JP2019049742A - Voice response device - Google Patents

Abstract

To make a voice response device for performing voice response for input voice easy to use.SOLUTION: A voice response system 100 configured such that a server 90 generates an appropriate response to the voice input in a terminal device 1 and the terminal device 1 outputs a response as voice includes a voice feature recording unit for recording the features of the input voice, a voice match determination unit for determining whether or not the features of the input voice match the features of the voice previously recorded by the voice feature recording unit, and a voice output unit configured to output a response different from that in the case where it is determined that the voice features match, when it is determined that the voice features do not match by the voice match determination unit.SELECTED DRAWING: Figure 1

Description

Cross-reference to related applications

  This international application claims priority based on Japanese Patent Application No. 2012-178454 filed on Aug. 10, 2012 in the Japanese Patent Office, and the international application is based on Japanese Patent Application No. 2012-178454. The entire contents are incorporated into this international application by reference.

  The present invention relates to a voice response device that causes voice to respond to input voice.

  As the above-mentioned voice response device, one is known which searches for an answer to an input question from a dictionary and outputs the searched answer as voice (see, for example, Patent Document 1). There is also known a technology for generating an answer to a question based on the content of a dialog with a user (see, for example, Patent Document 2).

Patent No. 4832097 Patent No. 4924950 gazette

In the above technology, only one question is set to make one answer specified by the dictionary.
It is an aspect of the present invention to make it easier for the user to use a voice response device that makes voice responses to input voice.

The invention of the first aspect is
A voice response device for making a voice response to input voice
A voice feature recording unit that records features of the input voice;
A voice match determination unit that determines whether the features of the input voice match the features of the voice previously recorded by the voice feature recording unit;
A voice output unit that outputs a response different from that in the case where it is determined that the voice features do not match if the voice match determination unit determines that the voice features do not match;
It is characterized by having.

  According to such a voice response device, when the person who has input the voice is different from before, it is possible to return a different response from the case where the person who has input the voice is the same as before. Therefore, it is possible to improve usability for the user as compared with the case where the person who has input the voice makes the same answer regardless of whether the person is the same as before.

By the way, in the above voice response apparatus, as in the invention of the second aspect,
A person specifying unit for specifying a person who has input a voice based on the feature of the input voice;
A control unit that controls the controlled unit in accordance with the input voice;
When the control unit receives an instruction contradicting from a different person, control may be performed by giving priority to an instruction by a person higher in priority according to the priority order set in advance for each person.

According to such a voice response device, even if a contradictory instruction is received from a different person, it is possible to control the controlled unit according to the priority.
In addition, in the case of receiving a contradictory instruction, a voice response may be used to point out the contradiction or present an alternative. At this time, when presenting an alternative, a response may be output in consideration of the weather and the like.

Furthermore, in the voice response device, as in the invention of the third aspect,
A person specifying unit for specifying a person who has input a voice based on the feature of the input voice;
And a schedule recording unit configured to record a schedule based on the input voice for each of the persons.

According to such a voice response device, a schedule can be managed for each person.
In the present invention, the schedule recording unit can be subordinate to the invention according to the second aspect except for the person specifying unit. In the present invention, the priority of the schedule may be changed according to the attribute of the schedule. The scheduled attributes are classified, for example, according to whether or not they can be changed (whether or not there is an influence on the other party).

  The schedule registered earlier according to the schedule attribute may be changed, or the schedule registered later may be registered at a vacant time. Also, when registering a schedule, consider the location of the previous schedule and the location of the later schedule, search for the time to move between them, and search for the time to move between them. A schedule to be registered later may be registered in consideration of it.

  In addition, when a plurality of persons share the same schedule (time zone in which a schedule is to be implemented), as in the case where a plurality of persons managed by the device make a meeting, the schedules of the plurality of persons are vacant. It is also possible to search for the time zone in which you are working and set up a meeting at this time. In addition, when there is no vacant time zone, the already registered schedule may be changed according to the schedule attribute.

When changing the schedule in this way, it is preferable to output that effect as a voice response.
Further, in the voice response device, as in the invention of the fourth aspect, the predetermined contact can not be heard if the input voice can not be heard (that is, if it can be estimated that there is an error as a sentence when converted into characters) You may ask to At this time, the inquiry source and the inquiry destination may be specified by using the position information.

  According to such a voice response device, for example, when the content that the child speaks can not be heard, the correct content is referred to by inquiring the mother, or when the content that the elderly person speaks can not be heard, by asking the family of the elderly person. It is possible to secure the accuracy of the input voice because it can be input from the

  Furthermore, in the voice response device, as in the invention of the fifth aspect, the voice response device is provided with a plurality of databases prepared in advance according to age information indicating the age or age group of the user (the person who input the voice). Alternatively, a database to be used may be selected according to the user's age information, and speech may be recognized according to the selected database.

  According to such a voice response device, the database to be referred to at the time of voice recognition is changed according to the age. Therefore, if words and phrases used frequently according to the age are registered, the voice recognition is performed. Accuracy can be improved.

In the voice response apparatus, as in the invention of the sixth aspect, the age of the user (the person who has input the voice) may be estimated, and the estimated age may be used as age information.
When the user's age is to be estimated, for example, it may be estimated according to the characteristics of the input voice (voice waveform, voice height, etc.), or the camera when the user inputs voice. It may estimate by imaging a user's face by an imaging part, such as.

