JP2010079103A - Voice interactive apparatus, program for the same, and voice interactive processing method - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a voice interactive apparatus for easily obtaining a way how to input a voice even if a user is not familiar with an information input which uses a voice. <P>SOLUTION: The voice interactive apparatus includes: a voice recognizing means for recognizing input contents based on the voice inputted into a voice input section; a scenario storing means for storing an interactive scenario where a voice guidance is associated with an input item which should be succeedingly inputted to the voice input section according to the voice guidance; a voice guidance outputting means for outputting the voice guidance according to the interactive scenario from a voice output section; and a display processing means for displaying the input item inputted to the voice input section according to the voice guidance on a display section according to the interactive scenario when the voice guidance is outputted from the voice output section. <P>COPYRIGHT: (C)2010,JPO&INPIT

Description

  The present invention relates to a voice interaction apparatus, a program thereof, and a voice interaction processing method. It relates to a processing method.

2. Description of the Related Art In recent years, a large number of voice interaction apparatuses that provide information and services required by users by interacting with users by voice have been proposed. In this type of voice interactive device, items to be input (input items) are presented to the user by outputting voice guidance, and the user's voice is recognized by recognizing the voice emitted by the user according to the voice guidance. A dialogue is performed (see, for example, Patent Document 1).
JP-A-11-212594

  However, in the conventional voice interactive apparatus, when the user is not used to inputting information by voice, it may be difficult to know how to input voice to the voice guidance.

  Therefore, the present invention provides a voice dialogue apparatus, a program thereof, and a voice dialogue processing method capable of grasping how to perform voice input even when the user is not accustomed to voice information input. The purpose is to provide.

  To achieve this object, the invention according to claim 1 is a scenario storage means for storing a dialogue scenario in which voice guidance is associated with input items to be input to the voice input unit in accordance with the voice guidance. Voice guidance output means for outputting voice guidance according to the dialogue scenario from the voice output unit, and when the voice guidance is output from the voice output unit, the voice guidance is input to the voice input unit according to the voice guidance. A speech processing unit for displaying input items to be displayed in a predetermined display format on a display unit according to the dialogue scenario, and a speech recognition unit for recognizing input contents based on speech input to the speech input unit It was a dialogue device.

  Further, the invention according to claim 2 is the voice dialogue apparatus according to claim 1, wherein a candidate display unit that displays a plurality of input content candidates recognized by the voice recognition unit on the display unit, and the candidate A selection unit that selects any one candidate from a plurality of input content candidates displayed by the display unit, and an input processing unit that determines the candidate selected by the selection unit as input content. And

  Further, the invention according to claim 3 is the voice interaction device according to claim 1 or 2, wherein the display processing means is configured to execute the operation when a predetermined period has elapsed after the input item displayed on the display unit is designated. An example of input for an input item is displayed.

  According to a fourth aspect of the present invention, in the voice interactive apparatus according to any one of the first to third aspects, the display processing means is configured such that an input item displayed on the display unit is designated. When a voice is input to the voice input unit, the display format of the input item is changed.

  According to a fifth aspect of the present invention, in the spoken dialogue apparatus according to any one of the first to third aspects, the display processing means is configured such that the input item displayed on the display unit is designated. When the voice recognition process is started, the display format of the input item is changed.

  The invention according to claim 6 is the voice interactive apparatus according to any one of claims 1 to 5, wherein the display processing means outputs a voice guidance by the voice guidance output means, and then a predetermined period has elapsed. In this case, the input item is displayed on the display unit.

  According to a seventh aspect of the present invention, in the voice interactive apparatus according to any one of the first to sixth aspects, the display processing means should be input to the voice input unit in accordance with the voice guidance. When an arbitrary input item is displayed on the display unit in addition to the essential input item, the display format is changed between the essential input item and the arbitrary input item.

  According to an eighth aspect of the present invention, in the spoken dialogue apparatus according to any one of the first to seventh aspects, the display processing means is configured such that the input content for the input item is a predetermined input content. An input item subordinate to the already displayed input item is displayed on the display unit.

  Further, the invention according to claim 9 is the voice interaction device according to any one of claims 1 to 8, wherein the voice recognition means has a plurality of voice recognition dictionaries, and the display unit The speech recognition dictionary corresponding to the input item displayed on the screen is selected to recognize the speech input to the speech input unit.

  The invention according to claim 10 is the voice interaction device according to any one of claims 1 to 9, wherein the display processing means should be input to the voice input unit according to the voice guidance. When there are a plurality of input items, each of the plurality of input items is displayed on the display unit in a predetermined display format, and the voice recognition means has a plurality of voice recognition dictionaries, Select a speech recognition dictionary corresponding to the input content of the input item that has already been recognized, and recognize the voice input to the voice input unit for the input item that has not been recognized among the plurality of input items, This is performed using the selected speech recognition dictionary.

  The invention described in claim 11 is a program that causes a computer to function as each unit of the voice interactive apparatus according to any one of claims 1 to 10.

  According to the twelfth aspect of the present invention, when the voice guidance according to the dialogue scenario stored in the storage unit is output from the voice output unit, and when the voice guidance is output from the voice output unit, In response to the input items to be input to the voice input unit in accordance with the recognized input content and the dialogue scenario in a predetermined display format, and based on the voice input to the voice input unit And a step of recognizing input contents.

  According to the present invention, when the voice guidance is output from the voice output unit, the input items to be input to the voice input unit according to the voice guidance are displayed on the display unit in a predetermined display format. Even if the user is not used to inputting information by voice, it is easy to understand what content should be input and how.

[1. Outline of Spoken Dialogue Processing Device]
An outline of a voice interaction apparatus according to an embodiment of the present invention will be described with reference to the drawings. FIG. 1 is an explanatory diagram showing a schematic configuration of a voice interaction device, FIG. 2 is an external view of the voice interaction device in this embodiment, and FIG. 3 is an explanatory diagram of a voice interaction processing method.

  The voice interaction device is capable of providing information and services required by the user by interacting with the user by voice. As shown in FIG. 1, the voice dialogue and voice according to the voice guidance are provided. Scenario storage means for storing a dialogue scenario in which input items to be input to the input unit are associated with each other, voice guidance output means for outputting voice guidance according to the dialogue scenario from a voice output unit, and the voice guidance is Display processing means for displaying input items to be input to the voice input unit according to the voice guidance in a predetermined display format according to the dialogue scenario when output from the voice output unit; and the voice input unit Voice recognition means for recognizing the input content based on the input voice.

  Here, the dialogue scenario in the present embodiment includes a plurality of set question items, a plurality of user answer items predicted for each question item, new question items and confirmation items for each answer item, etc. However, it is constructed according to various situations.

  In accordance with this dialogue scenario, voice guidance is output from the voice output unit, and input items to be input to the voice input unit in accordance with the voice guidance are displayed on the display unit in a display format that is easy for the user to understand. .

  Therefore, since the input content requested by the voice guidance is displayed as an input item on the display unit, the user can easily grasp what content should be input and how.

