JP2020042473A - Voice recognition terminal, voice reservation system and voice reservation program - Google Patents

Abstract

To provide a voice recognition terminal, a voice reservation system, and a voice reservation program capable of easily performing a reservation operation by voice input.SOLUTION: A voice recognition terminal according to an embodiment has a speaker, a microphone, and a processor. The processor outputs a search result of reservation candidates according to an essential search condition as voice via the speaker when information indicating the essential search condition set as an essential input item required for searching for the reservation candidates is detected from the recognition result of voice input via the microphone; outputs a re-search result of the reservation candidates to which an additional search condition is added as the voice via the speaker when the additional search condition is detected from the recognition result of the voice input via the microphone after outputting the search result of the reservation candidates; and makes a reservation based on a designated reservation candidate when an instruction for making the reservation with any one of the reservation candidates output via the speaker is detected from the recognition result of the voice input via the microphone.SELECTED DRAWING: Figure 1

Description

  Embodiments of the present invention relate to a voice recognition terminal, a voice reservation system, and a voice reservation program.

  2. Description of the Related Art In recent years, a speech recognition technology has been developed, and a system for operating a terminal used by each home or individual by speech recognition has been widely used. With the spread of such voice recognition technology, many people are accepting voice (voice) operation as one easy-to-use user interface. For example, there is a search system that can search for information such as a route search and a train timetable by voice.

  However, most of the conventional systems capable of performing voice operations simply replace information input to an existing system by voice input. In a system for inputting information by voice, a voice including necessary information is obtained from a user in a conversational manner. Therefore, as the number of input items increases, the process of inputting information takes longer. For example, the conventional reservation system for reserving seats is based on the premise that many input items are input, and it takes time to input all of these input items by voice. Therefore, the conventional reservation system has a problem that even if the conventional reservation system is compatible with voice input, it takes time to input information by voice, and the reservation process cannot be easily performed.

JP 2001-097217 A

  In order to solve the above-mentioned problems, an object of the present invention is to provide a voice recognition terminal, a voice reservation system, and a voice reservation program that can easily perform a reservation operation by voice input.

  According to the embodiment, the speech recognition terminal has a speaker, a microphone, and a processor. The speaker outputs sound. The microphone inputs voice. The processor controls the sound output from the speaker and recognizes the sound input from the microphone. The processor is configured to search for a reservation candidate based on the essential search condition when detecting information indicating an essential search condition set as an essential input item necessary for searching for a reservation candidate from a recognition result of a voice input by the microphone. Is output by voice through the speaker, and after the search result of the reservation candidate is output, if an additional search condition is detected from the recognition result of the voice input by the microphone, the reservation candidate to which the additional search condition is added is reproduced. A search result is output as voice by the speaker, and a reservation based on the specified reservation candidate when an instruction to make a reservation with any one of the reservation candidates output from the speaker is detected from a recognition result of the voice input by the microphone. I do.

FIG. 1 is a block diagram illustrating a configuration example of a voice recognition terminal and a voice reservation system according to the embodiment. FIG. 2 is a flowchart for explaining an operation example of the voice recognition terminal in the voice reservation system according to the embodiment. FIG. 3 is a flowchart for explaining an operation example of the voice recognition terminal in the voice reservation system according to the embodiment. FIG. 4 is a diagram illustrating an example of search conditions registered in advance for each user in the voice reservation system according to the embodiment. FIG. 5 is a diagram illustrating an example of essential input items necessary for a simple search and supplementary input items that can be specified in a detailed search in the voice reservation system according to the embodiment.

Hereinafter, embodiments will be described with reference to the drawings.
The voice reservation system according to the present embodiment is a system for reserving a train seat or the like by voice input by a user. Hereinafter, in the present embodiment, a system will be described in which a user reserves a seat of a train such as an express train or an express train using a voice recognition terminal called a smart speaker by a user's own operation.

FIG. 1 is a block diagram illustrating a configuration example of a voice reservation system according to the embodiment.
In the configuration example shown in FIG. 1, the voice reservation system includes a voice recognition terminal 1, a voice server 2, a voice database (DB) 3, a reservation server 4, a reservation DB 5, a train information DB 6, and the like.

