Method, apparatus, and computer program product for machine translation

An apparatus for machine-translating a first language into a second language, comprising:

• an example storing unit that stores therein a first-language example in association with a second-language example that is a translation of the first-language example;
• a receiving unit that receives a first-language speech;
• a first recognizing unit that recognizes the first-language speech as a first-language character string, and outputs a first recognition result;
• a second recognizing unit that recognizes the first-language speech as a most probable first-language example from among first-language examples stored in the example storing unit, calculates a first likelihood indicating a certainty of the most probable first-language example, and outputs a second recognition result including calculated first likelihood;
• and
• a retrieving unit that retrieves, when calculated first likelihood exceeds a first threshold, a second-language example corresponding to the second recognition result from the example storing unit.

A method of machine-translating a first language into a second language, comprising:

• outputting a first recognition result by recognizing a input first-language speech as a first-language character string;
• outputting a second recognition result by recognizing the input first-language speech as a most probable first-language example from among first-language examples stored in an example storing unit that stores therein a first-language example in association with a second-language example that is a translation of the first-language example;
• calculating a similarity between the first recognition result and the second recognition result;
• and
• retrieving, when calculated similarity exceeds a predetermined threshold, a second-language example corresponding to the second recognition result from the example storing unit.

A computer program product comprising a computer-usable medium having computer-readable program codes embodied in the medium that when executed cause a computer to execute:

• outputting a first recognition result by recognizing a input first-language speech as a first-language character string;
• outputting a second recognition result by recognizing the input first-language speech as a most probable first-language example from among first-language examples stored in an example storing unit that stores therein a first-language example in association with a second-language example that is a translation of the first-language example;
• calculating a similarity between the first recognition result and the second recognition result;
• and
• retrieving, when calculated similarity exceeds a predetermined threshold, a second-language example corresponding to the second recognition result from the example storing unit.