  Such a voice interaction device can be installed, for example, at the reception of a company or the like, and can be used as an automatic reception device that can be relayed to a department or employee desired by the customer by means of a voice conversation with a visitor visiting the company. It is. At this time, as a dialogue scenario, a visitor reception scenario is used in which voice guidance and expected answers are routed until a visitor is relayed to a predetermined department or specific employee.

  As shown in FIG. 2, the voice interactive apparatus according to the present embodiment applied to the automatic reception apparatus is installed in the reception counter 10, and on the front surface of the housing 1, a touch panel display 2 that is a display unit, A microphone 3 that is an audio input unit, a speaker 4 that is an audio output unit, and an infrared sensor 5 that detects the presence of a user are provided.

  An information processing apparatus such as a computer including a control unit, a storage unit (FIG. 1), and an input / output unit, and other devices are mounted on the housing 1 as necessary. The functions as the voice guidance output means, display processing means, and voice recognition means are mainly handled by the information processing apparatus as the control section, and the scenario storage means functions as a storage section including a hard disk of the information processing apparatus. I'm in charge. In addition, various dictionaries associated with the dialogue scenario are stored in the storage unit.

  The storage unit stores a voice interaction program for causing the information processing apparatus to function as each of the above-described means (see FIG. 13). According to the voice interaction processing program, the control unit is operated by the voice interaction device. Perform voice interaction processing. The voice interaction program is recorded on various storage media such as a CD, a DVD, a flexible disk, or a flash memory, and is read from these and stored in the storage unit.

  In this way, the information processing apparatus functioning as a voice recognition unit according to the voice dialogue program outputs voice guidance from the speaker 4 and collates voice signals input by the user from the microphone 3 with various dictionaries. Character string data corresponding to the utterance content is generated, and voice guidance is further output based on this data and the dialogue scenario, so that voice dialogue with the user proceeds.

  That is, in the voice dialogue processing method by the voice dialogue apparatus according to the present embodiment using such a program, as shown in FIG. 3, a procedure S1 for outputting voice guidance according to the dialogue scenario stored in the storage unit from the voice output unit. And when the voice guidance is output from the voice output unit, input items to be input to the voice input unit according to the voice guidance are displayed on the display unit according to the recognized input content and the dialogue scenario. There is a procedure S2 for displaying in a predetermined display format, and a procedure S3 for recognizing the input content based on the voice input to the voice input unit.

  For example, when a visitor is detected, voice guidance such as “Come here. Please tell me your business while checking the guidance on the screen. Please tell me your name and affiliation.” Is output from the speaker 4. Then, input items to be voice input by the customer are displayed on the touch panel display 2 according to the voice guidance. As the display mode of the input items, the customer's own company name input section and name input section are displayed in a predetermined display format. On the other hand, the visitor speaks while confirming the contents to be voice-input while looking at the displayed company name input section and name input section. The voice interactive apparatus recognizes an utterance (voice) input through the microphone 3. Then, voice guidance output and input item display, voice input by a visitor, and voice recognition by a voice dialogue device proceed according to a dialogue scenario (customer acceptance scenario).

  As described above, the voice interaction apparatus according to the present embodiment performs a display display corresponding to the voice guidance, instead of performing the dialogue progress only with the voice of the apparatus-side voice guidance and the user's utterance. Thus, even a user who is not used to inputting information by voice can easily grasp how to input voice. That is, when voice guidance is output from the speaker 4, the input items to be input to the microphone 3 in accordance with the voice guidance are displayed in a predetermined display format on the touch panel display 2, and the user views this display. The voice can be input.

  For voice guidance, the user performs voice input, but when the user inputs information by voice, the user grasps whether the information input is actually input to the voice interactive device. I can't do that, and I can be anxious.

  Therefore, the display processing means is configured to change the display format of the input item when an input of the voice is input to the voice input unit when the input item displayed on the display unit is designated. Here, “the input item displayed on the display unit is designated” means that the user touches (touches) the input item displayed on the touch panel display 2 and the control unit touches the input item. The signal is being detected.

  In other words, when the user inputs a voice while touching the displayed input item with a finger, when the voice dialogue apparatus accepts this voice input, the display format is changed and the voice is accepted by the user. This is notified. The input item may be specified with a single touch of a finger, but when voice input is performed so that the control unit can easily recognize the start and end timing of voice input, the user Is preferably provided on the condition that the touch state of the displayed input item is continued. In addition, from the viewpoint that the user's awareness of voice input can be clearly defined, it is desirable to keep the touch state during voice input. It is assumed that the touch state is kept during input.

  Therefore, when the user inputs the content of the input item in accordance with the voice guidance, what information should be input by touching the input item displayed on the touch panel display 2 with a finger? Can be confirmed visually, and it can be confirmed that the information inputted by voice is accepted by the apparatus, so that the voice interactive apparatus can be used with confidence. In addition, as a change mode of the display format, a change in shape, color, size, and the like can be considered, and there is no particular limitation. However, if it changes according to the volume of the user's voice, the user It is more preferable because it can recognize the level of loudness of one's own voice.

  Further, the display processing means can change the display format of the input item when the voice recognition process is started when the input item displayed on the display unit is designated.

  In other words, when the displayed input item is touched with a finger and a voice is input, the voice interactive device accepts this voice input, and when the voice recognition process is started and is being acquired as information, the input item is displayed. The format is further changed to notify the user that the user's voice input has been accepted. It should be noted that the change in the display format at this time is preferably different from the change when the voice input is accepted.

  By the way, in the processing executed by the display processing means, (1) when the user touches with his / her finger and inputs a voice, when the voice dialogue apparatus accepts this voice input, the display format is changed and the user is voiced. A process for notifying that an input has been accepted; and (2) a process for changing the display format of the input item when the voice recognition process is started when the input item displayed on the display unit is designated. However, either one of them or both may be implemented.

Further, it is preferable that the display processing means displays the input items on the display section when a predetermined period has elapsed after the voice guidance is output by the voice guidance output means. That is, the visitor is not initially distracted by the touch panel display 2, but first, listens to the voice guidance in a concentrated manner, and then shifts his eyes to the touch panel display 2.
[2. Outline of operation of spoken dialogue processing device]

  An outline of the operation of the voice interaction device described above will be described as a case where a customer visiting the company uses the voice interaction device. FIG. 4 is an explanatory diagram showing the flow of the operation of the voice interaction apparatus according to the present embodiment, FIG. 5 is an explanatory diagram showing an outline of the voice input processing, and FIGS. 6 to 9, 11 and 12 are displayed on the touch panel display 2. FIG. 10 is an explanatory diagram showing the procedure of the display format changing process according to the input level of voice input.

As shown in FIG. 4, when the visitor stands in front of the voice interaction device functioning as an automatic reception device, the voice interaction device detects the visit of the customer by the infrared sensor 5 (FIG. 2) (step S101).
The voice interaction apparatus that has detected the presence of a visitor refers to a visitor reception scenario that will be described in detail later, and acquires a predetermined utterance voice (line) and a list of input items (step S102).