  For example, the voice recognition terminal 1, the voice server 2, and the voice DB 3 constitute a voice recognition system. The speech recognition terminal 1, the reservation server 4 and the reservation DB 5 constitute a reservation system. Further, the train information DB 6 may be connected to an operation management server (not shown) or the like to configure an operation management system. Some or all of these systems are interconnected by, for example, a WAN (Wide Area Network), a VPN (Virtual Private Network), a LAN (Local Area Network), a dedicated line, or the like.

  The voice recognition terminal 1 is a device that inputs and outputs voice in a conversation format with a user. The voice recognition terminal 1 acquires a voice uttered by the user as an input voice, and converts the input information into a text by recognizing the obtained input voice. The voice recognition terminal 1 determines the content of a response to the user based on the information obtained by converting the input voice to text, and outputs the content of the response in voice. In the present embodiment, the speech recognition terminal 1 acquires information necessary for reserving a seat of an express train or an express train by voice from a user in a conversation format.

  The voice recognition terminal 1 is, for example, a voice input / output device called a smart speaker. However, the voice recognition terminal 1 may be any information processing device capable of inputting and outputting voice, and may be a mobile terminal such as a smartphone or a tablet PC, or a stationary computer. Further, the voice recognition terminal 1 may be provided in an electric appliance such as a television, a refrigerator, an air conditioner, and a lighting device.

  The voice server 2 is a computer that manages voice control such as voice input / output in a conversation format. The voice server 2 has a function of communicating with the voice recognition terminal 1. The voice server 2 is also connected to a voice DB 3 that stores voice information. The voice DB 3 stores information indicating the contents of a response to the voice input by the user. The voice server 2 refers to the voice DB 3 to determine a response content to the input voice. That is, the voice server 2 acquires the recognition result of the input voice from the voice recognition terminal 1, and determines the content of the voice response to the input voice with reference to the voice DB 3.

  As a configuration of the voice reservation system, the function of the voice server 2 described above may be provided in the voice recognition terminal 1 or the voice recognition terminal 1 holds information stored in the voice DB 3 as described above. You may do it. In such a configuration, the voice server 2 may be omitted, or the voice DB 3 may be provided in a memory in the voice recognition terminal 1.

  The reservation server 4 is a computer that manages reservation information. The reservation server 4 has a function of communicating with the voice recognition terminal 1. The reservation server 4 is connected to a reservation DB 5 and a train information DB 6. The reservation DB 5 is a database that stores reservation information. The reservation server 4 refers to the reservation information stored in the reservation DB 5 and writes the reservation information. The train information DB 6 is a database that stores information on trains to be reserved. The train information DB 6 may be a database managed by another operation management system or the like. In this case, the reservation server 4 may access the train information DB 6 via the server of the operation management system.

Next, a configuration of a control system in the voice recognition terminal 1 will be described.
In the configuration example shown in FIG. 1, the voice recognition terminal 1 includes a processor 11, a memory 12, a data memory 13, a communication unit 14, a microphone 15, a speaker 16, and the like.
The processor 11 is, for example, a CPU. The processor 11 implements various processes by executing a program. The memory 12 includes a volatile memory such as a RAM, a nonvolatile memory such as a ROM, a rewritable nonvolatile memory, and the like. The processor 11 implements various processes by executing a program stored in the memory 12 or the data memory 13.

  The data memory 13 is a rewritable nonvolatile storage device. The data memory 13 executes an application program executed by the processor. The data memory 13 also stores setting information and the like. For example, the data memory 13 has a storage area for storing a program for voice recognition, a program for reservation processing, and setting information. The speech recognition program is a program for the processor 11 to execute a speech recognition process for converting input speech into text data. The reservation processing program is a program for the processor 11 to make a reservation in cooperation with the reservation server 4.

  In the speech recognition realized by the processor 11 executing the speech recognition program, a plurality of input methods may be allowed. That is, the processor 11 may convert a keyword recognized from the input voice into information that can be used as a search condition. For example, "Monday" as a search condition is not only when the date is specified explicitly, but also when an event that can specify the date such as "Tomorrow", "Day after tomorrow", "Omisomika", "New Year's Day", "Next Monday", etc. For example, it is also possible to recognize the one designated by the above as a specific “month and day”.

  Such a method of specifying information as a search condition from a keyword can be realized by registering a correspondence between a keyword and information as a search condition in a database for speech recognition. Such a database for voice recognition may be arbitrarily added or deleted by a business operator or an administrator according to the operation situation.

