the source sound is output from the speaker 208, this guidance voice is propagated and input to the microphone 207 ( So-called guidance voice from the speaker 208 to the microphone 207). Thereby, guidance voice information (device voice information) as source sound information including amplitude (or frequency) corresponding to the inputted guidance voice is acquired, and pseudo guidance voice information is generated based on this guidance voice information. The

  Then, as shown in FIG. 15B, when the pseudo guidance voice information is generated as described above, the pseudo guidance voice (detected sound) is output from the speaker 208 based on the pseudo guidance voice information. The pseudo guidance voice is output from the speaker 208 within a predetermined time (for example, 1 [msec]) after the guidance voice that is the generation source of the pseudo guidance voice is input to the microphone 207.

  At this time, in particular, the timing at which the pseudo guidance voice is output may be the time when the output of the guidance voice that is the generation source of the pseudo guidance voice is finished. That is, when the guidance voice is output from the speaker 208, the visitor M who is operating the reception terminal 20 by dialogue does not utter because the guidance voice is output, and the guidance voice is output. It is common to utter along the content of the guidance voice after the end. Therefore, when the timing of outputting the pseudo guidance voice is set to the time when the output of the guidance voice is finished (before the visitor M speaks), the distance to the visitor M when speaking is set as the distance. There is an effect that it can be detected more accurately.

  The subsequent processing is almost the same as in FIGS. 7 and 8 of the above embodiment, and the reflected sound information corresponding to the reflected sound of the pseudo guidance voice is input to the microphone 207, and the distance to the visitor M is obtained. Detection is performed. Then, the gain of the microphone 207 is adjusted according to the detected distance.

  As described above, in this modification, the guidance voice is output from the speaker 208 and received by the microphone 207 in the reception process with the visitor M (or a reflected sound of the guidance voice as described later). Information is acquired, and pseudo guidance voice generated based on the obtained guidance voice information is output via the speaker 208. As a result, as in the modified example of (1), it is possible to perform accurate distance detection with high accuracy and to obtain an effect that it is relatively difficult for the visitor M to detect using sound.

  In addition, since the guidance voice generated through the speaker 208, in other words, generated inside the reception terminal 20, is used as the source sound, the guidance voice information including the amplitude (frequency) convenient for distance detection is used. A pseudo guidance voice (information) can be generated. That is, when generating pseudo guidance voice (information) based on such guidance voice information, for example, by selecting and using one having a large amplitude of guidance voice information, the processing speed of analysis (for example, filtering analysis or the like) is increased. Can be fast.

  In the above example, the guide voice is used as the source sound. However, the reflected sound of the guide voice may be used as appropriate.

(3) Others In the visitor reception system 1, the reception terminal 20 that detects the distance to the visitor M and adjusts the gain of the microphone 207 and the DB server 10 are separate devices. However, the present invention is not limited to this. A server equipped with a microphone and a speaker is installed near the entrance of the company, and only the server performs distance detection to the visitor M, adjustment of the gain of the microphone 207, and reception processing. Also good. Further, information stored in the HDD 150 such as the visitor reservation DB 151 and the employee DB 155 may be stored in the HDD 205 on the reception terminal 20 side, or a separate connection that can be connected to the reception terminal 20 via a network. It is good also as a structure which memorize | stores in a memory | storage device and reads required information during a reception process.

  In the above description, the voice input unit is configured by one microphone 207, but is not limited thereto, and may be configured by a plurality of (for example, two) microphones (a so-called array type microphone device).

  In the above, as a predetermined calculation process, the time required from the detection sound being output through the speaker 208 until the reflected sound is input to the microphone 207 is measured. The distance to the visitor M was detected from the relationship proportional to the distance (see Equation 1 above). However, the present invention is not limited thereto, and as a predetermined calculation process, the distance to the visitor M may be detected from the phase difference between the output detection sound and the input reflection sound.

  In addition, in the above, the arrow shown in each figure of FIG.4, FIG.5 etc. shows an example of the flow of a signal, and does not limit the flow direction of a signal.

  In addition, the flowcharts shown in FIGS. 10, 11, 13, and 14 do not limit the present invention to the procedure shown in the above-described flow, and the procedure can be added without departing from the spirit and technical idea of the invention. You may delete or change the order.

  In addition to those already described above, the methods according to the above-described embodiments and modifications may be used in appropriate combination.

  In addition, although not illustrated one by one, the present invention is implemented with various modifications within a range not departing from the gist thereof.

20 Reception terminal (dialogue device)
201 CPU
207 Microphone (voice input means)
208 Speaker (Audio output means)
217 Variable gain amplifier M Visitor (operator)

Claims (7)

An interactive device that an operator can operate in an interactive manner,
Voice input means for inputting voice;
Audio output means for outputting audio;
Source sound acquisition means for acquiring source sound information including a corresponding amplitude or frequency from a source sound that is input via the sound input means and serves as a generation source of detection sound for distance detection;
Detection sound output for outputting the detection sound generated based on the source sound information acquired by the source sound acquisition means via the audio output means within a predetermined time after the sound input means inputs the sound. Means,
Reflected sound acquisition means for acquiring corresponding reflected sound information from the reflected sound of the detected sound at the object input via the voice input means;
Based on the reflected sound information acquired by the reflected sound acquisition means, distance calculation means for performing a predetermined calculation process, detecting that the object is the operator and detecting the distance to the operator;
And a sensitivity adjusting means for adjusting a gain of the voice input means based on a detection result of the distance detecting means. The source sound acquisition means is
An utterance voice as the source sound is obtained by an utterance voice as the source sound uttered by the operator in the dialogue processing of the dialogue apparatus,
The detection sound output means includes
2. The dialogue apparatus according to claim 1, wherein the detection sound generated based on the utterance voice information acquired by the source sound acquisition means is output via the voice output means. The source sound acquisition means is
The apparatus sound information as the source sound information is acquired from the apparatus sound as the source sound or the reflected sound of the apparatus sound output from the sound output means and input by the sound input means in the dialogue process with the operator. ,
The detection sound output means includes
2. The dialogue apparatus according to claim 1, wherein the detection sound generated based on the apparatus sound information acquired by the source sound acquisition means is output via the sound output means. The source sound acquisition means is
The ambient sound information as the source sound information is obtained by the ambient sound as the source sound generated around the device and input by the voice input means,
The detection sound output means includes
2. The dialogue apparatus according to claim 1, wherein the detection sound generated based on the ambient sound information acquired by the source sound acquisition means is output via the voice output means.   The start control means for starting the adjustment of the gain by the sensitivity adjustment means when the distance to the operator detected by the distance detection means becomes a predetermined value or less. Item 5. The interactive device according to any one of items 4 to 5.   6. The apparatus according to claim 5, further comprising setting control means for setting a gain of the voice input means to a predetermined value or more when a distance to the operator detected by the distance detection means is larger than the predetermined value. Interactive device.   After a dialogue process is completed using the voice input unit whose gain is adjusted by the sensitivity adjustment unit, a termination control unit is configured to end gain adjustment of the voice input unit by the sensitivity adjustment unit when a predetermined period has elapsed. The interactive apparatus according to claim 6.

Images