In addition, when imaging the face of the user, identification of the user or age authentication may be performed.
Furthermore, the present invention may be applied to a face-to-face device such as a cash dispenser. In this case, identity verification such as age verification can be performed using the present invention.

In addition, the present invention may be applied to a vehicle. In this case, the configuration for identifying a person can be used as an alternative to the key of the vehicle.
Although the above-mentioned invention was explained as a voice response device, it may be constituted as a voice recognition device provided with the composition which recognizes the inputted voice. In addition, the invention of each aspect does not have to be premised on other inventions, and can be independent inventions as much as possible.

FIG. 1 is a block diagram showing a schematic configuration of a voice response system to which the present invention is applied. It is a block diagram which shows schematic structure of a terminal device. It is a flowchart which shows the voice response terminal process which MPU of a terminal device performs. It is a flowchart which shows the voice response server process (the 1) which the calculating part of a server performs. It is a flowchart which shows a voice response server process (the 2). FIG. 6A is an explanatory view showing a speech recognition DB. FIG. 6B is an explanatory view of the priority DB. It is a flow chart which shows schedule input processing among voice response server processing. It is explanatory drawing which shows an example of the schedule recorded on schedule DB. It is an explanatory view showing an example of a schedule attribute. It is a flow chart which shows operation input processing among voice response server processing. It is a flowchart which shows the change confirmation process of voice response server processes. It is a flowchart which shows the period designation processing of voice response server processing. It is a flowchart which shows the operation input process of a modification.

Embodiments of the present invention will be described below with reference to the drawings.
[Configuration of this embodiment]
The voice response system 100 to which the present invention is applied is a system configured such that the server 90 generates an appropriate response to the voice input in the terminal device 1 and the terminal device 1 outputs the response as voice. It is. Further, when the input voice includes a command, the control command is output to the target apparatus (the controlled unit 95). Furthermore, it also has a function to manage the schedule of the user.

  In detail, as shown in FIG. 1, in the voice response system 100, communication base stations 80 and 81 communicate with a plurality of terminal devices 1 and various devices (control units 95) such as air conditioners mounted on vehicles and the server 90. And the Internet 85 are configured to be able to communicate with each other. The terminal device 1 may be configured to directly communicate with another terminal device 1 or the controlled unit 95.

  The server 90 has a function as a normal server device. In particular, the server 90 includes an arithmetic unit 101 and various databases (DBs). The arithmetic unit 101 is configured as a known arithmetic device provided with a CPU, and memories such as ROM and RAM, and communicates with the terminal device 1 via the Internet network 85 based on a program in the memory, Various processing such as voice recognition and response generation for performing reading and writing of data in various DBs or talking with a user who uses the terminal device 1 is performed.

  As various DBs, as shown in FIG. 1, voice recognition DB 102, predictive conversion DB 103, voice DB 104, response candidate DB 105, personality DB 106, learning DB 107, preference DB 108, news DB 109, weather DB 110, priority DB 111, schedule DB 112, terminal An information DB 113, an emotion determination DB 114, a health determination DB 115, a report destination DB 117, and the like are provided. The details of these DBs will be described each time the processing is described.

Next, as shown in FIG. 2, the terminal device 1 is configured such that the behavior sensor unit 10, the communication unit 50, the notification unit 60, and the operation unit 70 are provided in a predetermined case.
The behavior sensor unit 10 includes the well-known MPU 31 (microprocessor unit), a memory 39 such as a ROM and a RAM, and various sensors. The MPU 31 is a sensor element constituting the various sensors to be inspected (humidity, wind speed, etc.) In order to be able to detect favorably, processing such as driving a heater for optimizing the temperature of the sensor element is performed.

  The behavior sensor unit 10 includes, as various sensors, a three-dimensional acceleration sensor 11 (3DG sensor), a three-axis gyro sensor 13, a temperature sensor 15 disposed on the back of the housing, and humidity disposed on the back of the housing A sensor 17, a temperature sensor 19 disposed in front of the housing, a humidity sensor 21 disposed in front of the housing, an illuminance sensor 23 disposed in front of the housing, and a rear surface of the housing A wetness sensor 25, a GPS receiver 27 for detecting the current position of the terminal device 1, and a wind speed sensor 29 are provided.

  The behavior sensor unit 10 also includes an electrocardiogram sensor 33, a heart sound sensor 35, a microphone 37, and a camera 41 as various sensors. Each of the temperature sensors 15 and 19 and the humidity sensors 17 and 21 measures the temperature or humidity of the air outside the housing as an inspection target.

  The three-dimensional acceleration sensor 11 receives accelerations applied to the terminal device 1 in three mutually orthogonal directions (vertical direction (Z direction), width direction of the case (Y direction), and thickness direction of the case (X direction)). Detect and output this detection result.

  The triaxial gyro sensor 13 has an angular velocity applied to the terminal device 1 in the vertical direction (Z direction) and in any two directions orthogonal to the vertical direction (the width direction of the housing (Y direction) and the housing The angular acceleration in the thickness direction (X direction) (each counterclockwise velocity in each direction is made positive) is detected, and the detection result is output.

  The temperature sensors 15, 19 include, for example, a thermistor element whose electric resistance changes in accordance with the temperature. In the present embodiment, the temperature sensors 15 and 19 detect the degree of Celsius temperature, and all temperature display described in the following description is performed at the degree of Celsius temperature.