  The communication unit 14 is an interface for communicating with the voice server 2 and the reservation server 4. The processor 11 is communicatively connected to the voice server 2 via the communication unit 14 and acquires information indicating a response content output by voice. In addition, the processor 11 requests the reservation server 4 connected for communication via the communication unit 14 to search for a reservation candidate, and also acquires information from the reservation server 4.

The microphone 15 inputs a voice. The microphone 15 is configured to collect a voice uttered by the user. The speaker 16 outputs sound. The processor 11 uses the microphone 15 and the speaker 16 to execute input / output of voice in a conversational manner.
Note that the voice recognition terminal 1 may include a display device such as a display having a touch panel, an operation device, and the like. Thus, the voice recognition device 1 may be configured so that the display device can input information (screen input) using the operation unit in accordance with the display content on the screen, in combination with the information input by voice using the microphone 15. In addition, the voice recognition device 1 may be configured to be able to provide guidance on a screen displayed by the display device, in combination with voice guidance using the speaker 16.

Next, a configuration of a control system in the reservation server 4 will be described.
In the configuration example shown in FIG. 1, the reservation server 4 includes a processor 41, a memory 42, a communication unit 43, and the like.
The processor 41 is, for example, a CPU. The processor 41 implements various processes by executing a program. The memory 42 includes a volatile memory such as a RAM, a nonvolatile memory such as a ROM, a rewritable nonvolatile memory, and the like. The processor 41 implements various processes by executing a program stored in the memory 42.

  In the present embodiment, the processor 41 has a function of searching for a reservation candidate (reservation search) according to a search condition supplied together with a search request from the voice recognition terminal 1. For example, the processor 41 searches for a reservation candidate according to a specified search condition by referring to reservation information stored in the reservation DB 5 and train information of each train stored in the train information DB 6.

  Further, the processor 41 determines the reservation information according to the reservation contents (reservation candidates) supplied together with the reservation request from the speech recognition terminal 1. For example, the processor 41 determines the reservation contents (selection of reservation candidates) by registering the reservation contents specified by the voice recognition terminal 1 in the reservation DB 5.

  The communication unit 43 is an interface for communicating with the voice recognition terminal 1. The processor 41 is communicatively connected to the voice recognition terminal 1 via the communication unit 43 and acquires a reservation candidate search request or a reservation request from the voice recognition terminal 1. The processor 41 also supplies the speech recognition terminal 1 with the result of the reservation search according to the search request or the registration result of the reservation content according to the reservation request as the reservation candidate.

Next, the operation of the voice reservation system according to the embodiment will be described.
FIG. 2 and FIG. 3 are flowcharts for explaining an operation example of the voice recognition terminal 1 in the voice reservation system according to the embodiment.
When starting the reservation processing, the processor 11 of the voice recognition terminal 1 activates a reservation processing program stored in the data memory 13 (ST10). For example, the processor 11 may always execute voice recognition and activate a program for reservation processing when a specific keyword or the like is recognized. For example, when a keyword such as “reservation application”, “express reservation”, or “seat reservation” is detected from the input voice, the processor 11 may start a program for reservation processing.

  When the program for the reservation process is started, the processor 11 performs a login process for authenticating the user (ST11). The login process may be biometric authentication based on the biological characteristics of the voice to be input, or may be authentication by a password (ID, secret question, etc.) to be input by voice.

  If the user is successfully authenticated by the log-in process, the processor 11 outputs a voice input guide for prompting the user to input a method of searching for a reservation candidate through the speaker 16 (ST12). For example, the voice input guidance prompting the user to enter a search method may be "How do you search for a train to reserve?", "Please tell us how to find a train to reserve." Do you want to search? Do you specify search conditions? " Further, the processor 11 may use not only voice input but also input such as a search method and a search condition on a screen using a display device having a touch panel.

  After outputting the voice input guidance, the processor 11 recognizes the voice input by the microphone 15 to determine a search method instructed by the user by voice (ST13). In the present embodiment, as a search method for searching for a reservation candidate, there are a method of searching with registered search conditions (usual search) and a method of directly specifying search conditions by a user (condition specification). . The registered search condition is a search condition stored as setting information for each user. The search condition specified by the user may be registered, or the search condition when a reservation was made in the past may be used. It may be registered.

