WO2008001549A1

WO2008001549A1 - Audio interaction device, audio interaction method and its program

Info

Publication number: WO2008001549A1
Application number: PCT/JP2007/059586
Authority: WO
Inventors: Toshiki Moriguchi; Tsuyoshi Nakano; Yasutaka Shindoh
Original assignee: Murata Kikai Kabushiki Kaisha
Priority date: 2006-06-26
Filing date: 2007-05-09
Publication date: 2008-01-03
Also published as: JP4505862B2; JPWO2008001549A1

Abstract

A distance and a direction to a user are confirmed at a checking point in an interaction scenario. If the user sits at a short distance in the front or in the slanting front, the interaction is continued but, if that is not the case, the interaction is set to stand by and if the distance to the user becomes short, a commencement of a talk reminds him or her of interaction. The interaction restarts in response to a reply to such a reminding talk but it ends if a plurality of replies cannot be obtained. Thus, user's interest can be confirmed during the interaction and, if his or her interest is low, the interaction can be terminated after the standing-by or reminding is carried out.

Description

Specification

Voice dialogue apparatus, voice dialogue method and program thereof

Technical field

[0001] The present invention relates to a voice dialogue apparatus such as a guidance robot, a voice dialogue method, and a dialogue program, which prevents the dialogue apparatus from speaking into the air and enables natural conversation with a person. It is in.

Background art

[0002] Patent Document 1 (Patent No. 3273620) discloses that a voice interaction device monitors human behavior, speaks when it stops in front of the interaction device for a predetermined time or more, and terminates the conversation when the user leaves. Disclosure.

Patent Document 1: Japanese Patent No. 3273620

Disclosure of the invention

Problems to be solved by the invention

[0003] An object of the present invention is to interrupt a dialogue when the user's interest is reduced during the dialogue, to resume the dialogue when the user's interest is recovered, and to terminate the dialogue when the user's interest does not return. is there.

An additional problem in the present invention is to confirm the user's intention to interact with the natural point in the conversation scenario, and to enable the user to interrupt the conversation with an uninterested user. An additional problem with the present invention is to make the interaction with the user more natural. An additional problem with the present invention is to eliminate the situation where the dialog device advances by ignoring the user when the voice dialog device guides the user.

Means for solving the problem

[0004] The present invention is a voice interactive device including a speaker and a microphone,

A confirmation means for confirming the user's interest by detecting the distance and direction to the user during the dialogue with the user;

A means for switching the operation of the voice interaction device according to the confirmation result of the confirmation means, so that the conversation is continued if the user's interest is high, and is waited if the user's interest is low; Talking to the user for the speaker power prompt during standby, confirming the intention to continue the dialog, continue the conversation, and if the confirmation of the intention to continue the dialog cannot be confirmed, prompt means for waiting further

And ending means for ending the dialogue when the degree of waiting exceeds a predetermined value.

[0005] The present invention is also a voice interaction method using a speaker and a microphone,

While interacting with the user, the user's interest is confirmed by detecting the distance and direction to the user,

If the user's interest is high, the dialogue is continued, and if the user's interest is low, the operation is switched according to the confirmation result.

While waiting, utter the speaker power reminder to the user, confirm the willingness to continue the dialogue, continue the dialogue, and if the intention to continue the dialogue cannot be confirmed, wait further.

The dialogue is terminated when the degree of waiting exceeds a predetermined value.

[0006] The present invention is a program for a voice interaction device comprising a speaker, a microphone, and a sensor for detecting a distance and direction to a user,

During the dialogue with the user, a voice command according to the confirmation result of the confirmation command is used so that the confirmation command for confirming the user's interest by the sensor and the conversation is continued if the user's interest is high, and waits if the user's interest is low. A command for switching the operation of the interactive device;

Talking to the user for the speaker power prompt during standby, confirming the intention to continue the dialog, continue the conversation, and if the confirmation of the intention to continue the dialog cannot be confirmed, a prompt instruction to wait further;

And a termination command for terminating the dialogue when the degree of waiting exceeds a predetermined value.

[0007] Preferably, a scenario of a dialog with the user describing the confirmation point of the user's interest is stored, and when the user's interest is low at the stored confirmation point, the dialog is interrupted and waited. The reminder means utter the reminder when the distance to the user decreases. Do.

Preferably, a moving means is provided so that the user can move with the user, and when the user is away from the user, the user is called from the speaker or the voice interactive device looks back at the user.

