SPEECH INTERACTIVE APPARATUS AND COMPUTER PROGRAM PRODUCT

• US 20120179473A1
• Filed: 03/19/2012
• Published: 07/12/2012
• Est. Priority Date: 09/25/2009
• Status: Active Grant

• Current Assignee: Kabushiki Kaisha Toshiba (Toshiba Corporation), Toshiba Digital Solutions Corporation (Toshiba Corporation)
• Original Assignee: Kabushiki Kaisha Toshiba (Toshiba Corporation)
• Inventors: Yano, Takehide
• Granted Patent: US 8,571,874 B2
• Time in Patent Office: Days
• US Class Current: 704/275
• CPC Class Codes: G10L 15/22 Procedures used during a speech recognition process, e.g. man-machine dialogue

Abstract

According to an embodiment, a speech interactive apparatus includes an output unit to output a first response; a receiving unit to receive a start instruction of a speech input as a reply to the first response; a response control unit to stop the output of the first response when the start instruction is received while the first response is being output; and a deciding unit to decide on a first determination period, which is used in determining whether a silent state has occurred, based on whether the start instruction is received while the first response is being output or based on the timing of receiving the start instruction. When the input speech is not input during a period starting from the reception of the start instruction till an elapse of the first determination period, the response control unit instructs the output unit to output the first response again.

Claims

1. A speech interactive apparatus comprising:

an output unit configured to output a first response which needs to be notified to a user;

a receiving unit configured to receive a start instruction of a speech input from the user as a reply to the first response;

a speech recognizing unit configured to perform speech recognition of an input speech that is input by the user;

an interactive unit configured to decide on a second response, which is determined according to a speech recognition result of the input speech, based on interaction information stored in a storing unit and instruct the output unit to output the second response;

a response control unit configured to stop the output of the first response when the start instruction is received while the first response is being output; and

a deciding unit configured to decide on a first determination period, which is used in determining whether or not a silent state has occurred, based on whether or not the start instruction is received while the first response is being output or based on the timing of receiving the start instruction, wherein

when the input speech is not input during a period starting from the reception of the start instruction till an elapse of the first determination period, the response control unit determines that the silent state has occurred and instructs the output unit to output the first response again.

2. The apparatus according to claim 1, wherein, shorter a period from a time of starting the output of the first response up to the time of receiving the start instruction, the deciding unit decides on the first determination time to be of a shorter duration.

3. The apparatus according to claim 2, further comprising a supplementary information control unit configured to, when the input speech is not input during a period starting from the timing of receiving the start instruction till the elapse of the first determination period, instruct the output unit to output supplementary information that contains information regarding the input speech that is input as the reply to the first response.

4. The apparatus according to claim 3, further comprising an interactive unit configured to decide on a response, which is determined according to the speech recognition result of the input speech, based on interaction information stored in a storing unit and instruct the output unit to output the response, wherein

when the input speech is not input during a period starting from the reception of the start instruction till an elapse of the first determination period that is set to be equal to or longer than a predetermined threshold value, the supplementary information control unit instructs the output unit to output supplementary information that contains information regarding the input speech that is input as a reply to the second response that is determined based on the interaction information.

5. The apparatus according to claim 3, wherein, longer the first determination period, the supplementary information control unit instructs the output unit to output the supplementary information at a later timing.

6. The apparatus according to claim 5, wherein, when the first determination period is shorter than a predetermined threshold value and when the input speech is not input during a period starting from the start of outputting the first response till the elapse of the first determination period, the supplementary information control unit instructs the output unit to instruct the supplementary information.

7. The apparatus according to claim 3, further comprising a supplementary information control unit configured to, when the input speech is not input during a period starting from the start of outputting the first response till the elapse of a second determination period that is shorter than the first determination period, instruct the output unit to output supplementary information that contains information regarding the input speech that is input as the reply to the first response.

8. A computer program product comprising a computer readable medium including programmed instructions, wherein the instructions, when executed by a computer, cause the computer to execute:

outputting a first response which needs to be notified to a user;

receiving a start instruction of a speech input from the user as a reply to the first response;

performing speech recognition of an input speech that is input by the user;

deciding on a second response, which is determined according to a speech recognition result of the input speech, based on interaction information stored in a storing unit and instructing an output unit to output the second response;

stopping the output of the first response when the start instruction is received while the first response is being output; and

deciding on a first determination period, which is used in determining whether or not a silent state has occurred, based on whether or not the start instruction is received while the first response is being output or based on the timing of receiving the start instruction; and

determining, when the input speech is not input during a period starting from the reception of the start instruction till an elapse of the first determination period, that the silent state has occurred and instructing the output unit to output the first response again.