  If it is determined that the usual search has been instructed (ACT 14, YES), the processor 11 determines whether the registration of the search condition has been instructed as the usual search (ACT 15). For example, when “always search” and “registration” are detected from the input voice, the processor 11 determines that registration of search conditions to be used as a normal search has been instructed.

  If it is determined that the registration of the search condition has been instructed (ST15, YES), processor 11 outputs a voice guidance for inputting the search condition to be registered by voice through speaker 16 (ST16). For example, the processor 11 outputs a guidance to the effect that "Please tell us the search conditions used in usual search."

After guiding the voice input of the search condition to be registered, the processor 11 recognizes the voice input by the microphone 15 and detects the search condition for the user to specify the registration by voice (ST17). When the search condition for which the user specifies registration is determined (ST18), the processor 11 stores the search condition for which registration is specified in the data memory 13 as setting information of the user (ST19). The user setting information may be stored in an accessible storage device, and may be stored in an external storage device such as a reservation DB 5 accessible via the reservation server 4, for example.
If the search condition has been registered, or if the registration of the search condition has been canceled (ST18, NO), the processor 11 returns to ST12 and executes the above-described processing again.

  When a scheduled search using registered search conditions is instructed as a usual search (ST15, NO), the processor 11 inputs a voice to specify a search condition to be used for the scheduled search from the registered search conditions. Is output by voice through the speaker 16 (ST20). For example, the processor 11 outputs, through the speaker 16, a voice guidance for inquiring the user which search condition among the registered search conditions to perform the reserved search.

  After guiding the selection of the search condition, the processor 11 recognizes the voice input by the microphone 15 to detect the registered search condition selected by the user by voice (ST21). For example, the processor 11 identifies the registered search condition corresponding to the detected identification word by detecting the identification word set in each registered search condition from the input voice.

FIG. 4 is a diagram illustrating an example of registered search conditions used as usual search.
Information indicating search conditions as shown in FIG. 4 is information stored for each user, and is stored, for example, in the data memory 13. As shown in FIG. 4, a plurality of search conditions may be stored for each user. Each search condition is stored in association with the identification word. The identification word is a word that can be recognized from the voice uttered by the user, and may be any word that can uniquely specify the registered search condition. For example, the identification word may be a word arbitrarily specified by the user. The search condition that can be registered may be any item. In the search condition to be registered, additional information can be set as information specified when executing the reserved search. For example, “month / day” may be set as additional information, and the user may specify “month / day” as additional information for each reservation search.

  However, in the present embodiment, it is assumed that essential input items described later are the minimum information required for searching for a reservation candidate (reservation search). Therefore, it is assumed that additional information is set in the search condition to be registered. For example, when “month / day” is not included in the search condition to be registered, “month / day” is set as additional information in the search condition. Further, the additional information may be arbitrarily set by the user. For example, if a condition (item) to be changed is set as additional information for each reservation search, the search conditions can be set by changing the additional information for each reservation search.

  When one search condition is specified from the registered search conditions (ST22, YES), the processor 21 determines whether additional information is necessary for the search condition (ST23). If the additional information is unnecessary (ST23, NO), the processor 11 proceeds to ST34 to be described later and executes a reservation search.

  When it is determined that additional information is necessary (ST23, YES), the processor 11 outputs a voice guidance for inputting the additional information by voice through the speaker 16 (ST24). For example, the processor 11 may specifically output voice of the additional information, and prompt the input of the additional information. After outputting the guidance, the processor 11 recognizes the voice input by the microphone 15 so that the user specifies the additional information by voice (ST25). For example, when “month / day” is set as the additional information, the processor 11 specifies “month / day” from the input voice. When the necessary additional information is specified (ST23, NO), the processor 11 proceeds to ST34.

  If there is only one search condition registered as a normal search, the process of ST20-22 may be omitted. Further, as the operation of the reservation system, search conditions that can be registered as usual searches may be limited to one. If only one search condition is registered as the usual search, the user can set the registered search condition simply by specifying the usual search, and can easily and quickly perform the scheduled search. .