  Next, a partition area corresponding to the input item is displayed on the touch panel display 2 (step S103), and voice guidance is uttered (step S104).

  For example, as shown in FIG. 6, on the touch panel display 2, there are a character image in which the receptionist is imaged, and a “company name input section” area and a “name input section” area of the visitor himself / herself who should input voice. In the area displayed in a circle, the characters “company” and “name (surname)” are displayed in a written form. Further, as if the receptionist is speaking, the speaker 4 outputs a voice guidance such as “Come on. Please tell me your name and affiliation first”.

  Then, the voice input process is executed based on the designation of the partitioned area and the voice input by the visitor (step S105), and then it is determined whether or not all the voice contents according to the visitor reception scenario have been recognized (step S106). If completed, the customer acceptance process is terminated as it is, and if not completed, the processes in steps S102 to S105 are repeated.

  Here, the voice input process of step S105 is performed in the procedure as shown in FIG. If you want to enter a company name after listening to the voice guidance “First, tell me your name and affiliation.” touch.

  The voice interactive device detects the touch operation of the visitor (step S201), and loads a voice recognition dictionary corresponding to the input item (S202). A plurality of types of dictionaries are stored in the storage unit in association with the customer reception scenario.

  That is, the information processing apparatus functioning as voice recognition means has a plurality of voice recognition dictionaries, and selects a voice recognition dictionary corresponding to an input item displayed on the touch panel display 2 to the microphone 3. The input voice is recognized.

  For example, if a visitor enters a name first and then a company name, touching the “name entry area” area to enter the name, the control unit will create a name dictionary for the whole country. When this is selected and loaded, and the "company name input section" area is touched, a company dictionary for the whole country is selected and loaded, and each voice recognition is performed.

  On the other hand, information that functions as a display processing means when there are a plurality of input items to be input to the voice input unit in response to voice guidance, such as the above-described “company name input section” area and “name input section” area. The control unit of the processing device displays each of the plurality of input items on the display unit in a predetermined display format, but the control unit responds to the input content of the input item already recognized among the plurality of input items. A speech recognition dictionary is selected, and speech input to the speech input unit is recognized using the selected speech recognition dictionary for an input item that has not been recognized among the plurality of input items. You can also.

  In other words, if the company name is input first and then the name is input next, the control unit will not respond to the input contents of the input item (company name) that has already been recognized, not the name dictionary for the whole country. The employee name dictionary of the company is selected and loaded as the voice recognition dictionary. Therefore, the range of the dictionary search is reduced, and the recognition accuracy and speed of the visitor's name can be improved.

  Next, the voice interactive apparatus changes the display area corresponding to the input item to a display indicating the input waiting state (step S203), and changes the system state to the input waiting state (step S204).

  If the “company name input section” area is designated, as shown in FIG. 7, the circle of the “company name input section” area displayed in a circle is displayed in multiple layers. This makes it possible to distinguish it from the “name input section” area that is displayed as it is.

  On the other hand, the visitor confirms that the “company name input section” area touched with the finger has changed to the input waiting state as shown in FIG. While consciously saying, “I am a voice input,” for example, I can input a voice with confidence, for example, “I am an industry.”

  When the voice interactive apparatus receives voice input and starts recognition processing of the input content, the voice interactive apparatus executes display format change processing (step S206).

  For example, as shown in FIG. 8, the color type and its density are changed to three levels (first color to third color) according to the input level of the voice input. By such display processing, the visitor can confirm that his / her voice is accepted.

  The change of the display process shown in FIG. 8, that is, the process of changing the display format in accordance with the input level of the voice input is performed according to the procedure shown in FIG.

  As shown in FIG. 10, first, the control unit of the voice interaction apparatus acquires a voice input level (step S300). Next, the sound input level allowable range data stored in advance in the storage unit is read (step S301). Then, it is determined whether or not the input level exceeds the allowable range (step S302). If it is determined that the number exceeds, the process proceeds to step S303, and the display area (for example, “company name input section” area) is set to the first color (for example, light blue) (FIG. 8A).

  If it is determined in step S302 that the input level does not exceed the allowable range, it is determined whether or not the input level falls below the allowable range (step S304). If it is determined that the value is lower, the process proceeds to step S305, and the display area (for example, “company name input section” area) is set to the second color (for example, normal density green) (FIG. 8). (b)).

  If it is determined in step S304 that the input level does not fall below the allowable range, the process proceeds to step S306, and the display area (for example, “company name input section” area) is changed to the third color (for example, high-density red). Etc.) (FIG. 8C).

  When the processing of steps S303, S305, and S306 is completed, it is determined whether or not the voice input by the visitor has been completed (step S307). If it is determined that it is not, the process returns to step S300.

  Further, in the display format changing process in step S206, when the voice recognition process is started with respect to “○ △ industrial” uttered by the customer, as shown in FIG. 9, light is emitted radially in the “company name input section” area. The display mode (display format) is changed by blinking. By such display processing, the visitor can confirm that his / her name is being recognized.

  Returning to FIG. 5, when the control unit of the voice interaction device finishes the display format change process (step S206), the visitor has started the voice interaction content recognition process on the voice interaction device side for his / her utterance. Since it can be confirmed by the change in the display format shown in FIG. 9, the finger is released from the touched “company name input section” area, and the utterance is ended.

  The control unit of the voice interaction apparatus performs input speech recognition processing using the dictionary loaded in step S202 (step S208), and displays the recognition results around the “company name input section” area in order of likelihood. (Step S209). For example, as shown in FIG. 11, a plurality of recognition result candidates having different sizes in order of likelihood are displayed in a partition area of a predetermined shape so as to surround a “company name input section displayed in a circle” area. Here, like the company name input section area, a circular section area is used, but it may be a rectangle or the like.

  On the other hand, the visitor confirms the recognition result candidates displayed around the “company name input section” area, and if there is a correct recognition result reflecting his / her voice input, the correct recognition is made on the touch panel display 2. Touch the part where the result is displayed. On the other hand, an NG button is displayed on the touch panel display 2 (not shown), and if it is determined that the recognition result does not reflect its own voice input, the NG button is touched and the voice input is performed again.

  By the way, when the recognition process of the voice input in step S208 is performed, there may be a single candidate instead of a plurality. In that case, only one section area of a predetermined shape indicating a plurality of recognition result candidates is displayed, and this functions as a confirmation button. Note that an OK button (not shown) may be displayed together with the NG button so that these can be used as a confirmation button.

  As described above, the control unit of the voice interactive apparatus functions as voice recognition means, and further displays candidate candidates for displaying a plurality of input content candidates recognized by the voice recognition means on the touch panel display 2, and the candidates. It also functions as a selection means for selecting any one candidate from a plurality of input content candidates displayed by the display means, and an input processing means for determining the candidate selected by this selection means as the input content.

  When the control unit of the voice interaction apparatus detects a touch signal on the touch panel display 2 indicating a correct recognition result from the visitor after the process of step S209, the control unit recognizes it in the company name input section area as shown in FIG. In addition to displaying the result, the circle in the “Company Name Input Section” area is displayed with a bold line to notify the visitor that the voice input has been confirmed.

