| 摘要 |
PROBLEM TO BE SOLVED: To permit high-precision voice recognition by translating a morpheme sequence of a 1st language outputted by a voice recognition part into a morpheme sequence of a 2nd language, converting the translated morpheme sequence of the 2nd language into a voice signal, and outputting it. SOLUTION: A translation device 101 receives the voice that a user speaks through a microphone 103, recognizes the received voice by a computer in a housing part 105, and displays the result of the voice recognition on a liquid crystal display 106. Then the computer in the housing part 105 translates the morpheme sequence of the 1st language outputted and displayed through the voice recognition into the morpheme sequence of the 2nd language, converts the translation result into the voice signal, and outputs it to a telephone line 108. The telephone line 108 is connected to a speech-partner's telephone set, etc., through exchanges and the speech partner is able to hear the voice signal sent through the telephone line 108 as a voice. |