  If it is determined that direct input of the search condition has been instructed (ACT 14, NO), the processor 11 outputs a voice guidance for inputting the search condition by the speaker 16 (ST31). After guiding the selection of the search condition, the processor 11 detects each item of the search condition specified by the user by voice by recognizing the voice input by the microphone 15 (ST32). The processor 11 detects each word specifying each item (search item) serving as a search condition from the input voice, and stores the detected item in the memory 12.

  When each item of the search condition is detected from the input voice, the processor 11 determines whether or not a predetermined required input item capable of performing the simple search of the reservation search has been obtained (ST33). In the present embodiment, essential input items as search conditions necessary for executing a simple search (minimum search conditions that allow a scheduled search) and supplementary input items as auxiliary search conditions that can be specified in advance are set in advance. It has been set.

FIG. 5 is a diagram showing a setting example of essential input items and supplementary input items designated as search conditions for a scheduled search.
The required input items are search items that are required at least for the reserved search. The supplementary input items are search items that can be specified as search conditions in addition to the required input items. In the present embodiment, a scheduled search using a search item as an essential input item is referred to as a simple search, and a scheduled search using search conditions in which a supplementary input item is added to the essential input item is referred to as a detailed search.

  In the example illustrated in FIG. 5, “month / date”, “departure time or arrival time”, “departure station”, “arrival station”, and “number of people” are set as the required input items. As the supplementary input items, “seat type”, “window side or aisle side”, “car number”, “seat number”, “vehicle type”, and the like are set. The supplementary input item may be any item as long as it does not contradict other search items, and any information regarding reservation may be set.

If it is determined that the essential input item has not been acquired (ST33, NO), the processor 11 returns to ST31 and accepts an input of a further search item.
If it is determined that the required input items have been obtained (ST33, YES), the processor 11 requests the reservation server 4 to perform a reservation search using the input search items (ST34). After requesting the reservation server 4 to make a reservation search, the processor 11 acquires a reservation candidate from the reservation server 4 as a result of the reservation search (ST35). Here, the processor 41 of the reservation server 4 searches the reservation DB 5 based on the specified search information in response to the reservation search request received from the voice recognition terminal 1 by the communication unit 43, thereby making a reservation possible candidate (reservation). Candidate). When a reservation candidate is selected, the processor 41 of the reservation server 4 supplies a reservation search result including the reservation candidate selected by the communication unit 43 to the speech recognition terminal 1 that has requested the reservation search.

  When the reservation search result is obtained from the reservation server 4, the processor 11 of the speech recognition terminal 1 guides the reservation search result and the fact that the search item can be added by voice output from the speaker 16 (ST36). In addition, when the processor 11 of the voice recognition terminal 1 acquires the reservation search result, the display device displays, on the screen, information such as information indicating that the reservation search result and the search item can be added, in addition to the voice guidance. You may do it. After guiding the reservation search result, the processor 11 recognizes a voice input by the microphone 15 to detect an instruction from the user such as confirmation of reservation or addition of a search item (ST37).

  If the user does not specify a reservation candidate (ST38, NO), processor 11 determines whether the user has instructed to input an additional search item (ST39). If no reservation candidate is specified and no additional search item is input (ST39, NO), the processor 11 returns to ST36 and waits for an instruction input by the user.

  If it is determined that the input of the additional search item has been designated (ST39, YES), processor 11 outputs a voice guidance to the effect of accepting the input of the additional search item through speaker 16 (ST40). For example, the processor 11 outputs a voice guidance indicating that an additional search item for performing a reservation search is to be specified. Further, the processor 11 may read out a search item that can be specified as a supplementary input item by voice, and prompt the user to input an additional search item.

  After outputting the guidance prompting the input of the additional search item, the processor 11 recognizes the voice input by the microphone 15 to specify the additional search information that the user inputs by voice (ST41). In this state, the processor 11 determines whether to execute the reservation search again according to the recognition result of the input voice (ST42).

  For example, when the processor 11 detects an additional search item from the input voice, the processor 11 may execute the reserved search again in response to a user's instruction to perform a search again. As a specific example, when the user utters “window side” and “re-search”, the processor 11 sets “window side” as an additional search item, and executes the reserved search again. Further, when the processor 11 detects an additional search item from the input voice, the processor 11 may execute the reserved search again at a predetermined timing.