  Thus, the voice interactive apparatus according to the present embodiment includes the touch panel display 2 as an interface so that the user can visually grasp the contents of the voice guidance in addition to the voice input unit and the voice output unit. As a result, a smoother dialog can be realized.

  Hereinafter, the voice interaction apparatus according to the present embodiment will be described more specifically with reference to the drawings. In the following, an explanation will be given by taking as an example the process of accepting a visitor to the company by means of a voice interaction device and relaying a visit to an employee desired by the visitor.

[3. Specific Configuration of Spoken Dialogue Processing Device]
FIG. 13 is a block diagram showing an electrical configuration of the voice interaction apparatus according to the present embodiment, and FIG. 14 shows a first visitor reception scenario that is an example of the interaction scenario file stored in the storage unit. Visitor reception scenario table, FIG. 15 is a visit reservation table showing the visit reservation data associated with the visitor reception scenario and stored in the storage unit, and FIG. 16 shows the flow of voice dialogue that proceeds according to the visitor reception scenario. FIG. 17 is an explanatory diagram of a second customer reception scenario table, and FIG. 18 is an explanatory diagram of a customer reception screen displayed on the touch panel display 2.

  As shown in FIG. 13, the voice interaction apparatus has a touch panel display 2 (display unit), a microphone 3 (voice input unit), and a speaker 4 (voice output unit) as described above with reference to FIG. 1. A control unit including a CPU 61, a ROM 62, a RAM 63, and the like, and a hard disk device 7 (hereinafter referred to as “HDD 7”) as a storage unit. The information processing apparatus 6 provided with is provided. Although not shown in the figure, the information processing apparatus 6 includes a display control circuit as a display processing unit that controls the touch panel display 2, an audio output circuit that controls audio output from the speaker 4, and the like.

  The HDD 7 stores a system program for controlling the entire voice dialogue apparatus, a voice dialogue program for performing voice dialogue processing, a dialogue scenario file, a voice recognition dictionary, voice data for speech, and the like. , A dialogue control program, a voice input program, an input level determination program, a voice recognition program, and the like. Note that a speech synthesis program can be used instead of the speech data for speech.

  Note that the information processing apparatus that performs main control, such as the HDD 7 that constitutes the storage unit, does not necessarily have to be stored in the housing 1, and may be provided in a separate workstation or server, for example. Good. In that case, a system configuration in which an apparatus including the housing 1 shown in FIG. 2 is used as a terminal apparatus and is connected to a workstation or a server wirelessly or by wire.

  In the dialogue scenario file, a plurality of dialogue scenarios corresponding to various situations are tabulated and stored. In the voice dialogue apparatus according to the present embodiment, the first customer reception scenario table shown in FIG. 14 or shown in FIG. The second visitor reception scenario table is referenced first.

  As shown in FIG. 14, the first visitor reception scenario table includes a plurality of ID columns for acquiring information necessary for determining specific items necessary for dealing with a visitor's business in a time series. Is provided. Each ID column is associated with a title, a speech uttered as voice guidance, and input items to be voice input by a customer who is a user according to the voice guidance. Furthermore, this input item requires an item name, a recognition dictionary (speech recognition dictionary) for recognizing the input item, a subordinate check for determining whether or not it is a subordinate item derived from the input item, and an input item. Is associated with a mandatory check that identifies whether it is an arbitrary or arbitrary one.

  Specifically, ID1 to ID4 are set in the first customer reception scenario table, the title in the column of ID1 is “customer identification 1”, and is output as voice guidance to identify the customer. that dialogue is "Hello. please tell me the company name and your name.". “Company name” and “name (last name)” are set as the item names of input items that the customer as a user inputs by voice. In addition, in this ID1, a flag indicating that neither “company name” nor “name (last name)” is subordinate in the subordinate check is set, and in the essential check, both are essential. A flag is raised.

  As the recognition dictionary, “company name dictionary” is set for “company name”, and “surname last name dictionary” is set for “name (surname)”. That is, the information processing apparatus functioning as voice recognition means has a plurality of voice recognition dictionaries, and appropriately selects a voice recognition dictionary corresponding to an input item displayed on the touch panel display 2 via the microphone 3. The input voice is recognized.

  In addition, the title of the column of ID2 is “department identification”, and the voice guidance line for identifying the department is “Tell me your department name”. “Department name” is set as the item name of the input item. As a recognition dictionary, a “part name dictionary” is set. Since the input item of ID2 is a subordinate item derived from “company name” which is the input item of ID1, a flag indicating subordination is set in the subordinate check, and the input item of ID2 also has a department. Since it is indispensable to specify, a flag indicating that it is indispensable in the essential check is set.

  The title in the column of ID3 is “customer identification 2”, and this is a subordinate item when a visitor cannot be identified by “customer identification 1”, so a flag indicating subordination is set in the subordinate check. The speech that is output as voice guidance is “I'm sorry, can I get the name below?”, And the item name of the input item that the visitor inputs by voice is set to “name (name)” ing. A “name dictionary” is set as the recognition dictionary.

  The title in the column of ID4 is “guidance specific”, which is a conclusion of a dialogue scenario, and no voice input from a visitor is assumed. Therefore, only a line is set in this ID4. There are two types of lines: “I will connect you to the person in charge now” and “No appointments are registered”.

  Further, the visit reservation table shown in FIG. 15 is stored in association with the visitor reception scenario table. As shown in the figure, “visit company name”, “visitor department”, “visitor name”, “visit date / time” For each item of "person in charge" and "person in charge telephone number", data relating to visitors who have made a visit reservation are collected.

[4. Progress of voice dialogue by voice dialogue processing device]
How the voice dialogue proceeds according to the first visitor reception scenario will be described with reference to FIG. When a visitor is detected by the infrared sensor 5, the voice interaction apparatus refers to the first visitor reception scenario and, as shown in the figure, first, a display area (“company name input section”) corresponding to the speech process and input items, as shown in the figure. The display process of the area, “name input section” area) is executed (step S400).

  It is determined whether or not the visitor's company is a group company as a result of the voice input by the visitor in the procedure described above (see FIG. 5) (step S401). This determination is based on the company name dictionary used when recognizing a visitor's voice input. That is, an identifier indicating whether or not a company is a group company is attached to company name data in the company name dictionary.

  If it is determined that the company is a group company, based on ID2 of the customer reception scenario table (FIG. 14), the voice interaction device outputs voice guidance “Please tell me your department” in step S402. At this time, as shown in FIG. 18, the department name (affiliation), which is an input item subordinate to the company name, is connected on the touch panel display 2 with a line so as to relate to the “company name input area” area. Is displayed.

  That is, the information processing apparatus of the voice interactive apparatus can display on the touch panel display 2 an input item subordinate to the input item that is already displayed when the input content for the input item is a predetermined input content. .

