Speech recognition apparatus and speech recognition system

1. A speech recognition apparatus comprising:

a speech detection unit configured to detect a speech input by a user;

an information providing unit configured to perform information provision to the user, using either first speech recognition information based on a recognition result of the speech by a first speech recognition unit or second speech recognition information based on a recognition result of the speech by a second speech recognition unit different from the first speech recognition unit; and

a selection unit configured to:

calculate a first user satisfaction level indicating a predicted value of a degree of satisfaction of the user with the information provision of a case of using the first speech recognition information, and calculate a second user satisfaction level indicating a predicted value of a degree of satisfaction of the user with the information provision of a case of using the second speech recognition information, on the basis of an elapsed time from the input of the speech to: acquisition of the first speech recognition information by the selection unit, acquisition of the second speech recognition information by the selection unit, or present, and

compare the first user satisfaction level with the second user satisfaction level to obtain a comparison result, and

select either the first speech recognition information or the second speech recognition information as speech recognition information to be used by the information providing unit on the basis of the comparison result.

2. The speech recognition apparatus according to claim 1 , wherein,

in a case where the first speech recognition information has been acquired first and the second speech recognition information has not been acquired yet, the selection unit

measures a first elapsed time regarding an elapsed time from the input of the speech to the acquisition of the first speech recognition information, and predicts a second elapsed time regarding an elapsed time from the input of the speech to acquisition of the second speech recognition information,

calculates the first user satisfaction level on the basis of the measured first elapsed time,

calculates the second user satisfaction level on the basis of the predicted second elapsed time, and

compares the calculated first user satisfaction level with the calculated second user satisfaction level, and determines whether to select the first speech recognition information on the basis of a comparison result.

3. The speech recognition apparatus according to claim 1 , wherein,

in a case where the first speech recognition information has been already acquired and the second speech recognition information has not been acquired yet, the selection unit

measures a third elapsed time regarding an elapsed time from the input of the speech to present,

calculates the first user satisfaction level and the second user satisfaction level on the basis of the measured third elapsed time, and

compares the calculated first user satisfaction level with the calculated second user satisfaction level, and determines whether to select the first speech recognition information on the basis of a comparison result.

4. The speech recognition apparatus according to claim 1 , wherein,

in a case where the first speech recognition information has been acquired first and the second speech recognition information has been acquired second, the selection unit

measures a second elapsed time regarding an elapsed time from the input of the speech to the acquisition of the second speech recognition information,

calculates the first user satisfaction level and the second user satisfaction level on the basis of the measured second elapsed time, and

compares the calculated first user satisfaction level with the calculated second user satisfaction level, and determines whether to select either the first speech recognition information or the second speech recognition information on the basis of a comparison result.

5. The speech recognition apparatus according to claim 1 , wherein

the selection unit further calculates the first user satisfaction level and the second user satisfaction level on the basis of at least one of a first domain and a second domain respectively corresponding to the first speech recognition information and the second speech recognition information, of a plurality of domains determined in advance according to attributes of the speech, and a first estimated accuracy rate and a second estimated accuracy rate obtained respectively corresponding to the first speech recognition information and the second speech recognition information.

6. The speech recognition apparatus according to claim 5 , wherein

at least one of the first speech recognition unit and the second speech recognition unit recognizes the speech, using any one of a plurality of dictionary data, and

the selection unit estimates at least one of the first domain and the second domain on the basis of the dictionary data used by the at least one of the first speech recognition unit and the second speech recognition unit for the recognition of the speech.

7. The speech recognition apparatus according to claim 5 , wherein

at least one of the first speech recognition information and the second speech recognition information includes intention estimation information indicating an estimation result of an intention of the user with respect to the speech, and

the selection unit estimates at least one of the first domain and the second domain on the basis of the intention estimation information.

8. The speech recognition apparatus according to claim 5 , wherein

the selection unit estimates the first domain and the second domain on the basis of an estimation history of the past first domain and the past second domain.

9. The speech recognition apparatus according to claim 5 , wherein

the selection unit determines the first estimated accuracy rate and the second estimated accuracy rate on the basis of at least one of the first domain and the second domain, reliability with respect to the first speech recognition information and reliability with respect to the second speech recognition information, and the elapsed time from the input of the speech.

10. The speech recognition apparatus according to claim 1 , wherein

each of the first speech recognition information and the second speech recognition information includes intention estimation information indicating an estimation result of an intention of the user with respect to the speech, and

the selection unit selects the intention estimation information included in either the first speech recognition information or the second speech recognition information.

11. The speech recognition apparatus according to claim 1 , wherein

the selection unit selects any one of

an adoption operation to adopt an input operation based on either the first speech recognition information or the second speech recognition information as an input operation of the user,

a confirmation operation to adopt an input operation based on either the first speech recognition information or the second speech recognition information as an input operation of the user after confirmation of the user, and

a rejection operation to reject both the input operation based on the first speech recognition information and the input operation based on the second speech recognition information without adoption, and

changes the method of information provision according to the selected operation.

12. A speech recognition system including a terminal device and a server,

the terminal device comprising:

a speech detection unit configured to detect a speech input by a user;

a first speech recognition unit configured to execute speech recognition processing for recognizing the speech and output first speech recognition information based on a recognition result of the speech;

a first communication control unit configured to transmit speech information based on the speech to the server and receive second speech recognition information transmitted from the server;

an information providing unit configured to perform information provision to the user, using either the first speech recognition information or the second speech recognition information; and

a selection unit configured to:

calculate a first user satisfaction level indicating a predicted value of a degree of satisfaction of the user with the information provision of a case of using the first speech recognition information, and calculate a second user satisfaction level indicating a predicted value of a degree of satisfaction of the user with the information provision of a case of using the second speech recognition information, on the basis of an elapsed time from the input of the speech to: acquisition of the first speech recognition information by the selection unit, acquisition of the second speech recognition information by the selection unit, or present, and

compare the first user satisfaction level with the second user satisfaction level to obtain a comparison result, and

select either the first speech recognition information or the second speech recognition information to be used by the information providing unit on the basis of the comparison result, and

the server comprising:

a second communication control unit configured to receive the speech information transmitted from the terminal device and transmit the second speech recognition information to the terminal device; and

a second speech recognition unit configured to execute speech recognition processing for recognizing the speech on the basis of the speech information and output the second speech recognition information based on a recognition result of the speech.