IP Library › Granted Patent US 12,106,756
Granted Patent B2
US 12,106,756 · App. 17/626,030 · Granted Oct 1, 2024

Interaction device

Inventors: Go Tanaka (Chiyoda-ku, JP); Yutaro Shiramizu (Chiyoda-ku, JP)
Assignee: NTT DOCOMO, INC.
G10L15/22A63F13/215A63F13/44A63F13/80G06N20/00G10L15/08A63F2300/1081A63F2300/638A63F2300/8094G10L2015/227
View Patent ↗
Loading inventors, assignments & file history…
Monitor This Case
Get email alerts when status or documents change.
Order Certified Copies
Most orders are placed with the USPTO same day — all within 24 business hours.
Order via The Patent Place →
Pre-filled with this patent's details
Quick Facts
Patent No.
US 12,106,756
App. No.
17/626,030
Granted
Oct 1, 2024
Kind
B2
Abstract

An interaction apparatus 10 includes an acquiring unit 13 for acquiring a system message, a setting unit 14 for setting a wait time for receiving an answer from a user based on the system message, a measuring unit 16 for measuring an elapsed time after the system message is output to the user, a user utterance acquiring unit 15 for acquiring a user utterance, and a controller 17 for controlling an interaction with the user so that an additional action corresponding to the system message is output to the user. The measuring unit 16 determines whether or not to subtract the elapsed time, based on the system message and the user utterance, when the user utterance not corresponding to the answer to the system message is acquired by the user utterance acquiring unit 15 after the system message is output.

Claims (38)

1. An interaction apparatus comprising:

processing circuitry configured to

acquire a system message for a user;

set a wait time for receiving an answer from the user based on the system message;

measure an elapsed time after the system message is output to the user;

acquire a user utterance uttered by the user; and

control an interaction with the user so that an additional action corresponding to the system message is output to the user when the measured elapsed time exceeds the wait time,

wherein the processing circuitry is configured to determine whether or not to subtract the elapsed time, based on the system message and the user utterance, when the user utterance not corresponding to the answer to the system message is acquired after the system message is output.

2. The interaction apparatus according to claim 1 ,

wherein the processing circuitry is configured to:

determine not to subtract the elapsed time when the user utterance is a filler and the system message meets a predetermined condition; and

determine to subtract the elapsed time when the user utterance is a filler and the system message does not meet the predetermined condition.

3. The interaction apparatus according to claim 2 ,

wherein the predetermined condition includes a condition that the system message corresponds to a system message in a game in which words along a predetermined theme are alternately answered between the user and a system.

4. The interaction apparatus according to claim 2 ,

wherein the processing circuitry is configured to determine whether or not the user utterance is the filler by using a detection model that is machine-learned to input a speech recognition result of the user utterance and to output an estimation result of whether or not the user utterance is the filler.

5. The interaction apparatus according to claim 2 ,

wherein the processing circuitry is configured to determine whether or not the user utterance is the filler based on a profile information on an attribute of the user.

6. The interaction apparatus according to claim 1 ,

wherein the processing circuitry is configured to determine whether or not to subtract the elapsed time by using a subtractive decision model that is machine-learned to input an information based on the system message and a speech recognition result of the user utterance and to output an information indicating whether or not to subtract the elapsed time.

7. The interaction apparatus according to claim 6 ,

wherein the subtractive decision model is machine-learned to further input a profile information on an attribute of the user and output the information indicating whether or not to subtract the elapsed time.

8. The interaction apparatus according to claim 1 ,

wherein the processing circuitry is configured to set the wait time by using a decision model that is machine-learned to input an information based on the system message and output the wait time corresponding to the system message.

9. The interaction apparatus according to claim 8 ,

wherein the decision model is machine-learned to further input a profile information on an attribute of the user and output the wait time.

10. The interaction apparatus according to claim 9 ,

wherein the processing circuitry includes a classification model that is machine-learned to input the information based on the system message and output a classification information indicating whether or not the system message is a message which requests the user to answer, and

wherein the processing circuitry is configured to:

acquire the classification information by inputting the information based on the system message into the classification model; and

acquire and set the wait time by inputting the information based on the system message into the decision model when the acquired classification information indicates that the system message is the message which requests the user to answer.

11. A method, performed by processing circuitry of an interaction apparatus, comprising:

acquiring a system message for a user;

setting a wait time for receiving an answer from the user based on the system message;

measuring an elapsed time after the system message is output to the user;

acquiring a user utterance uttered by the user; and

controlling an interaction with the user so that an additional action corresponding to the system message is output to the user when the measured elapsed time exceeds the wait time,

wherein the method further includes determining whether or not to subtract the elapsed time, based on the system message and the user utterance, in response to the user utterance not corresponding to the answer to the system message is acquired after the system message is output.

Assignments (1)
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded Jan 10, 2022
From: TANAKA, GO; SHIRAMIZU, YUTARO
To: NTT DOCOMO, INC.
Reel/Frame 058609/0317 →
Priority Claims (1)
JP 2019-130279 · Jul 12, 2019 · national
Continuity (1)
Related Publication 20220254346A1 · Aug 11, 2022