Dialogue establishing utterances without content words

Patent metadata

Field	Value
Publication number	US-11232789-B2
Application number	US-201716301665-A
Country	US
Kind code	B2
Filing date	May 19, 2017
Priority date	May 20, 2016
Publication date	Jan 25, 2022
Grant date	Jan 25, 2022

Abstract

The present invention keeps a dialogue continuing for a long time without causing uncomfortable feeling to a user. A dialogue system 10 includes at least an input part 1 that receives a user utterance, which is an utterance from a user and a presentation part 5 that presents the utterance. The input part 1 receives a user utterance performed by the user. A presentation part 5 - 1 presents a dialogue-establishing utterance which does not include any content words. A presentation part 5 - 2 presents a second utterance associated with a generation target utterance, which is one or more utterances performed before user utterance that includes at least the user utterance, after the dialogue-establishing utterance.

First claim

What is claimed is:

1. A dialogue method carried out by a dialogue system that performs a dialogue with a user, the dialogue method comprising:
an utterance receiving step in which an input part receives a user utterance performed by the user,
a first presentation step in which a presentation part presents a dialogue-establishing utterance, wherein the dialogue-establishing utterance is presented to the user utterance, the dialogue-establishing utterance is a nodding, and no utterances are presented between the user utterance and the dialogue-establishing utterance; and
a second presentation step in which the presentation part presents, after the dialogue-establishing utterance, a second utterance which is an utterance determined with a predetermined rule that considers words included in a generation target utterance, wherein the generation target utterance is one of (i) the user utterance and (ii) the user utterance and one or more utterances performed before the user utterance, and no utterances are presented between the dialogue-establishing utterance and the second utterance,
wherein
the first presentation step and the second presentation step are executed where an utterance candidate generated in association with the generation target utterance does not meet a prescribed standard, wherein the prescribed standard is whether or not the utterance candidate has inappropriate contents as a reply to the user utterance,
the generation target utterance includes L words,
the utterance candidate is generated using M words included in the L words, and
the second utterance is generated using N words included in the L words, where N is an integer equal to or greater than 1, M and L are integers equal to or greater than 2, and N<M≤L.

2. A dialogue method carried out by a dialogue system that performs a dialogue with a user, the dialogue method comprising:
an utterance receiving step in which an input part receives a user utterance performed by the user;
a first presentation step in which a presentation part presents a dialogue-establishing utterance, wherein the dialogue-establishing utterance is presented to the user utterance, the dialogue-establishing utterance is a nodding, and no utterances are presented between the user utterance and the dialogue-establishing utterance; and
a second presentation step in which the presentation part presents, after the dialogue-establishing utterance, a second utterance which is an utterance determined with a predetermined rule that considers words included in a generation target utterance, wherein the generation target utterance is one of (i) the user utterance and (ii) the user utterance and one or more utterances performed before the user utterance, and no utterances are presented between the dialogue-establishing utterance and the second utterance,
wherein
the first presentation step and the second presentation step are executed where an utterance candidate generated in association with the generation target utterance does not meet a prescribed standard, wherein the prescribed standard is whether or not the utterance candidate has inappropriate contents as a reply to the user utterance,
the second utterance is the utterance candidate that does not meet the prescribed standard.

3. The dialogue method according to claim 1 or 2 , wherein
the dialogue system comprises a plurality of agents,
in the first presentation step, any one agent of the plurality of agents presents the dialogue-establishing utterance,
in the second presentation step, an agent different from the agent presenting the dialogue-establishing utterance of the plurality of agents presents the second utterance.

4. A dialogue system that performs a dialogue with a user, the dialogue system comprising:
an input part that receives a user utterance performed by the user;
an utterance determination part that determines a dialogue-establishing utterance and a second utterance which is an utterance determined with a predetermined rule that considers words included in a generation target utterance, wherein the dialogue-establishing utterance is presented to the user utterance, the dialogue-establishing utterance is a nodding, and the generation target utterance is one of (i) the user utterance and (ii) the user utterance and one or more utterances performed before the user utterance; and
a presentation part that presents the dialogue-establishing utterance determined by the utterance determination part and presents the second utterance determined by the utterance determination part immediately after presenting the dialogue-establishing utterance, wherein no utterances are presented both between the user utterance and the dialogue-establishing utterance and between the dialogue-establishing utterance and the second utterance
wherein
the presentation part presents the dialogue-establishing utterance and the second utterance where an utterance candidate generated in association with the generation target utterance does not meet a prescribed standard, wherein the prescribed standard is whether or not the utterance candidate has inappropriate contents as a reply to the user utterance,
the generation target utterance includes L words,
the utterance candidate is generated using M words included in the L words, and
the second utterance is generated using N words included in the L words, where N is an integer equal to or greater than 1, M and L are integers equal to or greater than 2, and N<M≤L.

5. A dialogue apparatus that determines an utterance presented by a dialogue system comprising at least an input part that receives a user utterance which is an utterance from a user and a presentation part that presents an utterance, the dialogue apparatus comprising processing circuitry configured to:
determine a dialogue-establishing utterance, wherein the dialogue-establishing utterance is presented to the user utterance, the dialogue-establishing utterance is a nodding, and no utterances are presented between the user utterance and the dialogue-establishing utterance; and
determine a second utterance which is an utterance presented after the dialogue-establishing utterance and is an utterance determined with a predetermined rule that considers words included in a generation target utterance, wherein the generation target utterance is one of (i) the user utterance and (ii) the user utterance and one or more utterances performed before the user utterance, and no utterances are presented between the dialogue-establishing utterance and the second utterance
wherein
the dialogue-establishing utterance and the second utterance are presented where an utterance candidate generated in association with the generation target utterance does not meet a prescribed standard, wherein the prescribed standard is whether or not the utterance candidate has inappropriate contents as a reply to the user utterance,
the generation target utterance includes L words,
the utterance candidate is generated using M words included in the L words, and
the second utterance is generated using N words included in the L words, where N is an integer equal to or greater than 1, M and L are integers equal to or greater than 2, and N<M≤L.

6. A non-transitory computer readable recording medium including computer executable instructions that make a dialogue system perform a dialogue method, wherein the dialogue system performs a dialogue with a user, the dialogue method comprising:
an utterance receiving step in which an input part receives a user utterance performed by the user;
a first presentation step in which a presentation part presents a dialogue-establishing utterance, wherein the dialogue-establishing utterance is presented to the user utterance, the dialogue-establishing utterance is a nodding, and no utterances are presen

Assignees

• Nippon Telegraph & Telephone
• Univ Osaka

Inventors

• Sugiyama Hiroaki
• Meguro Toyomi
• Yamato Junji
• Yoshikawa Yuichiro
• Ishiguro Hiroshi
• Iio Takamasa
• Arimoto Tsunehiro

Classifications

• G06F3/167 Primary

  Audio in a user interface, e.g. using voice commands for navigating, audio feedback · CPC title

• G10L13/00

  Speech synthesis; Text to speech systems · CPC title

• G10L15/00

  Speech recognition (G10L17/00 takes precedence) · CPC title

• G06F3/16

  Sound input; Sound output (speech processing G10L) · CPC title

• G10L2015/223

  Execution procedure of a spoken command · CPC title