  The humidity sensors 17 and 21 are configured, for example, as a known polymer film humidity sensor. The polymer film humidity sensor is configured as a capacitor in which the amount of water contained in the polymer film changes according to the change in relative humidity, and the dielectric constant changes.

The illumination sensor 23 is configured as, for example, a known illumination sensor provided with a phototransistor.
The wind speed sensor 29 is, for example, a well-known wind speed sensor, and calculates the wind speed from the power (heat radiation amount) required to maintain the heater temperature at a predetermined temperature.

  The heart sound sensor 35 is configured as a vibration sensor that captures the vibration of the user's heart due to pulsation, and the MPU 31 considers the vibration due to the pulsation or the vibration in consideration of the detection result of the heart sound sensor 35 and the heart sound input from the microphone 37. Identify noise from other vibrations and noise.

The wetness sensor 25 detects water droplets on the surface of the housing, and the electrocardiographic sensor 33 detects the user's heartbeat.
The camera 41 is disposed in the casing of the terminal device 1 such that the outside of the terminal device 1 is an imaging range. In particular, in the present embodiment, the camera 41 is disposed at a position at which the user of the terminal device 1 can be imaged.

  The communication unit 50 includes the well-known MPU 51, the wireless telephone unit 53, and the contact point memory 55, and can obtain detection signals from various sensors constituting the action sensor unit 10 through an input / output interface (not shown). It is configured. Then, the MPU 51 of the communication unit 50 executes processing according to the detection result by the behavior sensor unit 10, an input signal input through the operation unit 70, and a program stored in the ROM (not shown).

  Specifically, the MPU 51 of the communication unit 50 has a function as an operation detection device that detects a specific operation performed by the user, a function as a position relationship detection device that detects a position relationship with the user, and a row by the user. The function as an exercise load detection apparatus which detects the exercise load which is to be performed, and the function which transmits the processing result by MPU51 are performed.

  The wireless telephone unit 53 is configured to be able to communicate with, for example, a base station of a mobile phone, and the MPU 51 of the communication unit 50 outputs the processing result of the MPU 51 to the notification unit 60 or via the wireless telephone unit 53. Then, the information is sent to a preset destination (the contact recorded in the contact memory 55).

  The contact point memory 55 functions as a storage area for storing location information of a user's visit destination. In the contact memory 55, information of a contact (such as a telephone number) to be contacted when an abnormality occurs in the user is recorded.

  The notification unit 60 includes, for example, a display 61 configured as an LCD or an organic EL display, an illumination 63 made of an LED capable of emitting light in seven colors, and a speaker 65, for example. The respective units constituting the notification unit 60 are drive-controlled by the MPU 51 of the communication unit 50.

Next, the operation unit 70 includes a touch pad 71, a confirmation button 73, a fingerprint sensor 75, and a relief request lever 77.
The touch pad 71 outputs a signal corresponding to the position and pressure touched by the user (user or the user's guardian).

  The confirmation button 73 is configured such that the contact point of the built-in switch is closed when the user presses it, and the communication unit 50 can detect that the confirmation button 73 has been pressed. There is.

The fingerprint sensor 75 is a known fingerprint sensor, and is configured to be able to read a fingerprint using, for example, an optical sensor. In addition, instead of the fingerprint sensor 75, for example, a sensor that recognizes the shape of a vein of the palm, or the like that can recognize human physical characteristics (means that can perform biometrics authentication: specifying an individual) Any means can be adopted.

It also has a relief request lever 77 connected to a predetermined contact when operated.
[Process of this embodiment]
The process implemented in such a voice response system 100 will be described below.

  The voice response terminal process implemented in the terminal device 1 is a process of receiving a voice input by the user and sending this voice to the server 90, and when receiving a response to be output from the server 90, reproducing the response by voice . This process is started when the user inputs that voice input is to be performed via the operation unit 70.

  Specifically, as shown in FIG. 3, first, the input from the microphone 37 is received (ON state) (S2), and imaging (recording) by the camera 41 is started (S4). Then, it is determined whether there is a voice input (S6).

  If there is no voice input (S6: NO), it is determined whether or not a timeout has occurred (S8). Here, the timeout indicates that the allowable time for waiting for processing has exceeded, and in this case, the allowable time is set to, for example, about 5 seconds.

If it has timed out (S8: YES), it will shift to the processing of S30 mentioned later. Moreover, if it has not timed out (S8: NO), it will return to the process of S6.
If there is a voice input (S6: YES), the voice is recorded in the memory (S10), and it is determined whether the voice input is completed (S12). Here, when the voice is interrupted for a predetermined time or more, or when it is input that the voice input is to be terminated through the operation unit 70, it is determined that the voice input is finished.

  If the voice input is not completed (S12: NO), the process returns to S10. If the voice input is completed (S12: YES), data such as an ID for identifying itself, voice, and a captured image are transmitted to the server 90 as a packet (S14). The process of transmitting data may be performed between S10 and S12.

Subsequently, it is determined whether transmission of data is completed (S16). If the transmission is not completed (S16: NO), the process returns to S14.
If the transmission is completed (S16: YES), it is determined whether the data (packet) to be transmitted in the voice response server process described later has been received (S18). If the data has not been received (S18: NO), it is determined whether or not a timeout has occurred (S20).

If it has timed out (S20: YES), the process proceeds to the process of S30 described later. Moreover, if it has not timed out (S20: NO), it will return to the process of S18.
If data is received (S18: YES), a packet is received (S22). In this process, a response to character information is acquired.