The invention's effect

[0008] Whether or not the user is interested in the dialogue with the voice interactive device can be evaluated to some extent depending on the distance between the user and the voice interactive device and whether the user is on the front side of the interactive device. . Therefore, according to the present invention, during the dialogue with the user, if the user leaves the dialogue device or moves from the front side to the side of the dialogue device, the confirmation means detects and waits, Make a prompt utterance at the timing to check whether you intend to continue the dialogue. If the response is not obtained even if the reminder is uttered multiple times, or if the waiting time exceeds a predetermined value or the user has left the dialogue device, the dialogue is terminated. For this reason, if the user loses interest during the dialogue, the dialogue can be terminated, so that the voice dialogue device can be prevented from talking into the air. Along with this, it is possible to reduce the stress of workers resident in the vicinity of the voice interaction device. In addition, if you do not get a response after a single prompting utterance, other prompting utterances that do not end the conversation immediately, or that you have waited for a sufficient amount of time from the user, etc. Wait for the condition to be met and end the conversation. For this reason, it does not give the user the impression that the conversation has ended suddenly.

[0009] When a confirmation point is described in a dialogue scenario and the user's interest is confirmed at the confirmation point, the interest can be confirmed at a break in the dialogue scenario. If there is a break in the scenario, it will not become unnatural even if the dialogue is interrupted. If the user's interest is low, the conversation is interrupted and waits for a predetermined time, for example. Therefore, if the user does not intend to interact with the user, the conversation can be interrupted and waited until the interest is restored.

Furthermore, if the user utters a reminder when the distance to the user is reduced, a natural conversation can be made as if the user had approached the voice dialogue device and made a prompt utterance in response.

[0010] When the voice interactive device guides the user, the distance from the user is increased. If you call from the speaker or the voice interaction device looks back to the user, you can prevent the user from naturally leaving the distance while keeping pace with the user's pace.

[0011] [Fig. 1] Block diagram of the guidance robot of the embodiment

[Fig.2] Diagram showing guidance and confirmation algorithm for user's will of dialogue in the embodiment

[Figure 3] Flowchart representing the algorithm of Figure 2

FIG. 4 is a diagram showing an example of the interest level evaluation table in the embodiment.

FIG. 5 is a diagram showing an example of a standby condition evaluation table in the embodiment.

[Fig. 6] Diagram showing the attitude control algorithm of the guidance robot in the embodiment

[FIG. 7] A flow chart showing an algorithm for maintaining contact with the user during guidance movement in the embodiment.

Explanation of symbols

[0012] 2 Guidance robot 4 User 6 Laser range sensor

8 CCD camera 10 Image recognition unit 12 User detection unit

14 Microphone 16, 22 Amplifier 18 Voice recognition unit

20 Speaker 24 Dialogue control unit 26 Scenario storage unit

28 History storage unit 30 Behavior control unit 32 Walking control unit

34 Attitude control unit 41 Scenario selection process 42 Guidance process

43 Confirmation process 44 Standby process 45 Reminder process 46 End process

50, 52 Tape Nore

BEST MODE FOR CARRYING OUT THE INVENTION

[0013] The following is an optimal example for carrying out the present invention.

Example

[0014] FIGS. 1 to 7 will be described with reference to the guidance robot 2 as an example and its operation. 4 is a user and the person with whom the guidance robot 2 interacts, and 6 is a laser range sensor that detects the distance to the user 4 and the direction viewed from the guidance robot 2. CCD camera 8 Takes a face image, etc., and recognizes the direction of the face and line of sight in the image recognition unit 10 to recognize whether the user is looking at the guidance robot 2 or the other side . The user detection unit 12 evaluates the degree of interest of the user regarding the dialogue with the guidance robot 2 based on the recognition results of the sensor 6 and the image recognition unit 10. The degree of interest may be a two-level evaluation such as presence / absence, or a multi-level evaluation such as 0 to 10 or the like. Furthermore, the user's interest can be either instantaneous values at each point in time or the past interest level can be accumulated. If the past interest degree is accumulated, if the user shows interest for a long time, then the user's interest gradually increases. The degree of interest can be reduced.