  According to this voice guidance “Please tell me your department”, the visitor can also input the department to which he / she belongs by touching the department (affiliation) part of the touch panel display 2 according to the procedure shown in FIG. Then, the information processing apparatus recognizes the voice and generates visitor data, and refers to the visitor reservation data (step S403). Then, it is determined whether the generated visitor data corresponds to the visitor reservation data (step S404). This visitor reservation data is data of the visit reservation table shown in FIG.

  If there is no visitor data specified by voice recognition in the visitor reservation table data, it is determined that there is no reservation, and the process proceeds to step S405. In this process, the information processing apparatus outputs a voice message such as “No appointment is registered.” Then, in step S406, the voice conversation progresses to another conversation scenario, for example, a business confirmation scenario. The process ends.

  After that, the business confirmation scenario is read and the business confirmation process proceeds. In this business confirmation scenario, for example, “No appointment is registered, but what kind of business The voice guidance is output, and the input items (for example, meeting, delivery, sales, etc.) to be input in accordance with the voice guidance are displayed on the touch panel display 2.

  On the other hand, if the visitor data does not correspond to any of the visitor reservation data in step S404, it is determined whether or not there is data having the same surname (same last name) as the visitor data in the visitor reservation data (step S407).

  If there is no data with the same surname, the process proceeds to step S410. On the other hand, if there is data with the same surname, in step S408, "I'm sorry, can you give me a lower name?" Output guidance. At this time, the name (first name) which is an input item subordinate to the name (last name) is displayed on the touch panel display 2 so that the department name (affiliation) shown in FIG. 18 is related to the “company name input section” area. It is displayed in a form similar to that connected by a line.

  When the visitor again inputs the lower name by voice while touching the name (name) portion of the touch panel display 2 according to the procedure shown in FIG. 5, the information processing apparatus recognizes the voice and inputs the visitor data. Regenerate and go to visit reservation data. Then, it is determined whether or not the regenerated customer data corresponds to the aforementioned customer reservation data (step S409).

  If the same name data does not exist, the process proceeds to step S405. If the same name data exists, in step S410, the voice message “I will connect to the person in charge now” is output and the customer reception scenario Finish the voice dialogue process for accepting visitors.

  Note that the voice interaction apparatus according to this embodiment is used as an automatic reception apparatus, and therefore has a configuration including predetermined communication means capable of communicating with employees. For example, if the communication means is a telephone, after the process of step S410, the voice dialogue apparatus refers to the person in charge telephone number of the visit reservation table and connects to the person in charge and informs the visitor by voice, or the voice dialogue apparatus. Through the microphone 3 and the speaker 4, communication control processing can be performed so that the visitor can directly talk with the person in charge (employee). In other words, the customer reception process of the automatic reception apparatus using the voice interaction process includes a process for contacting the person in charge.

  By the way, in the voice dialogue processing referring to the first visitor reception scenario table described above, when the visitor is finally transferred to the person in charge, the visit reservation table (FIG. 15) is referred to and associated with the prospective visitor. I was trying to relay to the person in charge.

  In this case, data indicating the person in charge is required in the visit reservation table, but it is an example of a dialogue scenario file in order to be able to accept a visitor by voice dialogue even if there is no data of the person in charge in the visit reservation table The second visitor reception scenario table may be used.

  The second visitor reception scenario table is different from the first visitor reception scenario table in that the name of the person in charge is input to the visitor by voice.

  That is, as shown in FIG. 17, ID1 to ID3 are set in the second visitor reception scenario table, and the title in the column of ID1 is “customer identification”, which is the same as that shown in FIG. The same lines and input items (item name, recognition dictionary, subordinate check, mandatory check) as “customer identification 1” of ID 1 of the 1 visitor reception scenario table are set.

  In the ID2 column, the same title “department identification” is set as in the ID2 column of the first visitor reception scenario table, and the lines and input items (item name, recognition dictionary, subordinate check, essential check) are the same. .

  The title in the column of ID3 is “specify person in charge”, and the speech that is voice-output as voice guidance to identify the customer is “Please tell me the name of the person in charge”. “Name of person in charge” and “Department name” are set as item names of input items that are input by a customer as a user. In the subordinate check of ID3, a flag is set indicating that neither the “person in charge” nor the “department name” is subordinate. On the other hand, in the essential check, a flag indicating that the “person in charge” is essential is set, but a flag indicating that the “department name” is optional is set.

  In addition, as a recognition dictionary, a “person name dictionary” is set as the “person name”, and a “part name dictionary” is set as the “part name”.

[5. Specific voice dialogue processing by voice dialogue processing device]
Hereinafter, with reference to FIG. 19 to FIG. 25, the customer acceptance process by the voice dialogue process according to the second visitor acceptance scenario table will be described. FIG. 19 is a flowchart showing an example of a voice dialogue process that proceeds according to a customer acceptance scenario, FIGS. 20 and 21 are flowcharts showing a subroutine of the voice dialogue process, and FIGS. 22 to 25 are explanatory diagrams of screens displayed on the touch panel display. It is. The following processing is executed by the automatic reception device to which the voice interactive device having the electrical configuration shown in FIG. 13 is applied. The power is already turned on, the system program is started, and the voice interactive program is executed. This is a processing flow from the time when all the initial settings and the like are read and the visitor is located in front of the device.

  As shown in FIG. 19, when a visitor is detected by the infrared sensor 5 (step S500), the CPU 61 in the apparatus reads the second visitor reception scenario table (FIG. 17) from the dialogue scenario file and stores the visitor identification (ID1). Data is loaded (step S501).

  A dialogue with the visitor is executed according to this visitor identification (ID1) (step S502). By this step S502, the company name and name of the visitor are specified. The dialog execution process executed in step S502 will be described in detail later with reference to the subroutines shown in FIGS.

  When the company name is recognized by specifying the visitor, the CPU 61 determines whether or not this company name is its own group company (step S503). That is, the CPU 61 determines with reference to the company name dictionary defined in the second visitor reception scenario table (see the description of step S401 in FIG. 16).

  If the company name is a group company, the CPU 61 loads the department identification (ID2) data from the second visitor reception scenario table (step S504), and executes a dialog with the visitor according to the department identification (ID2) ( Step S505). The processing in step S505 is similar to that in step S502, and this will also be described in detail later.

  Next, the CPU 61 refers to the visit reservation data (FIG. 15) in the visit reservation table (step S506). Then, it is determined whether or not the visitor data generated by recognizing the voice input of the visitor is in the visit reservation data (step S507).

  If it is determined that there is visitor data in the visit reservation data, the CPU 61 proceeds to step S510. If it is determined that there is no visitor data, the CPU 61 determines the person in charge (ID3) from the second visitor reception scenario table in step S508. ) Is then loaded, and then a dialog with the visitor is executed according to the person-in-charge identification (ID3) (step S509), and then the process proceeds to step S510. Note that the processing in step S509 is similar to that in step S502, and will be described in detail later.

