Information processor, information processing method, and program

Patent metadata

Field	Value
Publication number	US-2020105265-A1
Application number	US-201816620992-A
Country	US
Kind code	A1
Filing date	Jun 7, 2018
Priority date	Jun 20, 2017
Publication date	Apr 2, 2020
Grant date	—

Abstract

The present technology relates to an information processor, an information processing method, and a program that allow information to be presented using a character that allows a user to feel attached. An information processor of the present technology acquires a context, and selects, on the basis of the context, a predetermined character to be used for presenting information to the user from among a plurality of characters for which attributes defining different features are set. The attribute set for the selected predetermined character is merged with the attribute set for the main character, and information is presented to the user using the main character having a feature defined by the merged attribute. The present technology can be applied to an agent device operated by a voice UI.

First claim

1 . An information processor comprising:
an acquisition portion that acquires a context; and a character selection portion that selects, on a basis of the context, a predetermined character to be used for presenting information to a user from among a plurality of characters for which attributes defining different features are set.

2 . The information processor according to claim 1 further comprising:
a management portion that manages a main character selected as the main character; and
an output controller that presents information to the user using the main character.

3 . The information processor according to claim 2 , wherein
the character selection portion merges the attribute set for the predetermined character with the attribute set for the main character, and the output controller presents information to the user using the main character having a feature defined by the merged attribute.

4 . The information processor according to claim 3 , wherein
the character selection portion merges a plurality of types of the attributes set for the main character and the predetermined character by assigning predetermined weights to the same types of attributes.

5 . The information processor according to claim 4 , wherein
the feature of each of the plurality of characters and the main character is defined by the attribute of at least any one of speech expression, speech content, tone, or appearance.

6 . The information processor according to claim 5 , wherein
the output controller presents information to the user on a basis of information regarding a rule of the speech content or on a basis of model information of the speech content obtained by performing learning based on a dialogue corpus.

7 . The information processor according to claim 1 , wherein
the acquisition portion analyzes text data input by the user or speech voice data of the user, and acquires the context.

8 . The information processor according to claim 1 , wherein
the character selection portion selects the predetermined character from among the plurality of characters managed by an external device.

9 . The information processor according to claim 1 further comprising a character management portion that manages the plurality of characters, wherein
the character selection portion selects the predetermined character from among the plurality of characters managed by the character management portion.

10 . The information processor according to claim 1 , wherein
the acquisition portion acquires a content selected by the user as the context, and the character selection portion selects the predetermined character according to the content selected by the user.

11 . The information processor according to claim 1 , wherein
the acquisition portion acquires a content to be presented to the user as the context, and the character selection portion selects the predetermined character according to the content to be presented to the user.

12 . The information processor according to claim 1 , wherein
the acquisition portion acquires a profile of the user as the context, and the character selection portion selects the predetermined character according to the profile of the user.

13 . The information processor according to claim 1 , wherein
the acquisition portion acquires a service to be used by the user as the context, and the character selection portion selects the predetermined character according to the service.

14 . The information processor according to claim 1 , wherein
the acquisition portion acquires, as the context, an emotion of the user recognized on a basis of at least any one of an analysis result of a face image of the user, an analysis result of a speech voice of the user, or an analysis result of a speech content of the user, and the character selection portion selects the predetermined character according to the emotion of the user.

15 . The information processor according to claim 3 further comprising a preference learning portion that learns a preference of the user with respect to the main character having a feature defined by the merged attribute, wherein
the management portion adjusts the attribute set for the main character on a basis of a learning result of the preference of the user.

16 . The information processor according to claim 3 , wherein
the character selection portion selects a plurality of characters as the predetermined character, and merges the attribute set for each of the plurality of selected predetermined characters with the attribute set for the main character.

17 . An information processing method comprising the steps of:
acquiring a context; and selecting, on a basis of the context, a predetermined character to be used for presenting information to a user from among a plurality of characters for which attributes defining different features are set.

18 . A program that causes a computer to perform processing comprising the steps of:
acquiring a context; and selecting, on a basis of the context, a predetermined character to be used for presenting information to a user from among a plurality of characters for which attributes defining different features are set.

Assignees

• Sony Corp

Inventors

• Kuriya Shinobu
• Tsuruta Yasushi
• Nishikawa Kana
• Ogawa Hiroaki
• Ohmura Junki
• Touyama Keisuke

Classifications

• G10L13/10

  Prosody rules derived from text; Stress or intonation · CPC title

• G10L15/1815

  Semantic context, e.g. disambiguation of the recognition hypotheses based on word meaning · CPC title

• G10L25/63

  for estimating an emotional state · CPC title

• G10L2015/223

  Execution procedure of a spoken command · CPC title

• G10L25/60

  for measuring the quality of voice signals · CPC title