  As for the additional search items, it is assumed that the reservation search can be performed without inputting all the supplementary input items unless the information is inconsistent. For the additional search items, priorities for searching for a reservation candidate may be given according to the user's specification. For example, when a seat reservation is specified as an additional search item, the window side may be specified with priority (or the passage side if there is no window side), or only the window side may be specified as a reservation target.

  If it is determined that the reservation search is to be performed again under the search condition including the additional search item detected from the input voice (ST42, YES), the processor 11 returns to ST36 and includes the already specified required input item. The scheduled search is executed again using the sum of the search conditions and the additional search items as the search conditions for the search.

  According to such a detailed search, it is possible to narrow a reservation candidate by setting a more detailed condition after executing a reservation search easily by a simple search. Further, in the detailed search, it is not necessary to input all the specific information, and the reserved search can be performed as an additional search item among the supplementary input items that the user arbitrarily specifies. As a result, it is possible to quickly execute a simple search and also execute a reservation search based on detailed search conditions according to a user's request.

  When the reservation content is specified from the reservation search result (one of the reservation candidates is specified) (ST38, YES), the processor 11 requests the reservation server 4 to make a reservation based on the reservation content specified by the user (ST43). The reservation server 4 stores the reservation contents in the reservation DB 5 in response to the reservation request from the voice recognition terminal 1, thereby completing the registration of the reservation based on the reservation contents requested by the user. When the registration of the reservation is completed, the reservation server 4 sends a notification to the effect that the registration of the reservation content has been confirmed to the speech recognition terminal 1 that has issued the reservation request.

  Upon receiving notification that the contents of the reservation have been registered (ST44), the processor 11 of the voice recognition terminal 1 determines whether or not to execute payment of the fee for the reservation (ST45). For example, the processor 11 may determine whether or not to execute payment of a fee for a registered reservation in response to a user request.

  In addition, the processor 11 may set in advance a period for accepting only the reservation and a period for executing the settlement together with the reservation on the basis of the current date and time. As an example of this case, if the departure time of the train is less than 30 minutes from the current date and time, settlement is executed together with the reservation, and if the departure time of the train is 30 minutes or more from the current date and time, only the reservation is accepted. It may be possible. When only a reservation is accepted, the reservation may be determined by separately executing settlement by the predetermined date and time.

  If it is determined that the settlement of the fee for the reservation content is executed (ST45, YES), the processor 11 setstle the fee for the reservation content by a predetermined settlement method (ST46). The voice recognition terminal 1 may execute voice payment as payment processing. For example, credit card information may be registered as user information, and payment may be executed by voiceprint authentication or authentication using a security code. Further, assuming a login process by user authentication, as a settlement process, points (remaining amount) that can be used for settlement of a fee are accumulated in advance for each user, and points for the fee are subtracted from the points of the user. Payment may be performed.

  When judging not to execute the payment of the fee for the reservation contents (ST45, NO), the processor 11 accepts only the registered reservation and ends the processing. If only the reservation is accepted, the fee for the reservation may be settled separately. For example, the processor 11 may issue a reservation number for a registered reservation, and separately execute a settlement process by a predetermined date based on the reservation number.

  As described above, according to the present embodiment, the speech recognition terminal first requests the reservation server to search for a reservation candidate based on an essential input item detected from input speech as a simple search, and searches for a reservation candidate from the reservation server. Output the result in audio. When an additional search condition is specified after outputting the search result of the reservation candidate by the simple search, the voice recognition terminal requests the reservation server for a re-search result of the reservation candidate with the search condition further adding the additional search condition. Then, the search result of the reservation candidate from the reservation server is output in voice. When an instruction is given to make a reservation with one of the reservation candidates selected from the search results of the reservation candidates output by voice, the voice recognition terminal requests the reservation server to make a reservation based on the designated reservation candidate.

  Thus, according to the present embodiment, the speech recognition terminal can obtain a reservation candidate by a simple speech input. Further, in the voice recognition terminal according to the embodiment, it is possible to search for a reservation candidate by specifying an additional search condition according to a user's request.

Further, the voice recognition terminal according to the present embodiment further receives a search for a reservation candidate based on a search condition registered for each user, and performs a search for a reservation candidate based on the registered search condition according to an instruction from the user. Request to
Thus, the speech recognition terminal according to the present embodiment can search for a reservation candidate using a pre-registered search condition with a simple voice input, and can register a complicated search condition in a simple manner by registering it in advance. It is possible to quickly obtain a reservation candidate desired by the user by voice input.