  In step S510, the CPU 61 obtains the person in charge telephone number of the visit reservation table (step S510). Then, the person in charge is contacted via the telephone (step S511), the effect of the visitor is announced by voice, and the visitor reception process is terminated. Alternatively, in step S511, as described above, the communication control process is performed so that the visitor can directly talk with the person in charge (employee) via the microphone 3 and the speaker 4 of the voice interaction device (automatic reception device). The customer acceptance process may be terminated when communication is completed.

  Here, the execution process of the dialog with the visitor in the above step S502, step S505, and step S509 will be described in detail with reference to FIG. 20 and FIG. Since the processes in steps S502, S504, and S509 all have the same process flow, the process in step S502 will be basically described below, and the processes in steps S505 and S509 will be described. Will be described as appropriate.

  As shown in FIG. 20, according to the visitor identification (ID1) in the second visitor reception scenario table, the CPU 61 first executes an input screen display process and a speech reproduction process (step S600).

  The processing in step S600 is executed by a subroutine shown in FIG. That is, first, a list of input items is acquired from the corresponding data in the dialogue scenario table (step S700). If it is the process of step S501 in FIG. 19, the visitor identification (ID1) of the second visitor reception scenario table is specified, and if it is the process of step S504, the department identification (ID2) of the second visitor reception scenario table is specified, step S509. If it is the process, the department identification (ID3) of the second visitor reception scenario table is acquired.

  Next, the CPU 61 determines whether or not there is an item subordinate to another input item in the acquired input item of ID1 (or ID2 or ID3) (step S701). If there are subordinate items, the process proceeds to step S704. If there are no subordinate items, in step S702, the drawing initialization process is performed to delete the input area already displayed, and the above-described ( Outline of voice input processing using FIG. 5) A confirmation button is displayed (step S703), and the process proceeds to step S704.

  The CPU 61 performs the processing shown in steps S704 to S711 for all the input items of the acquired ID1 (or ID2 or ID3) (if ID1, company name and last name), if ID2, only the department name, ID3 If so, input processing to the person in charge and department name) is started.

  That is, first, the input item is checked for necessity (step S705). If it is required, the input area for the input item is displayed in the required size, but if it is not required, the input for the input item is displayed. The area is drawn with an arbitrary size (step S707).

  As shown in FIG. 17, the company name, first name (last name), and the department name in ID2 are all input items in ID1, and as a result, in the dialog execution process according to ID1 in this visitor reception process, step S706 is performed. Processing will be done.

  On the other hand, among the person in charge name and department name in ID3, the department name is arbitrary. Therefore, as shown in FIG. 22, the CPU 61 displays a “department name input section” composed of a small area with respect to the essential “person name input section” area.

  As described above, when the CPU 61 of the division processing apparatus displays an arbitrary input item on the touch panel display 2 in addition to the essential input items to be input to the voice input unit according to the voice guidance, Control to change the display format with any input item. By performing such display control, it becomes possible for a visitor to visually determine which item should be input by voice even when a plurality of input items are displayed on the touch panel display 2. Usability as an interactive device is improved.

  Next, the CPU 61 determines whether or not the input item of ID1 (or ID2 or ID3) is subordinate to another input item (step S708), and specifies the subordinate input item (step S709).

  If it is subordinate to the company name of ID1 as in the department name in ID2, for example, as shown in FIG. 18, it is between the input areas of the subordinate input items ("company name input section" area and A line representing the relation is drawn in the “part name input section” area (step S710). On the other hand, if it is not dependent, this input process is terminated and the process proceeds to step S712 (step S711).

  As shown in FIG. 17, since the company name and name (last name), the person in charge name and the department name, which are input items in ID1 and ID3, are neither subordinates, they follow ID1 in this customer acceptance process. In the dialog execution process, steps S709 and S710 are not performed.

  In step S712, the CPU 61 obtains the line set in the visitor identification (ID1) of the second visitor reception scenario table. In other words, to get the words of "Please tell me hello. Company name and your name.".

  Then, the acquired speech is output from the speaker 4 as voice guidance (step S713), and the CPU 61 moves the process to step S601 in FIG.

  In step S601, a voice input is made by a visitor (visitor). This is in accordance with the process flow described in the outline of the voice input process (see FIG. 5).

  That is, when a predetermined input item area on the touch panel display 2 is touched by a visitor, a dictionary for speech recognition corresponding to the input item is loaded, and the display area corresponding to the input item is set in an input waiting state. Is displayed (see FIG. 7), and the system state transitions to the input waiting state. Then, the CPU 61 accepts the voice input in that state (see FIG. 8), and further executes the display format changing process when the input content recognition process is started (see FIG. 9). Then, the input speech is recognized using the previously loaded dictionary, and if there are a plurality of recognition results, they are displayed in order of likelihood (see FIG. 11).

  When the visitor confirms the recognition result displayed on the touch panel display 2 and presses an OK button, for example (step S602), the CPU 61 determines whether or not all essential input items in ID1 (or ID2 or ID3) have been input. If it is determined (step S603) and it is determined that there is an input item that has not been input, for example, an utterance that prompts the user to input an essential item such as “No item of ○○ has been input” is output (step 603). In step S604, the process proceeds to step S601, on the other hand, if it is determined in step S603 that all the essential input items have been input, the CPU 61 ends the processes in step S502, step S505, and step S509, and then proceeds to step S503 or step S509. The process proceeds to S506 or step S510.

  Note that the speech data relating to the utterance in step S604 is stored as utterance voice data in the HDD 7 in association with each ID of each dialogue scenario.

  As described above, if you use a voice interaction device to accept customers, what kind of content should be input and how even if the user is not used to inputting information by voice? Since it becomes easy to grasp, anyone can use it with peace of mind.

  By the way, in the embodiment described above, during the processing of step S600 and step S601 shown in FIG. 20, specifically, after outputting voice guidance in step S713 of FIG. 21, it corresponds to the item input by the visitor. When a predetermined time elapses before the utterance is started (see FIG. 5) while the display area to be touched on the touch panel display 2, the information processing apparatus functioning as a display processing unit is displayed on the touch panel display 2. In addition, an input example of an item to be input can be displayed.

  That is, when a predetermined period elapses after the input item displayed on the touch panel display 2 is specified, a voice interactive device that can display an input example for the input item can be provided.

  As a display form of an input example, for example, in the case of inputting a company name, as shown in FIG. 23, “△ industrial”, “XX sales”, “Δ × trading” is displayed next to the “company name input section” area. Examples of input are listed in. Therefore, when the user (visitor) is worried about whether he / she has to speak to a corporation or limited company when inputting the company name by voice, the user touches the area that he / she touches. Since the input examples are displayed close to each other, it is possible to prevent the time required for voice input from being unnecessarily long.

  If there are items subordinate to the input items as in the process of step S402 in FIG. 16, as shown in FIG. 18, the input partition areas are connected with lines so as to be related to each other. Although displayed, for example, when there are three or more input section areas on the touch panel display 2, as shown in FIG. 24, the input section area indicating the subordinate items (the “department input section” area in FIG. 24) However, it can be displayed as if it were an animation.