[0015] 14 is a microphone, 16 is an amplifier, and the voice recognition unit 18 recognizes the user's voice. 20 is a speaker, 22 is its amplifier, and the dialogue control unit 24 changes the scenario according to the degree of user interest from the user detection unit 12 and the user's voice input from the voice recognition unit 18. And controls the dialog by interrupting and ending the dialog. Reference numeral 26 denotes a scenario storage unit, which stores dialog scenarios and confirms the degree of interest of the user as a confirmation point in the scenario. The history storage unit 28 stores the history of scenario processing and prevents the same part of the scenario from being repeatedly processed. The history storage unit 28 changes the branch destination and jump destination in the scenario according to the history of the conversation up to now, changes the expression even when processing the same address in the scenario, and the first and second times Change the expression of the reminder utterance. For example, in the first reminder, “Can I continue the explanation?”, In the second reminder, “Let's finish the story?” Reference numeral 30 denotes an action control unit that causes the guidance robot 2 to walk with its feet and wheels via the walking control unit 32 and controls a hand and a neck (not shown) via the posture control unit 34. The control units 30 to 34 may not be provided.

FIG. 2 shows the main processing in the embodiment. The scenario selection process 41 controls the branching of the scenario according to the voice input from the user or the user's answer to the reminder process 45. The reminder process 45 is an input that there is no voice input from the user, and that the user's interest remains low. In the guidance process 42, guidance may be given to the user using a speaker, and another dialogue may be performed instead of the guidance. The confirmation process 43 uses the user detection unit 12 to confirm the degree of interest of the user. When the degree of interest of the user at the confirmation point is less than or equal to a predetermined value, the confirmation process 43 starts the standby process 44 and waits for a predetermined time. During this time, the user When the level of mind increases, the prompting process 45 utters a reminder, and if a response to continue the dialogue is obtained, the guidance is continued, or the address is assumed to be of interest to the user. To jump. Further, when a reply indicating that the dialogue is to be terminated is obtained, the guidance is terminated. If an answer meaning continuation of the dialog cannot be obtained, the process further waits. If the user utters a reminder for a predetermined number of times, waits for a predetermined period of time, or the user leaves, the dialog is terminated in a termination process 46. Note that when the end of the dialogue is selected in the scenario selection process 41 or when the guidance is terminated, the dialogue is terminated in the termination process 46. Each process in Fig. 2 can be interpreted as an instruction. In that case, Fig. 2 can be said to show the voice dialogue program of the embodiment.

FIG. 3 shows an algorithm for confirming the degree of interest of the user. Here, it is assumed that guidance is provided within the university department plan, and guidance is provided in Step 11. During this guidance, when the confirmation point in the dialogue scenario is reached, the distance from the user to the guidance robot (a short distance is a factor of great interest), the force in front of the guidance robot (a factor of great interest), It is diagonally in front (a factor with a high degree of interest), it is on the side (a factor with a low level of interest), or the face direction and line of sight (a factor with a high degree of interest when looking at the robot side, the line of sight is off And the degree of interest) (step 12). The reason for evaluating the user's interest only at the confirmation point is to avoid suddenly terminating the conversation in the middle of the sentence when the distance to the user increases while reading the guidance sentence. By placing confirmation points at scenario breaks or long sentence breaks, the conversation can be interrupted at its natural location. In step 13, the degree of interest is evaluated, and if the degree of interest is high, the scenario is continued in step 14, and if the degree of interest is low, the process waits (step 15).

[0018] If the user approaches the guidance robot side during standby, or the user's face direction or line of sight faces the guidance robot side, it can be evaluated that the degree of interest has increased. In this case, a reminder is made (steps 18 and 19). If the user's interest level does not increase, a reminder utterance is made every predetermined time from the start of standby (step 20). Then, if an answer is obtained that means that you want to continue the dialogue in response to the reminder (step 21), the scenario continues. The reminder utterance is "Let's continue the explanation?" If there is a positive response, or if a positive response is detected from the user's gesture, etc., the willingness of dialogue is assumed. Furthermore, if there is an answer that means the end of the dialogue, such as “No” or “Okay,” it ends (step 22). If there is no affirmative answer or no answer to the end of the dialogue, the process returns to the standby of step 15. You may change the message of the reminder utterance between the first time and the second time. For example, if you are interrupting an explanation for the faculty A, the first reminder will ask “Do you want to continue the explanation?” The second reminder will ask, “Do you want to explain other faculties?” In addition, it is okay to suggest a scenario change to the user. Detected that the user's interest has declined, such as when the willingness to talk is not displayed even after multiple reminder utterances, when the waiting time exceeds the specified time, or when the user leaves If so (step 16), terminate the conversation in step 17.

FIG. 4 shows a table 50 for evaluating the degree of interest of the user. For example, the user's face or line of sight is evaluated based on whether the user's distance or user orientation is front, diagonally front, or side force, such as whether the user is approaching, stopping, or retreating. Calculate the degree. In the case of Fig. 4, the user's face is slanted and the distance is slightly far is the factor of interest, and the orientation is front and approaching is the factor of great interest.

