Speech recognition device, speech recognition method, non-transitory recording medium, and robot

A robot comprising:

• one or more controllable moving mechanism;
• and
• a processor configured to:
• determine a distance to a speech source;
• specify one or more of a plurality of registered expressions to be recognizable based on the distance determined,
• wherein the one or more of the registered expressions specified are associated with a predetermined distance range;
• attempt to recognize a speech signal based on the one or more of the plurality of registered expressions specified as recognizable;
• and
• in response to failing to recognize the speech signal, control the one or more controllable moving mechanism to move the robot based on the distance determined and the predetermined distance range.

A speech recognition method comprising:

• determining a distance to a speech source;
• specifying one or more of a plurality of registered expressions to be recognizable based on the distance determined,
• wherein specifying the one or more of the plurality of registered expressions to be recognizable based on the distance determined comprises:
• maintaining a first registered expression to be recognizable;
• and
• switching a second registered expression from a recognizable registered expression into an unrecognizable registered expression based on the distance determined, wherein the second registered expression is acoustically similar to the first registered expression;
• and
• recognizing a speech signal based on the one or more of the plurality of registered expressions specified as recognizable.

A non-transitory computer-readable recording medium having stored thereon a program executable by a computer, the program controlling the computer to perform functions comprising:

• determining a distance to a speech source;
• specifying one or more of a plurality of registered expressions to be recognizable based on the distance determined,
• wherein specifying the one or more of the plurality of registered expressions to be recognizable based on the distance determined comprises:
• maintaining a first registered expression to be recognizable;
• and
• switching a second registered expression from a recognizable registered expression into an unrecognizable registered expression based on the distance determined, wherein the second registered expression is acoustically similar to the first registered expression;
• and
• recognizing a speech signal based on the one or more of the plurality of registered expressions specified as recognizable.

A speech recognition device comprising:

• a processor configured to:
• determine a distance to a speech source;
• specify one or more of a plurality of registered expressions to be recognizable based on the distance determined;
• provide a weighting parameter to each of the one or more of the plurality of registered expressions specified to be recognizable based on the distance determined, wherein the weighting parameter indicates an easiness of recognition of the each of the one or more of the plurality of registered expressions specified to be recognizable;
• and
• recognize a speech signal based on the weighting parameter provided to the each of the one or more of the plurality of registered expressions specified as recognizable.

A speech recognition method comprising:

• determining a distance to a speech source;
• specifying one or more of a plurality of registered expressions to be recognizable based on the distance determined;
• providing a weighting parameter to each of the one or more of the plurality of registered expressions specified to be recognizable based on the distance determined, wherein the weighting parameter indicates an easiness of recognition of the each of the one or more of the plurality of registered expressions specified to be recognizable;
• and
• recognizing a speech signal based on the weighting parameter provided to the each of the one or more of the plurality of registered expressions specified as recognizable.

A non-transitory computer-readable recording medium having stored thereon a program executable by a computer, the program controlling the computer to perform functions comprising:

• determining a distance to a speech source;
• specifying one or more of a plurality of registered expressions to be recognizable based on the distance determined;
• providing a weighting parameter to each of the one or more of the plurality of registered expressions specified to be recognizable based on the distance determined, wherein the weighting parameter indicates an easiness of recognition of the each of the one or more of the plurality of registered expressions specified to be recognizable;
• and
• recognizing a speech signal based on the weighting parameter provided to the each of the one or more of the plurality of registered expressions specified as recognizable.