Then, it is determined whether the reception is completed (S24). If the reception has not been completed (S24: NO), it is determined whether or not a timeout has occurred (S26).
If it has timed out (S26: YES), the notification unit 60 outputs that an error has occurred, and ends the voice response terminal process. Moreover, if it has not timed out (S26: NO), it will return to the process of S22.

If the reception is completed (S24: YES), the response based on the received packet is output from the speaker 65 by voice (S28). When such processing ends, the voice response terminal processing ends.

  Subsequently, a voice response server process performed by the server 90 (external device) will be described with reference to FIG. The voice response server process is a process of receiving a voice from the terminal device 1, performing voice recognition for converting the voice into character information, and generating a response to the voice and returning it to the terminal device 1.

  As the details of the voice response server process, as shown in FIG. 4 (and FIG. 5), first, it is determined whether or not a packet from any of the terminal devices 1 has been received (S42). If a packet has not been received (S42: NO), the process of S42 is repeated.

  If the packet is received (S42: YES), the terminal device 1 of the communication partner is specified (S44). In this process, the terminal device 1 is identified by the ID of the terminal device 1 contained in the packet.

  Subsequently, an image captured by the camera 41 included in the packet is acquired (S70), and the feature of the audio included in the packet is detected (S72). In this process, features such as voice waveform features (voiceprints) and sound levels are detected.

  Subsequently, the age group of the person who has input the voice is specified from the features of the captured image and the voice obtained by imaging the user (S74). In this process, the tendency of the voice feature and the age group is stored in advance in the speech recognition DB 102, and the age group is specified by referring to the speech recognition DB 102. In addition, a known technique for estimating the age of the user from the captured image is used in combination.

  Next, a person is specified from these voice features (S76). Here, in the voice recognition DB 102, the feature of the voice of each person is stored in advance in correspondence with the name of the person. In this process, the person is specified by referring to the voice recognition DB 102.

  When recording the characteristics of each person's voice, for example, only the name is input by the user as voice or characters, and the voice feature is captured when the name is input or when voice input is performed thereafter. Recording should be done. In addition, the technique of personal identification by an image may be used in the present embodiment.

  Subsequently, the voice and the detected voice feature are recorded in the voice recognition DB 102 (S78), and a database to be used when performing voice recognition is selected (S80). Here, as shown in FIG. 6A, the speech recognition DB 102 targets an infant DB targeting four years old, a child DB targeting five to ten years, and a teenager (10 to 19 years old). And a youth DB for 20's (20 to 29), a middle-aged DB for 30's to 50's (30 to 59), a middle age DB for 60's and over ing.

  Each DB is provided with a dictionary database that associates speech waveforms with letters (sounds or words) for recognizing speech as letters. Then, in each DB, the manner in which the user speaks (the tendency of the audio feature), the word having a tendency to be used for the age group, and the like are recorded as different information for each age group.

  In particular, in each DB, the age range to be used is set narrower as the younger age. The reason for this is that the younger the age, the more likely you are to change the manner in which you speak and the ability to create new words so that you can respond immediately to these changes.

  In the process of S80, one database (any of those shown in FIG. 6A) matching the age group is selected and set according to the estimated user age. Subsequently, the voice contained in the packet is recognized (S46).

  Here, in the prediction conversion DB 103, words that are likely to be used subsequently to a certain word are associated. In this process, a known speech recognition process is performed by referring to a selected database of the speech recognition DB 102 and the predictive conversion DB 103 to convert speech into character information.

  Subsequently, an object in the captured image is specified by performing image processing on the captured image (S48). Then, the emotion of the user is determined based on the waveform of the voice, the ending of the word, etc. (S50).

  In this process, the user refers to the emotion determination DB 114 in which the waveform (voice color) of the speech and the word ending are associated with the categories of emotions such as anger, joy, embarrassment, sadness, and exaltation. It is determined whether the emotion of the above corresponds to any category, and the determination result is recorded in the memory. Subsequently, by referring to the learning DB 107, a word which the user often speaks is searched, and a portion where the character information generated by the speech recognition is ambiguous is corrected.

  In addition, in the learning DB 107, characteristics of the user such as a word that the user often speaks and a habit of pronunciation are recorded for each user. In addition, addition and correction of data to the learning DB 107 are performed in a conversation with the user. Further, also in the predictive conversion DB 103, the emotion determination DB 104, etc., data may be stored by being divided into each age group, as in the speech recognition DB 102.

Subsequently, the corrected character information is specified as the input character information (S54). Then, as a result of these processes, it is determined whether the voice can be recognized as character information (S82).
In this process, when there is a defect as a sentence (for example, when there is a grammatical error), it is considered that even if the sentence is completed, it can not be recognized. If it can not be recognized as text information (S 82: NO), a predetermined voice (for example, the following words) is sent to a predetermined contact address (contact address set for each terminal device 1) registered in advance in the report destination DB 117 Can not be recognized. Since the recorded voice is played back, please give a sentence such as “speak the correct sentence.” And the voice input by the user to make an inquiry (S 84).

  In this process, for example, when a child having a bad tongue is a user, if it can not be recognized as character information, the mother terminal device 1 registered as a predetermined contact is inquired or an old man becomes a user In that case, ask the family.