  In addition, as the voice conversation proceeds, the items to be input by the user (visitor) may be “Yes” or “No” as shown in FIG.

  In such a case, the user (visitor) does not perform voice input, and the voice interaction apparatus may accept input only by touching the input section area indicating the input item. That is, input of simple contents or important contents is completed only by touch input by the touch panel display 2 without uttering.

  If it is a user, even if it is necessary to input contents that the other person does not want to hear, it is possible to use the voice interactive apparatus with peace of mind.

  In addition, it is assumed that a visitor suddenly visits without making an appointment. In such a case, according to the first visitor reception scenario and the second visitor reception scenario described above, in the first visitor reception scenario, as described with reference to FIG. Is output (see step S405), and it will be carried over to the business confirmation scenario, etc., and even if it follows the second visitor reception scenario (see FIGS. 19 to 21), the visitor May take time because his name and affiliation must be input first.

  Therefore, for example, an appointmentless reception scenario is prepared in advance, and a “no appointment” button area may be displayed on the touch panel display 2. When the “no appointment” button area is touched, the no appointment reception scenario is read, and a speech such as “Please tell me the employee you want to contact.” Is output from the speaker 4 as voice guidance. It is. At that time, as a screen displayed on the touch panel display 2, the form shown in FIG. 25 can be considered.

  FIG. 26 is an explanatory diagram of a screen displayed on the touch panel display 2 in this case. As shown in the figure, the department name and the employee name are displayed as required items in the input partition area of a predetermined size. Arbitrary input items such as group names are smaller than the input area. That is, also in this case, when displaying an arbitrary input item on the touch panel display 2 in addition to the essential input item to be input to the voice input unit according to the voice guidance, the required input item and the optional input item Follow the control to change the display format.

  In the no appointment reception scenario, although not shown in the table, the department name, first name (surname), and group are set as the item name of the input item. Needless to say, there is a flag indicating that it is optional.

  By the way, in embodiment mentioned above, although the voice interactive apparatus detected the user (visitor) with the infrared sensor 5, what is the apparatus and means as long as it can detect a user? May be used. It may be a switch that outputs a visit signal to the control unit by operating the visitor.

From the above-described embodiments, the following voice dialogue apparatus can be realized.
(1) A dialogue scenario in which voice guidance and input items (for example, item names shown in FIGS. 14 and 17) to be input to a voice input unit (for example, the microphone 3) according to the voice guidance are associated with each other is stored. Scenario storage means (for example, HDD 7), voice guidance output means (for example, CPU 61 or voice output circuit of information processing device 6) for outputting voice guidance according to the dialogue scenario from a voice output unit (for example, speaker 4), When the voice guidance is output from the voice output unit, input items to be input to the voice input unit according to the voice guidance are displayed on the display unit (for example, the touch panel display 2) according to the predetermined scenario (for example, touch panel display 2). 6 to 9, FIG. 11, 12, FIG. 18, FIG. 22 to FIG. 26) display processing means (for example, information A speech recognition means (for example, the CPU 61 of the information processing device 6, a voice dialogue program, a dialogue scenario file) that recognizes the input content based on the voice inputted to the voice input unit. Voice recognition dictionary and speech data for speech).

  (2) In the above (1), the display processing means (for example, the CPU 61 or the display control circuit of the information processing device 6) displays the input items (for example, FIG. 14 and FIG. 14) displayed on the display unit (for example, the touch panel display 2). When the voice input to the voice input unit (for example, the microphone 3) is performed when the item name shown in FIG. 17 is specified, the display format of the input item is changed (for example, the processing ( See FIG. 7 and FIG.

  (3) In the above (1), the display processing means (for example, the CPU 61 or the display control circuit of the information processing apparatus 6) displays the input items (for example, FIG. 14 and FIG. 14) displayed on the display unit (for example, the touch panel display 2). When the voice recognition process is started when the item name shown in FIG. 17 is designated, the voice interactive device changes the display format of the input item (see, for example, FIG. 9).

  (4) In any one of the above (1) to (3), the display processing means (for example, the CPU 61 or the display control circuit of the information processing apparatus 6) is connected to the voice guidance output means (for example, the CPU 61 of the information processing apparatus 6). Or a voice output circuit), a voice interaction device that displays the input items on the display unit (for example, the touch panel display 2) when a predetermined period has elapsed after the voice guidance is output.

  (5) In any one of the above (1) to (4), the voice recognition means (for example, the CPU 61 of the information processing device 6, a voice dialogue program, a dialogue scenario file, a voice recognition dictionary, speech voice data, etc.) is recognized. Candidate display means for displaying a plurality of input content candidates displayed on the display unit (for example, the CPU 61 or the display control circuit of the information processing device 6), and a plurality of input content candidates displayed by the candidate display means, Selection means for selecting any one candidate (for example, a confirmation button such as an OK button) and input processing means for determining the candidate selected by the selection means as input contents (for example, the CPU 61 of the information processing apparatus 6) Spoken dialogue device provided.

  (6) In any one of the above (1) to (5), the display processing means (for example, the CPU 61 or the display control circuit of the information processing device 6) is configured to perform the voice input unit (for example, the microphone 3) according to the voice guidance. ) In addition to the required input items to be input (for example, the person in charge name shown in FIG. 22), when displaying an arbitrary input item (for example, the department name shown in FIG. 22) on the display unit, A voice interactive apparatus that changes a display format between an input item and the arbitrary input item.

  (7) In any one of the above (1) to (6), the display processing means (for example, the CPU 61 or the display control circuit of the information processing device 6) is an input item displayed on the display unit (for example, the touch panel display 2). When a predetermined period has elapsed since the designation of the item (for example, the item name shown in FIG. 14 or FIG. 17), an input example for the input item (for example, XX industry, XX sales, or △ X trading shown in FIG. 23) A voice interaction device to display.

  (8) In any one of the above (1) to (7), the display processing means (for example, the CPU 61 or the display control circuit of the information processing apparatus 6) is configured to input the input items (for example, the items shown in FIGS. 14 and 17). When the input content for the name is a predetermined input content, a voice dialogue for displaying on the display unit (for example, touch panel display 2) an input item (for example, a department in FIG. 18) subordinate to the input item that has already been displayed. apparatus.

  (9) In any one of the above (1) to (8), the voice recognition means (for example, the CPU 61 of the information processing device 6, a voice dialogue program, a dialogue scenario file, a voice recognition dictionary, and voice data for speech) It has a plurality of voice recognition dictionaries (for example, name dictionary, company dictionary, employee name dictionary, etc.), and input items (for example, in FIG. 14 and FIG. 17) displayed on the display unit (for example, touch panel display 2). A voice dialogue apparatus that selects a voice recognition dictionary corresponding to the item name and recognizes a voice input to the voice input unit (for example, the microphone 3).

