Voice recognition apparatus

1. A voice recognition apparatus comprising:

a voice recognition unit that perfoms voice recognition;

a control unit that controls receipt of a voice input to said voice recognition unit according to a timeout time which defines an end of the receipt of a voice input, and for performing recognition according to a result of the voice recognition acquired by said voice recognition unit;

an environmental condition detecting unit that detects an environmental condition;

a timeout time control unit that changes said timeout time according to the environmental condition detected by said environmental condition detection unit; and

a receipt restart control unit that determines whether to restart the receipt of a voice input according to the environmental condition detected by the environmental condition detection unit after said receipt of a voice input times out.

2. The voice recognition apparatus according to claim 1 , wherein the timeout time includes a first timeout time which defines a time interval from a start of the receipt of a voice to a start of utterance, and a second timeout time which defines a time interval from said start of the receipt of a voice to an end of the utterance, and a third timeout time which defines a time interval succeeding said end of the utterance.

3. The voice recognition apparatus according to claim 1 , wherein the environmental condition detecting unit detects, as said environmental condition, a noise level in a surrounding environment, the timeout control unit reduces the timeout time when said noise level is equal to or higher than a predetermined level, and the restart control unit grants the restart of the receipt of a voice input when said noise level is lower than the predetermined level.

4. The voice recognition apparatus according to claim 1 , wherein the environmental condition detecting unit detects, as the environmental condition, at least one of current position information indicating a current position of a moving object and traveling state information indicating a traveling state of the moving object, the timeout control unit compares a preset criterion with said environmental condition and changes the timeout time, and the restart control unit compares said criterion with said environmental condition and determines whether to grant the restart of the receipt of a voice input.

5. The voice recognition apparatus according to claim 4 , wherein the current position information indicates the current position of the moving object, and includes a type of a road along which said moving object is traveling and a type of an area in which said moving object is traveling.

6. The voice recognition apparatus according to claim 4 , wherein at least a traveling velocity of the moving object and a gear position in a transmission of the moving object are included in the traveling state information.

7. The voice recognition apparatus according to claim 1 , wherein the environmental condition detecting unit has a degree-of-difficulty-in-dialog table in which dialog contents and degrees of difficulty which are defined for the dialog contents respectively are included, said environmental condition detecting unit acquires a degree of difficulty corresponding to said dialog content on a basis of said degree-of-difficulty-in-dialog table, the timeout control unit changes the timeout time according to the degree of difficulty acquired by said environmental condition detecting unit, and the restart control unit determines whether to grant the restart of the receipt of a voice input according to said degree of difficulty.

8. A voice recognition method comprising:

performing voice recognition;

controlling receipt of a voice input according to a timeout time which defines an end of the receipt of a voice input, and for performing recognition according to a result of the voice recognition acquired by said performing voice recognition step;

detecting, by utilizing an environmental condition detecting unit, an environmental condition;

changing said timeout time according to the environmental condition; and

determining whether to restart the receipt of a voice input according to the environmental condition detected by the environmental condition detecting step after said receipt of a voice input times out.

9. The voice recognition method according to claim 8 , wherein the timeout time includes a first timeout time which defines a time interval from a start of the receipt of a voice to a start of utterance, and a second timeout time which defines a time interval from said start of the receipt of a voice to an end of the utterance, and a third timeout time which defines a time interval succeeding said end of the utterance.

10. The voice recognition method according to claim 8 , further comprising:

detecting, as said environmental condition, a noise level in a surrounding environment;

reducing the timeout time when said noise level is equal to or higher than a predetermined level; and

granting the restart of the receipt of a voice input when said noise level is lower than the predetermined level.

11. The voice recognition method according to claim 8 , further comprising:

detecting, as the environmental condition, at least one of current position information indicating a current position of a moving object and traveling state information indicating a traveling state of the moving object;

comparing a preset criterion with said environmental condition and changing the timeout time;

comparing said criterion with said environmental condition; and

determining whether to grant the restart of the receipt of a voice input.

12. The voice recognition method according to claim 11 , wherein the current position information indicates the current position of the moving object, and includes a type of a road along which said moving object is traveling and a type of an area in which said moving object is traveling.

13. The voice recognition method according to claim 11 , wherein at least a traveling velocity of the moving object and a gear position in a transmission of the moving object are included in the traveling state information.

14. The voice recognition method according to claim 8 , wherein the environmental condition detecting unit has a degree-of-difficulty-in-dialog table in which dialog contents and degrees of difficulty which are defined for the dialog contents respectively are included, said method further comprising:

acquiring a degree of difficulty corresponding to said dialog content on a basis of said degree-of-difficulty-in-dialog table;

changing the timeout time according to the degree of difficulty acquired by said acquiring step; and

determining whether to grant the restart of the receipt of a voice input according to said degree of difficulty.