  Subsequently, it is determined whether the packet to which the voice has been input by the inquired party has been received (S86). If the packet is not received (S86: NO), this process is repeated. If a packet is received (S86: YES), the process returns to S54.

  Further, if the character information can be recognized in the process of S82 (S82: YES), it is determined whether the character information is for inputting a schedule (S88). If the schedule is not to be input (S88: NO), the process proceeds to S92 described later.

  If the schedule is to be input (S88: YES), schedule input processing is performed to manage the schedule (S90). In this process, as shown in FIG. 7, first, the schedule of the specific person who is the target of inputting the schedule is extracted (S102).

  In this process, as shown in FIG. 8, schedule data in which a specific person and time are arranged in a matrix is extracted from the schedule DB 112 (S102), and the input schedule (time zone, schedule content, location information) Is temporarily registered (S104).

  Subsequently, it is determined whether there is a conflict in the schedule (schedule) (S106). For example, for Mr. A, as shown in FIG. 8, a meeting schedule has already been registered from 10 o'clock on September 1st, but if it is instructed to put another schedule at this same time, a conflict will occur. Judge as yes.

  If there is a conflict (S106: YES), the process proceeds to S128 described later. Moreover, if there is no competition (S106: NO), the place where the schedule is implemented is extracted for the schedule before and after (S108).

  Then, in consideration of the time and place where the schedule before and after is implemented, the movement time until moving to the place where the temporarily registered schedule is implemented is calculated (S110). In this process, the time required for movement is calculated using, for example, a known transfer guidance program. For example, to move from Marunouchi in Tokyo to Nagoya requires approximately 2 hours of travel time.

  Subsequently, it is determined whether it is possible to move between the place where the temporarily registered schedule is implemented and the place where the preceding and subsequent schedules are implemented (S122). In this process, the time required for movement is compared with the length of idle time, and if the idle time is long, movement is permitted.

  If it is possible to move (S122: YES), this schedule is officially registered in the schedule DB 112 (S124), the fact that the registration is completed is recorded (S126), and the schedule input process is ended.

  In addition, if the movement is not possible (S122: NO), it is determined whether or not the schedule before or after or the temporarily registered schedule can be changed (S128). Here, for each schedule recorded in the schedule DB 112, a schedule attribute is set, and as the schedule attribute, as shown in FIG. 9, a level according to the importance is set. For example, at level A, an appointment (a promise of a visit) with a customer corresponds, and a change in schedule is prohibited.

Also, at level B, for example, appointments other than the customer correspond in the company, and it is impossible to change the schedule. Also, at Level C, private events correspond and it is possible to change the schedule.
Here, in the present process, when registering a schedule (in the case of the process of S124), the schedule attribute is recognized from the content, and the schedule attribute is also registered. Also, with regard to temporarily registered schedules, this process recognizes schedule attributes.

  If it is possible to change the previous or subsequent schedule or the temporarily registered schedule (S128: YES), a proposed change is presented (S130). Here, the proposal for change is to move a changeable schedule (that is, a schedule belonging to level C), and to allow a user (target person) to move between places where there is no conflict and the schedule is implemented. To present.

Then, the change flag is set to ON (S132), and the schedule input process is ended.
Also, if it is not possible to change the previous or subsequent schedule or the temporarily registered schedule (S128: NO), the fact that the schedule has been duplicated is recorded (S134), and the schedule input process is ended.

  When such a schedule input process is completed, the process returns to FIG. 5 to determine whether the character information is an operation input that is an instruction for operating the controlled unit 95 (S92). If it is not operation input (S92: NO), it will transfer to the process of S96 mentioned later.

  If it is an operation input (S 92: YES), the operation input processing is performed (S 94). This process is a process of controlling the operation of the controlled unit 95 according to the input voice. Specifically, as shown in FIG. 10, first, the content of the command is recognized (S202). The command contents include, for example, changing the reception channel and volume of the television receiver corresponding to the controlled unit 95, increasing the set temperature of the air conditioner of the vehicle corresponding to the controlled unit 95 by 1 ° C., etc. Do.

  Subsequently, it is determined whether or not there has been a previous instruction (for example, one in the past predetermined time (for example, within 10 minutes) to the same controlled unit 95) (S204). If there is no past command to the same controlled unit 95 (S204: NO), the process proceeds to the process of S216 described later.

  If there is a command in the past for the same controlled unit 95 (S204: YES), the command in the past is extracted (S206), and it is determined whether there is any contradiction with the command in the past (S208). Here, contradiction, for example, when the air conditioner of the vehicle corresponds to the controlled portion 95, there is a past command to lower the set temperature by 1 ° C., while the air conditioner of the vehicle contradicts this. This corresponds to the case where a command to increase the set temperature by 1 ° C. is input.

  Further, for example, when the television receiver corresponds to the controlled part 95, the volume is further changed immediately after the reception channel is changed, when an instruction to change to another reception channel is received, or immediately after the volume is changed. The case where the command to do is input etc. corresponds.

  If there is no contradiction (S208: NO), the process proceeds to S216. In addition, when there is a contradiction (S208: YES), it is determined whether or not the person who has input the contradictory command matches (S210). If the person who inputs the contradictory command does not match (S210: NO), the priority order of the person who has input these contradictory commands is acquired (S212).

  Here, in the priority DB 111, as shown in FIG. 6B, a person and a priority are stored in association with each other. For example, when Mr. A and Mr. B respectively input contradictory commands, the first place of Mr. A and the fourth place of Mr. B are acquired from priority DB 111.