  (10) In any one of the above (1) to (9), the display processing unit (for example, the CPU 61 or the display control circuit of the information processing device 6) is configured to perform the voice input unit (for example, the microphone 3) according to the voice guidance. ), When there are a plurality of input items (for example, item names shown in FIG. 14 and FIG. 17), each of the plurality of input items is displayed on the display unit (for example, touch panel display 2) in a predetermined display format. The voice recognition means (for example, the CPU 61 of the information processing device 6, the voice dialogue program, the dialogue scenario file, the voice recognition dictionary, and the voice data for speech) are a plurality of voice recognition dictionaries (eg, name dictionary, company dictionary, employee). A name dictionary, etc.), and selects a speech recognition dictionary corresponding to the input content of the input item already recognized among the plurality of input items. Yet the recognition of speech input to the speech input unit for the input item not recognized, voice dialogue system carried out using the selected speech recognition dictionary of the input items.

  In addition, from the embodiment described above, it is possible to realize a program that causes a computer to function as each unit of the voice interaction device according to any one of (1) to (10).

  Furthermore, from the above-described embodiment, the procedure S1 for outputting voice guidance according to the conversation scenario stored in the storage unit (for example, HDD 7) from the voice output unit (for example, the speaker 4), and the voice guidance is the voice output unit ( For example, when output from the speaker 4), the input item (for example, the item name shown in FIGS. 14 and 17) to be input to the voice input unit (for example, the microphone 3) according to the voice guidance is recognized. A procedure S2 for displaying in a predetermined display format on a display unit (for example, the touch panel display 2) according to the input content and the dialogue scenario, and a procedure S3 for recognizing the input content based on the voice input to the voice input unit. The voice interaction processing method possessed can be realized.

  As mentioned above, although this invention was demonstrated through embodiment applied to the automatic reception apparatus mainly installed in a company etc., this invention is not limited to embodiment mentioned above, It communicates with a user by an audio | voice. Therefore, it can be widely used as a voice interactive apparatus that provides information and services required by users.

It is the schematic of the voice dialogue processing method concerning this embodiment. It is a block diagram of the visitor reception apparatus in this embodiment. It is a figure which shows an example of the visit handling database memorize | stored in the visitor reception apparatus in this embodiment. It is explanatory drawing which shows the flow of operation | movement of the voice interactive apparatus which concerns on this embodiment. It is explanatory drawing which shows the outline | summary of an audio | voice input process. It is explanatory drawing of the screen displayed on a touchscreen display. It is explanatory drawing of the screen displayed on a touchscreen display. It is explanatory drawing of the screen displayed on a touchscreen display. It is explanatory drawing of the screen displayed on a touchscreen display. It is explanatory drawing which shows the procedure of the display format change process according to the input level of audio | voice input. It is explanatory drawing of the screen displayed on a touchscreen display. It is explanatory drawing of the screen displayed on a touchscreen display. It is a block diagram which shows the electric constitution of the voice interactive apparatus which concerns on this embodiment. It is a visitor reception scenario table which tabulated and showed the 1st visitor reception scenario which is an example of the dialogue scenario file memorize | stored in the memory | storage part. It is the visit reservation table which tabulated and showed the visit reservation data linked | related with the visitor reception scenario and memorize | stored in the memory | storage part. It is explanatory drawing which shows the flow of the voice dialogue which progresses according to a visitor reception scenario. It is explanatory drawing of a 2nd visitor reception scenario table. It is explanatory drawing of the screen displayed on a touchscreen display. It is a flowchart which shows an example of the voice dialogue process which progresses according to a visitor reception scenario. It is a flowchart which shows the subroutine of the same voice interaction process. It is a flowchart which shows the subroutine of the same voice interaction process. It is explanatory drawing of the screen displayed on a touchscreen display. It is explanatory drawing of the screen displayed on a touchscreen display. It is explanatory drawing of the screen displayed on a touchscreen display. It is explanatory drawing of the screen displayed on a touchscreen display. It is explanatory drawing of the screen displayed on a touchscreen display in other embodiment.

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 Case 2 Touch panel display 3 Microphone 4 Speaker 5 Infrared sensor 6 Information processing device 7 Hard disk device 10 Reception counter

Claims (12)

Scenario storage means for storing a dialogue scenario in which voice guidance and input items to be input to the voice input unit according to the voice guidance are associated with each other;
Voice guidance output means for outputting voice guidance according to the dialogue scenario from a voice output unit;
Display processing means for displaying input items to be input to the voice input unit according to the voice guidance in a predetermined display format according to the dialogue scenario when the voice guidance is output from the voice output unit;
Voice recognition means for recognizing input content based on voice input to the voice input unit;
A spoken dialogue apparatus characterized by comprising: Candidate display means for displaying a plurality of input content candidates recognized by the voice recognition means on the display unit;
Selection means for selecting any one candidate from a plurality of input content candidates displayed by the candidate display means;
The voice interactive apparatus according to claim 1, further comprising: an input processing unit that determines a candidate selected by the selection unit as input content.   The spoken dialogue apparatus according to claim 1, wherein the display processing unit displays an input example for the input item when a predetermined period has elapsed after the input item displayed on the display unit is specified. . The display processing means includes
The display format of the input item is changed when voice is input to the voice input unit when the input item displayed on the display unit is specified. The spoken dialogue apparatus according to any one of claims. The display processing means includes
The display format of the input item is changed when the voice recognition process is started when the input item displayed on the display unit is designated. The spoken dialogue apparatus described.   6. The display unit according to claim 1, wherein the display processing unit displays the input item on the display unit when a predetermined period has elapsed after the voice guidance is output by the voice guidance output unit. The spoken dialogue apparatus according to any one of claims.   In addition to the essential input items to be input to the voice input unit in response to the voice guidance, the display processing means is configured to display the optional input item and the optional input item when displaying any input item on the display unit. The spoken dialogue apparatus according to claim 1, wherein a display format is changed depending on an input item.   The display processing means displays an input item subordinate to an input item already displayed on the display unit when the input content for the input item is a predetermined input content. 8. The voice interactive apparatus according to any one of 7 above.   The voice recognition means has a plurality of voice recognition dictionaries, selects a voice recognition dictionary corresponding to an input item displayed on the display unit, and recognizes a voice input to the voice input unit. The spoken dialogue apparatus according to claim 1, wherein the voice dialogue apparatus is performed. When there are a plurality of input items to be input to the voice input unit according to the voice guidance, the display processing unit displays the plurality of input items on the display unit in a predetermined display format,
The voice recognition means has a plurality of voice recognition dictionaries, selects a voice recognition dictionary corresponding to the input content of the input item already recognized among the plurality of input items, and stores the plurality of input items. 10. The speech input to the speech input unit is recognized for input items that have not yet been recognized, using the selected speech recognition dictionary. The voice interactive apparatus described in 1.   A program for causing a computer to function as each means of the voice interactive apparatus according to any one of claims 1 to 10. Outputting voice guidance from the voice output unit according to the dialogue scenario stored in the storage unit;
When the voice guidance is output from the voice output unit, displaying input items to be input to the voice input unit according to the voice guidance in a predetermined display format on the display unit according to the dialogue scenario;
Recognizing input content based on voice input to the voice input unit;
A voice dialogue processing method comprising:

Images