[0020] To determine whether or not to end the dialogue, use the standby condition evaluation table 52 in Fig. 5, and enter the total waiting time from the confirmation point and the number of reminders, and continue waiting according to these values. Decide whether to end the conversation.

FIG. 6 shows an attitude control algorithm of the guidance robot in the embodiment. Guidance mouth Bot 2 has a face and head attached, and can change the direction around the neck to see and look back at the user. For one or more users, it detects the distance and orientation, the orientation of the user's face, and whether it is approaching or moving toward an increasing distance (step 25). From these, the degree of interest is evaluated for each user, and the robot looks at that user by turning his head or looking toward the user with the highest interest (step 26). When there is only one person, look at the user, except when the user moves away.

FIG. 7 shows an algorithm in the case where the guidance robot guides the user while walking. While walking, the guidance robot stops at important points and interacts with the user. Guidance is given, but the details are omitted. The guidance robot checks the distance to the user while moving (step 31). If the distance is excessive, it stops or slows down and waits for the user to approach (step 33). If the guidance robot stops or slows down and the distance to the user is too long (step 34), rotate the guidance robot's neck and call backwards or simply call Encourage the user to come closer by looking back (step 35). While maintaining the distance to the user within the appropriate range, guidance is given with the user, and if the user is lost (step 36), the guidance is terminated (step 37).

In the embodiment, the following effects can be obtained.

(1) Since the conversation is interrupted when the user loses interest, it is possible to prevent the guidance robot from speaking into the air and reduce the stress on the surrounding people.

(2) The user's will of dialogue can be confirmed not only at the start of the dialogue but also during the dialogue. In particular, the will of the dialogue can be confirmed naturally at the point where the scenario breaks.

(3) When the user's willingness to interact is low, the dialog is interrupted and the user's willingness to wait is waiting.

(4) To confirm the user's will of dialogue, the will of the dialogue can be confirmed naturally, taking into account the distance to the user, the orientation with the user, whether the user is approaching, the user's face and line of sight.

(5) When the user's interest is considered to have increased, such as when the distance to the user is reduced, the user utters a reminder and the person approaches the guidance robot. You can have a natural dialogue in a breeze.

(6) Even when the guidance robot guides the user, it can move while keeping the distance within a certain range according to the user's pace.

Claims

The scope of the claims

[1] A voice interaction device comprising a speaker and a microphone,

A means for switching the operation of the voice interaction device according to the confirmation result of the confirmation means, so that the conversation is continued if the user's interest is high, and is waited if the user's interest is low;

Talking to the user for the speaker power prompt during standby, confirming the intention to continue the dialog, continue the conversation, and if the confirmation of the intention to continue the dialog cannot be confirmed, prompt means for waiting further

A voice dialogue device comprising: ending means for ending the dialogue when the degree of standby exceeds a predetermined value.

[2] Provide a means for memorizing user interaction scenarios describing points of confirmation of user interest,

2. The voice interactive apparatus according to claim 1, wherein when the user's interest is low at the stored confirmation point, the dialog is interrupted and waited.

[3] The spoken dialogue apparatus according to claim 1, wherein the prompting means utters a prompt when the distance to the user is reduced.

[4] A moving means is provided so that the user can move with the user, and when the user is away from the user, the user is called from the speaker or the voice interactive device looks back at the user. The spoken dialogue apparatus according to claim 1, wherein

[5] A voice interaction method using a speaker and a microphone,

During the standby, utter the above speaker power prompt to the user, confirm the willingness to continue the dialogue, continue the dialogue, and if the intention to continue the dialogue cannot be confirmed, wait further.

Voice conversation method, characterized in that the conversation is terminated when the degree of waiting exceeds a predetermined value. Law.

[6] Memorize the user interaction scenario with the confirmation points of interest of the user,

6. The voice interaction method according to claim 5, wherein if the user's interest is low at the stored confirmation point, the conversation is interrupted and waited.

[7] A program for a voice interaction device comprising a speaker, a microphone, and a sensor for detecting a distance and direction to a user,

During the standby, the speaker is urged to utter the speaker power prompt, the dialogue is continued if the intention to continue the dialogue is confirmed, and the prompt to wait further if the intention to continue the dialogue cannot be confirmed, and

A spoken dialogue program comprising: an end command for terminating a dialogue when the degree of standby exceeds a predetermined value.

[8] Provide an instruction to memorize the user interaction scenario that describes the confirmation points of interest of the user,

8. The spoken dialogue program according to claim 7, wherein when the user's interest is low at the stored confirmation point, the dialogue is interrupted and waited.