  Subsequently, the instruction from the one with the highest priority is set. For example, Mr. A, who has the first priority, gives a command to "increase the air conditioner setting temperature of the vehicle by 1 ° C." and Mr. B, who has the fourth priority, decreases the setting temperature of the vehicle air conditioner by 1 ° C. If the command is issued, Mr. A's directive is applied and Mr. B's directive becomes invalid.

  Then, the set command is transmitted to the controlled unit 95 (S218), and the operation input process is ended. If the person who inputs the contradictory command matches in the process of S210 (S210: YES), the most recently input command is set (S216), the process of S218 described above is performed, and the operation is performed. End input processing.

  When such an operation input process is completed, the process returns to FIG. 5, and it is determined whether the change flag is set to ON (S96). If the change flag is OFF (S96: NO), the process proceeds to S56 described later.

  If the change flag is ON (S96: YES), a change confirmation process is performed (S98). The change confirmation process is a process of confirming the user's intention whether the schedule may be changed like the proposed change, when the proposed change of the schedule is presented.

In the change confirmation process, as shown in FIG. 11, first, it is determined whether or not there is an answer to the proposed change (S402). If there is no response to the proposed change (S402: NO), it is determined whether the registration for this schedule has been completed (S404). In other words, it is determined whether the registration has been completed by, for example, the user re-entering the schedule.

  If the registration for this schedule has been completed (S404: YES), the change flag is set to OFF (S412), and the change confirmation process is ended. If the registration for this schedule has been completed (S404: NO), the change confirmation process is ended.

  Further, in the process of S402, if there is an answer to the proposed change (S402: YES), for example, it is determined whether or not an answer indicating that the proposed change is acceptable is obtained (S406). If the answer indicating that the proposed change is acceptable is obtained (S406: YES), the presented proposed change is registered as a schedule (S408), and the registration completion is recorded (S410). Then, the process of S412 described above is performed, and the change confirmation process is ended.

For example, if an answer indicating that it is not good according to the proposed change is obtained (S406: NO), another proposed change is presented (S414), and the change confirmation process is ended.
When such a change confirmation process is completed, the process returns to FIG. 5, and a response is obtained from the response candidate DB 105 by searching the response candidate DB 105 for a sentence similar to the character information as an input (S56). Here, in the response candidate DB 105, character information as an input and an output as a response are uniquely associated.

  For example, when a schedule is input, when registration is completed, a response such as “registration completed.” Is output, and when it is recorded that a schedule is duplicated, “a schedule is duplicated”. Response is output. In addition, when the change flag is turned on, the response to the generated change proposal is such as "This move is not possible considering the schedule before and after this schedule ... ... How is it like?" It is output.

  If a command is input and control is performed as instructed, a response such as "OK" is output, and if control can not be performed as instructed, a response such as "Inconsistent instructions" It is output.

  Also, when the text information "Today's ※ weather" is input, a voice "Today's ※ weather is ※" is output. However, the part “※” is acquired by accessing the weather DB 110 in which the area name is associated with the weather forecast for several days in the area.

  Subsequently, the response content is converted into voice (S62). In this process, based on the database stored in the voice DB 104, a process of outputting the response content (character information) as voice is performed.

Then, the generated response (voice) is transmitted as a packet to the terminal device 1 of the communication partner (S64). The packet may be transmitted while generating the voice of the response content.
Subsequently, the contents of the conversation are recorded (S68). In this process, the input character information and the output response content are recorded in the learning DB 107 as conversation content. At this time, keywords (words recorded in the speech recognition DB 102) included in the conversation content and characteristics at the time of pronunciation are recorded in the learning DB 107.

When such processing ends, the voice response server processing ends.
[Effect by this embodiment]
In the voice response system 100 described in detail above, the server 90 (arithmetic unit 101) records the features of the input voice, and the features of the input voice match the features of the previously recorded voice. It is determined whether or not. Then, if it is determined that the voice features do not match, the server 90 outputs a response different from that in the case where it is determined that the voice features match.

  According to such a voice response system 100, when the person who has input the voice is different from before, it is possible to return a response different from the case where the person who has input the voice is the same as before. Therefore, it is possible to improve usability for the user as compared with the case where the person who has input the voice makes the same answer regardless of whether the person is the same as before.

  Further, in the voice response system 100, the server 90 specifies the person who has input the voice based on the feature of the input voice, and controls the controlled unit 95 according to the input voice. At this time, when the server 90 receives a contradictory instruction from a different person, control is performed by prioritizing an instruction from a higher priority person in accordance with the priority order set in advance for each person.

  According to such a voice response system 100, even when a contradictory instruction is received from a different person, control of the controlled unit 95 can be performed according to the priority.

Furthermore, in the voice response system 100, the server 90 records a schedule based on the input voice for each person.
According to such a voice response system 100, a schedule can be managed for each person.

  Also, in the voice response system 100, the server 90 changes the priority of the schedule according to the attribute of the schedule. Here, the scheduled attribute is classified, for example, according to whether or not it can be changed (whether or not there is an influence on the other party).

  Then, the schedule registered earlier according to the schedule attribute is changed, or the schedule registered later is registered as the vacant time. Also, when registering a schedule, consider the location of the previous schedule and the location of the later schedule, search for the time to move between them, and search for the time to move between them. Register the schedule to be registered later in consideration.