  Although several embodiments of the present invention have been described, these embodiments are provided by way of example and are not intended to limit the scope of the invention. These new embodiments can be implemented in other various forms, and various omissions, replacements, and changes can be made without departing from the spirit of the invention. These embodiments and their modifications are included in the scope and gist of the invention, and are also included in the invention described in the claims and their equivalents.

  DESCRIPTION OF SYMBOLS 1 ... Voice recognition terminal, 4 ... Reservation server, 11 ... Processor (1st processor), 12 ... Memory, 13 ... Data memory, 14 ... Communication part (1st communication part), 15 ... Microphone, 16 ... Speaker 41: processor (second processor), 42: memory, 43: communication unit (second communication unit).

Claims (6)

A speaker for outputting sound,
A microphone for inputting audio,
A processor that controls the sound output by the speaker and recognizes the sound input by the microphone,
The processor comprises:
When information indicating a required search condition set as a required input item required for searching for a reservation candidate is detected from a recognition result of a voice input by the microphone, a search result of the reservation candidate based on the required search condition is output to the speaker. Output by voice,
After the search result of the reservation candidate is output, when an additional search condition is detected from the recognition result of the voice input by the microphone, a re-search result of the reservation candidate to which the additional search condition is added is output as the voice through the speaker. And
Making a reservation based on the designated reservation candidate when an instruction to reserve with any one of the reservation candidates output from the speaker is detected from the recognition result of the voice input by the microphone;
Voice recognition terminal. The processor adds, as the additional search condition, any number of search conditions detected when an arbitrary number of search conditions among a plurality of search conditions set as supplementary input items are detected from a speech recognition result. Outputting the re-detection result of the reservation candidate in voice by the speaker;
The voice recognition terminal according to claim 1. The processor may further include, when detecting an instruction to search for a reservation candidate based on a registered search condition from a recognition result of a voice input by the microphone, the search result of the reservation candidate using the registered search condition to the speaker. Output by voice,
The voice recognition terminal according to claim 1. The processor specifies additional information required for searching for a reservation candidate using the registered search condition, and detects the registered search condition when the additional information is detected from a recognition result of a voice input by the microphone. And outputting a search result of a reservation candidate using the additional information and the speaker by voice.
The voice recognition terminal according to claim 3. In a voice reservation system having a voice recognition terminal and a reservation server,
The voice recognition terminal,
A first communication unit that communicates with the reservation server;
A speaker for outputting sound,
A microphone for inputting audio,
A first processor that recognizes voice input by the microphone,
The first processor comprises:
When information indicating a necessary search condition necessary for searching for a reservation candidate is detected from a recognition result of a voice input by the microphone, a search result of the reservation candidate based on the essential search condition obtained from the reservation server is obtained by the speaker. Output by voice,
When the additional search condition is detected from the recognition result of the voice input by the microphone after outputting the search result of the reservation candidate, the re-search result of the reservation candidate to which the additional search condition obtained from the reservation server is added is obtained. Sound output by the speaker,
Requesting the reservation server to make a reservation based on the designated reservation candidate when the instruction to reserve with any one of the reservation candidates output from the speaker is detected from the recognition result of the voice input by the microphone;
The reservation server,
A second communication unit that communicates with the voice recognition terminal;
A second processor that searches for a reservation candidate according to a search condition received from the voice recognition terminal via the second communication unit, and transmits a search result of the reservation candidate to the voice recognition terminal,
Voice reservation system. A computer that connects a speaker that outputs audio and a microphone that inputs audio,
Detecting information indicating a necessary search condition necessary for searching for a reservation candidate from the recognition result of the voice input by the microphone,
When the information indicating the essential search condition is detected, the search result of the reservation candidate according to the essential search condition is output as voice by the speaker,
After the search result of the reservation candidate is output, the additional search condition is detected from the recognition result of the voice input by the microphone,
When the additional search condition is detected, a re-search result of the reservation candidate to which the additional search condition has been added is output as a sound by the speaker,
Detecting a reservation instruction from any one of the reservation candidates output by the speaker from the recognition result of the voice input by the microphone;
Making a reservation based on the designated reservation candidate when detecting a reservation instruction with the one reservation candidate;
Voice booking program to do that.