DIALOG DETECTING APPARATUS, DIALOG DETECTING METHOD, AND COMPUTER PROGRAM PRODUCT

• US 20080243494A1
• Filed: 03/11/2008
• Published: 10/02/2008
• Est. Priority Date: 03/28/2007
• Status: Active Grant

• Current Assignee: Kabushiki Kaisha Toshiba (Toshiba Corporation), Toshiba Digital Solutions Corporation (Toshiba Corporation)
• Original Assignee: Kabushiki Kaisha Toshiba (Toshiba Corporation)
• Inventors: Okamoto, Masayuki, Hattori, Masanori, Tsuboi, Sogo, Cho, Kenta, Umeki, Hideo, Iketani, Naoki, Nishimura, Keisuke
• Granted Patent: US 8,306,823 B2
• Time in Patent Office: Days
• US Class Current: 704/214
• CPC Class Codes: G10L 15/22 Procedures used during a speech recognition process, e.g. man-machine dialogue

Abstract

A speech receiving unit receives a user ID, a speech obtained at a terminal, and an utterance duration, from the terminal. A proximity determining unit calculates a correlation value expressing a correlation between speeches received from plural terminals, compares the correlation value with a first threshold value, and determines that the plural terminals that receive the speeches whose correlation value is calculated are close to each other, when the correlation value is larger than the first threshold value. A dialog detecting unit determines whether a relationship between the utterance durations received from the plural terminals that are determined to be close to each other within an arbitrarily target period during which a dialog is to be detected fits a rule. When the relationship is determined to fit the rule, the dialog detecting unit detects dialog information containing the target period and the user ID.

Claims

1. A dialog detecting apparatus capable of connecting to a plurality of terminals and capable of obtaining a speech of a user, comprising:

a speech receiving unit that receives the speeches from the plurality of the terminals, each speech accompanying with a terminal ID identifying one of the terminals and an utterance duration of one of the speeches;

a proximity determining unit that calculates a correlation value expressing a correlation between speeches received by the plurality of terminals, compares the correlation value with a predetermined first threshold value, and determines that the plurality of terminals which receive a plurality of speeches whose correlation value is calculated are close to each other, when the correlation value is larger than the first threshold value; and

a dialog detecting unit that determines whether a relationship between utterance durations fits a predetermined rule, the utterance durations being received from the plurality of terminals that are determined to be close to each other in an arbitrarily target period, and detects dialog information containing the target period during which the relationship is determined to fit the rule and the terminal IDs received from the plurality of terminals that are determined to be close to each other.

2. The apparatus according to claim 1, wherein the dialog detecting unit determines whether a rate of an overlapping period during which utterance durations received from the plurality of terminals are overlapped to the target period fits the rule expressing that the rate is smaller than a predetermined second threshold value.

3. The apparatus according to claim 2, wherein the dialog detecting unit determines whether the rate of the overlapping period to the target period is smaller than the second threshold value and whether a rate of a non-utterance duration during which the utterance duration is not present within the target period to the target period fits the rule expressing that the rate is smaller than a predetermined third threshold value.

4. The apparatus according to claim 1, further comprising a schedule receiving unit that receives an input of schedule information containing a schedule of an action of a user and the terminal ID of the terminal that the user relevant to the action uses, and

the proximity determining unit calculates the correlation value between speeches corresponding to the terminal ID contained in the received schedule information within the speeches received from the plurality of terminals.

5. The apparatus according to claim 1, further comprising a speech storage unit that stores the terminal ID, the speech, and the utterance duration that are received by the speech receiving unit in association with each other, wherein

the proximity determining unit calculates a correlation value between the speeches stored in the speech storage unit, compares the calculated correlation value with the first threshold value, and determines that the plurality of terminals that receive speeches whose correlation value is calculated are close to each other, when the correlation value is larger than the first threshold value, and

the dialog detecting unit obtains from the speech storage unit utterance durations received from the plurality of terminals that are determined to be close to each other, and determines whether a relationship of the utterance durations obtained during the target period fits the rule.

6. The apparatus according to claim 1, further comprising a dialog storage unit that stores the dialog information detected by the dialog detecting unit.

7. The apparatus according to claim 6, further comprising an operation receiving unit that receives operation information containing the operation content to the terminal and an operation time and date in association with each other, wherein

the dialog storage unit stores the dialog information during the target period containing the operation time and date of the received operation information, and the operation content of the received operation information in association with each other.

8. The apparatus according to claim 7, wherein the operation receiving unit receives the operation information associated with a data ID that identifies data used at the terminal as the operation content, and

the dialog storage unit stores the dialog information during the target period containing the operation time and date of the received operation information, and the data ID of the received operation information in association with each other.

9. The apparatus according to claim 1, further comprising:

a speech receiving unit that receives an input of a speech of a user; and

a user information receiving unit that receives the terminal ID, wherein

the proximity determining unit calculates the correlation value between the speech input from the speech receiving unit and the speech received from the terminal, compares the correlation value with the first threshold value, and determines that the terminal which receives the speech and whose correlation value is calculated and the dialog detecting apparatus are close to each other, when the correlation value is larger than the first threshold value, and

the dialog detecting unit determines whether a relationship between the utterance duration that is received from the terminal determined to be close to each other and the utterance duration of speech that is input from the speech receiving unit during the target period fits the rule, and detects the dialog information containing the target period when the relationship is determined to fit the rule, a terminal ID received from the terminal determined to be close to each other, and a terminal ID received by the user-information receiving unit.

10. A dialog detecting method performed in a dialog detecting apparatus capable of connecting to a plurality of terminals and capable of obtaining a speech of a user, comprising:

receiving the speeches from the plurality of the terminals, each speech accompanying with a terminal ID identifying one of the terminals, and an utterance duration of one of the speeches;

calculating a correlation value expressing a correlation between speeches received by the plurality of terminals;

comparing the correlation value with a predetermined first threshold value;

determining that the plurality of terminals which receive a plurality of speeches whose correlation value is calculated are close to each other, when the correlation value is larger than the first threshold value;

determining whether a relationship between utterance durations fits a predetermined rule, the utterance durations being received from the plurality of terminals that are determined to be close to each other in an arbitrarily target period; and

detecting dialog information containing the target period during which the relationship is determined to fit the rule and the terminal identifications received from the plurality of terminals that are determined to be close to each other.

11. A computer program product having a computer readable medium including programmed instructions for detecting a dialog by connecting to a plurality of terminals capable of obtaining a speech of a user, wherein the instructions, when executed by a computer, cause the computer to perform:

receiving the speeches from the plurality of the terminals, each speech accompanying with a terminal ID identifying one of the terminals, and an utterance duration of one of the speeches;

calculating a correlation value expressing a correlation between speeches received by the plurality of terminals;

comparing the correlation value with a predetermined first threshold value;

determining that the plurality of terminals which receive a plurality of speeches whose correlation value is calculated are close to each other, when the correlation value is larger than the first threshold value;

determining whether a relationship between utterance durations fits a predetermined rule, the utterance durations being received from the plurality of terminals that are determined to be close to each other in an arbitrarily target period; and

detecting dialog information containing the target period during which the relationship is determined to fit the rule and the terminal identifications received from the plurality of terminals that are determined to be close to each other.