  Further, in the voice response system 100, the server 90, in the case where a plurality of persons share the same schedule, as in the case where a plurality of persons managed by the system 100 make a meeting, the schedule of the plurality of persons Search for open time zones and set up meetings at this time. In addition, if the schedule is not available, the already registered schedule is changed according to the schedule attribute.

According to such a voice response system, the usability can be further improved.
Furthermore, in the voice response system 100, when changing the schedule in this way, the server 90 outputs that effect as a voice response. According to such a voice response system, confirmation can be made to the user when changing the schedule.

  Further, in the voice response system 100, the server 90 inquires the contents of a statement to a predetermined contact when the input voice can not be heard (that is, when there is an error as a sentence when converted into characters). Also, the voice that can not be heard is recorded, the voice is transmitted to a predetermined contact, and the voice is input again to the contact person.

According to such a voice response system 100, for example, when the content spoken by the child can not be heard, the voice input by the inquiry to the mother or when the content spoken by the old person can not be heard, the query to the family of the elderly It is possible to secure the accuracy. At this time, the inquiry source and the inquiry destination may be specified using position information.

  Further, in the voice response system 100, the server 90 is provided in advance with a plurality of databases prepared in accordance with age information indicating the age or age group of the user (the person who has input the voice). The database to be used is selected according to the person's age information, and the speech is recognized according to this selected database.

  According to such a voice response system 100, the database to be referred to at the time of voice recognition is changed according to the age, so if words and word phrases frequently used according to the age are registered, the voice The accuracy of recognition can be improved.

Further, in the voice response system 100, the age of the user (the person who has input the voice) is estimated, and the estimated age is used as age information.
When the user's age is to be estimated, for example, it may be estimated according to the characteristics of the input voice (voice waveform, voice height, etc.), or the camera when the user inputs voice. And so on by imaging the face of the user by the imaging unit.

In addition, when imaging the face of the user, identification of the user and age verification are performed.
According to such a voice response system 100, voice recognition can be performed more accurately.

Other Embodiments
The embodiment of the present invention is not limited to the above embodiment, and can take various forms within the technical scope of the present invention.

  For example, in the schedule input process shown in FIG. 7, a period (such as a schedule and a time zone) is designated as “between September 1 and 3”, and the schedule setting is left to the voice response system 100. May be In this case, for example, a period such as "a meeting with Mr. B and Mr. C for one hour on September 1" is designated between the processing of S102 and the processing of S104. It is determined whether a schedule setting request has been made (S103).

  If there is no such schedule setting request (S103: NO), the processing of the above-mentioned S104 and thereafter is performed. Further, if there is such a schedule setting request (S103: YES), a period designation process is performed (S136), and when this process ends, the schedule input process is ended.

  In the period designation process, as shown in FIG. 12, first, it is determined whether the input request requires adjustment of a schedule of a plurality of persons. For example, if Mr. A enters "A meeting with Mr. B, Mr. C for one hour on Sept. 1.", it is necessary to refer not only to Mr. A but also to Mr. B and Mr. C's schedule. Because of this, it is determined that it is necessary to adjust the schedule of multiple persons.

  If it is necessary to adjust the schedule of multiple persons (S302: YES), the schedule of target persons other than the user (Ms. B and C involved in a schedule other than Mr. A who has input the voice) is extracted (S304) , And proceeds to the process of S306.

  If it is not necessary to adjust the schedule of a plurality of people (S302: NO), it is determined whether there is a time when the schedule of all the target persons is available within the designated period (S306).

  If it is a time for which everyone's schedule is open (S306: YES), this schedule is registered in the schedule DB 112 (S308), recording completion of registration is recorded (S310), and the schedule input process is ended. Also, if there is no time when all the plans are open (S306: NO), information (time, place) about the previous and subsequent plans is extracted (S312).

Subsequently, it is determined whether or not the schedule before and after can be changed (S314). If it is possible to change the schedule before and after (S314: YES), a proposed change is presented (S316).
Then, the change flag is set to ON (S318), and the schedule input process is ended. Also, if it is not possible to change the schedule before and after (S314: NO), the fact that the schedule has been duplicated is recorded (S320), and the schedule input process is ended.

In addition, although the process of S110 and S112 was abbreviate | omitted in a period designation process, you may implement these processes.
Furthermore, in the operation input process, as shown in FIG. 13, instead of the processes of S212 and S214, a weather forecast may be acquired (S232), and an alternative plan may be set according to the weather forecast (S234). For example, as a result of acquiring the weather forecast, if the temperature tends to rise from now on, we propose a proposal to lower the set temperature of the air conditioner, and if the temperature tends to fall from now on, the proposal to raise the set temperature of the air conditioner suggest. Also, if it is likely to rain from now on, we will propose to close the window.

Even in this way, usability can be improved.
Although voice recognition is used as a configuration for inputting character information in the above embodiment, the present invention is not limited to voice recognition, and may be input using an input unit (operation unit 70) such as a keyboard or a touch panel. Further, although the operation of “converting the input voice into character information” is performed by the server 90, the operation may be performed by the terminal device 1.

  Furthermore, in the voice response system 100, the computing unit 101 learns (records and analyzes) the user's actions (speaking, moving location, things shown on the camera) to compensate for the lack of words in the user's conversation. You may do so.

  For example, in response to the question that "the curry is good" for the question "Today is the hamburger?" However, the reason for saying that curry is good is transmitted.

