CONVERSATION MEMBER OPTIMIZATION APPARATUS, CONVERSATION MEMBER OPTIMIZATION METHOD, AND PROGRAM

• US 20180075395A1
• Filed: 07/10/2017
• Published: 03/15/2018
• Est. Priority Date: 09/13/2016
• Status: Active Grant

• Current Assignee: Honda Motor Co., Ltd. (Honda Motor Company)
• Original Assignee: Honda Motor Co., Ltd. (Honda Motor Company)
• Inventors: Mizumoto, Takeshi, Hagiya, Shunichi
• Granted Patent: US 10,699,224 B2
• Time in Patent Office: Days
• CPC Class Codes: G06F 16/9024 Graphs; Linked lists G06F16/9027 takes precedence
• CPC Class Codes: G06F 17/16 Matrix or vector computation , e.g. matrix-matrix or matrix-vector multiplication, matrix factorization matrix transposition G06F7/78
• CPC Class Codes: G06Q 10/04 Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem" market predictions or forecasting for commercial activities G06Q30/0202
• CPC Class Codes: G06Q 10/063114 Status monitoring or status determination for a person or group
• CPC Class Codes: G06Q 10/06315 Needs-based resource requirements planning or analysis
• CPC Class Codes: G10L 15/14 using statistical models, e.g. Hidden Markov Models [HMMs] G10L15/18 takes precedence
• CPC Class Codes: G10L 15/222 Barge in, i.e. overridable guidance for interrupting prompts

Abstract

A conversation member optimization apparatus includes: a conversation analyzing unit configured to recognize sound signals in a plurality of conferences held in the past for each conference and analyze conversations in the conferences for each conference; an estimating unit configured to estimate feature quantities in combinations of the members who participate in the conferences to be held on the basis of results analyzed by the conversation analyzing unit and in accordance with a member selecting instruction used to select members who participate in conferences to be held; and a member selecting unit configured to select the members on the basis of the results estimated by the estimating unit.

Claims

1. A conversation member optimization apparatus comprising:

a conversation analyzing unit configured to recognize sound signals in a plurality of conferences held in the past for each conference and analyze conversations in the conferences for each conference;

an estimating unit configured to estimate feature quantities in combinations of members who will participate in a conference to be held on the basis of results analyzed by the conversation analyzing unit and in accordance with a member selecting instruction used to select members to participate in the conference to be held; and

a member selecting unit configured to select the members on the basis of the results estimated by the estimating unit.

2. The conversation member optimization apparatus according to claim 1, wherein types indicating dominance relationships between the members who participate in the conferences are stored, and

the member selecting unit selects the members to match desired types.

3. The conversation member optimization apparatus according to claim 1, wherein the feature quantities include at least one of

an amount of conditional utterances serving as an amount of utterances according to a combination of one member and another member,

an interrupt matrix indicating whether another member interrupts and one member speaks, and

activity probability serving as probability of another member speaking after one member.

4. The conversation member optimization apparatus according to claim 1, wherein the member selecting instruction includes

the total number of participating members in the conference to be held and determination information indicating at least one member who is determined to participate in advance among participating members in the conference to be held.

5. The conversation member optimization apparatus according to claim 3, wherein the member selecting instruction includes the total number of participating members in the conference to be held and determination information indicating at least one member who is determined to participate in advance among participating members in the conference to be held,

members other than the member who is determined to participate in advance are candidates for selectable members, and

the estimating unit repeatedly performs selection until the one member among the members who are the candidates is included in the total number of participating members in the conference to be held so that the amount of conditional utterances when participating members in the conference to be held are combined with members who are candidates to be selected are combined one by one is maximized.

6. The conversation member optimization apparatus according to claim 3, wherein the member selecting instruction includes the total number of participating members in the conference to be held and determination information indicating at least one member who is determined to participate in advance among participating members in the conference to be held,

members other than the member who is determined to participate in advance are candidates for selectable members, and

the estimating unit repeatedly performs selection until the one member among the members who are the candidates is included in the total number of participating members in the conference to be held so that the amounts of conditional utterances when participating members in the conference to be held are combined with members who are candidates to be selected are combined one by one and sums of values obtained by converting the interrupt matrix into scalar values and values obtained by converting the activity probability into scalar values are maximized.

7. The conversation member optimization apparatus according to claim 3, wherein the member selecting instruction includes the total number of participating members in the conference to be held and determination information indicating at least one member who is determined to participate in advance among participating members in the conference to be held,

members other than the member who is determined to participate in advance are candidates for selectable members,

the estimating unit selects a member for whom the amounts of conditional utterances when participating members in the conference to be held are combined with members who are candidates to be selected are combined one by one are a first threshold value or more,

selects a member for whom the values obtained by converting the interrupt matrix into the scalar values when participating members in the conference to be held are combined with members who are candidates to be selected are combined one by one are a second threshold value or more, and

selects a member for whom the values obtained by converting the activity probability into the scalar values when participating members in the conference to be held are combined with members who are candidates to be selected are combined one by one are a third threshold value or more, and

wherein the member selecting unit repeatedly performs selection until the one member among the members who are the candidates is included in the total number of participating members in the conference to be held on the basis of the results selected by the estimating unit.

8. The conversation member optimization apparatus according to claim 6, wherein the estimating unit selects a function used for converting the interrupt matrix into scalar values in accordance with selection conditions included in the member selecting instruction when the interrupt matrix is converted into the scalar values.

9. A conversation member optimization method comprising:

a conversation analysis step of recognizing, by a conversation analyzing unit, sound signals in a plurality of conferences held in the past for each conference and analyzing conversations in the conferences for each conference;

an estimation step of estimating, by an estimating unit, feature quantities in combinations of members who will participate in a conference to be held on the basis of results analyzed in the conversation analysis step and in accordance with a member selecting instruction used to select members to participate in the conference to be held; and

a member selection step of selecting, by a member selecting unit, the member on the basis of the results estimated in the estimation step.

10. A program causing a computer to execute:

a conversation analysis step of recognizing sound signals in a plurality of conferences held in the past for each conference and analyzing conversations in the conferences for each conference;

an estimation step of estimating feature quantities in combinations of members who will participate in a conference to be held on the basis of results analyzed in the conversation analysis step and in accordance with a member selecting instruction used to select members to participate in the conference to be held; and

a member selection step of selecting the member on the basis of the results estimated in the estimation step.