Also, such a configuration can be implemented during a telephone call, and can also be configured to freely participate in the user's conversation.
Furthermore, in the voice response system 100, the server 90 may obtain a response candidate from a predetermined server or the Internet.

According to such a voice response system 100, response candidates can be obtained not only from the server 90 but also from any device connected via the Internet, a dedicated line or the like.
Furthermore, the present invention may be applied to a face-to-face device such as a cash dispenser. In this case, identity verification such as age verification can be performed using the present invention.

  In addition, the present invention may be applied to a vehicle. In this case, the configuration for identifying a person can be used as an alternative to the key of the vehicle. Although the above-mentioned invention was explained as voice response system 100, it may constitute as a voice recognition device which recognizes inputted voice.

In the above embodiment, the server 90 performs the main processing while the terminal device 1 and the server 90 communicate with each other, i.e., a so-called cloud system. However, a part or all of the processing (the processing shown in the flowchart) It may be implemented in In this case, the process related to communication between the terminal device 1 and the server 90 can be omitted.

The controlled unit 95 corresponds to any device that performs control according to an external command.
Furthermore, in the voice response system 100, the sound to be emitted may include an identification sound which is a sound indicating mechanical sound. This is to make it possible to distinguish between mechanical sound and human speech. In this case, it is preferable that the identification sound includes an identifier indicating which device emits the sound. In this way, generation sources of a plurality of types of mechanical sounds can be identified.

  Such identification sound may be an audible sound or an inaudible sound. When the identification sound is an inaudible sound, an identifier may be embedded in the voice by using a digital watermark technology.

  In the above embodiment, the response corresponding to the input voice is output as voice. However, the response is not limited to voice input, and the response corresponding to this input may be voice output. For example, a camera for detecting a change in the shape of the user's mouth may be provided, and means for estimating what kind of language the user is speaking may be provided based on the shape of the user's mouth.

  In this case, the correspondence relationship between the shape of the mouth and the sound may be prepared as a database, the sound may be estimated from the shape of the mouth, and the words may be estimated from the sound. According to such a configuration, the user can input voice without actually making a sound.

In addition, the shape of the mouth may be used as an aid when performing input using voice. In this manner, voice recognition can be performed more reliably even if the user's slippage is bad.
Furthermore, in preparation for the case where the user can not input the voice, the user may be configured to be able to perform an input instead of the voice by selecting the history of the input by the user on the display. In this case, the history may be simply displayed in the order of newness, and it is estimated that the possibility of being used is high in consideration of the usage frequency of the input content included in the history and the time zone in which the input content is input. The contents may be displayed in order.

  When the terminal device 1 is mounted on a vehicle, a specific action such as unlocking is performed in response to a call from the owner (user) in response to a call to the vehicle. It is also good. In this way, the voice can be used as a key, and even if the owner of the vehicle loses sight of the vehicle in a large parking lot or the like, the vehicle can be found by calling the vehicle.

[Relationship between the configuration of the present invention and the configuration of the embodiment]
The voice response system 100 in the present embodiment corresponds to an example of the voice response device in the present invention. Further, among the processes executed by the server 90, the process of S74 corresponds to an example of the person specifying unit in the present invention, and the process of S78 corresponds to an example of the audio feature recording unit in the present invention.

  Further, the process of S210 corresponds to an example of the speech match determination section in the present invention, and the processes of S214 and S216 correspond to an example of the speech output section in the present invention. The processes of S208 and S218 correspond to an example of the control unit in the present invention, and the process of S90 corresponds to an example of the schedule recording unit in the present invention.

  DESCRIPTION OF SYMBOLS 1 ... Terminal device, 10 ... Behavior sensor unit, 11 ... 3 dimensional acceleration sensor, 13 ... 3 axis gyro sensor, 15 ... Temperature sensor, 17 ... Humidity sensor, 19 ... Temperature sensor, 21 ... Humidity sensor, 23 ... Illuminance sensor, 25: wetness sensor, 27: GPS receiver, 29: wind speed sensor, 33: electrocardiogram sensor, 35: heart sound sensor, 37: microphone, 39: memory, 41: camera, 50: communication unit, 53: wireless telephone unit, 55: contact memory, 60: notification unit, 61: display, 63: illumination, 65: speaker, 70: operation unit, 71: touch pad, 73: confirmation button, 75: fingerprint sensor, 77: rescue request lever, 80: Communication base station, 85: Internet network, 90: Server, 95: Controlled unit, 100: Voice response system, 101: Arithmetic unit.

Claims (3)

A voice response device for making a voice response to input voice
A voice feature recording unit that records features of the input voice;
A voice match determination unit that determines whether the features of the input voice match the features of the voice previously recorded by the voice feature recording unit;
A voice output unit that outputs a response different from that in the case where it is determined that the voice features do not match if the voice match determination unit determines that the voice features do not match;
A voice response device comprising: In the voice response device according to claim 1,
A person specifying unit for specifying a person who has input a voice based on the feature of the input voice;
A control unit that controls the controlled unit in accordance with the input voice;
The voice response device characterized in that, when the control unit receives an contradictory instruction from a different person, control is performed by prioritizing an instruction from a person higher in priority according to a priority order set in advance for each person. In the voice response device according to claim 1,
A person specifying unit for specifying a person who has input a voice based on the feature of the input voice;
A schedule recording unit for recording a schedule based on the input voice for each of the persons;
A voice response device comprising:

Images