Speech translation apparatus, method and program

A speech translation apparatus comprising:

• an acquisition unit configured to acquire speech in a first language as a speech signal;
• a display configured to display a first language word string which is a result of speech recognition successively performed for the speech signals, and a second language word string which is a translation of the first language word string;
• a detection unit configured to detect a location on the display indicated by a user;
• a selection unit configured to select, in accordance with the location, at least one of the first language word string and the second language word string;
• and
• a presentation unit configured to present one or more similar examples which are examples in the first language and similar to the first language word string and one or more translation examples which are translations of the similar examples in the second language,
• wherein the display further displays the presented similar examples and translation examples.

A speech translation method comprising:

• acquiring speech in a first language as a speech signal;
• performing speech recognition on the speech signal to obtain a first language word string which is a result of the speech recognition successively;
• translating the first language word string into a second language to obtain a second language word string which is a result of translation;
• searching for at least one similar example for each first language word string, and, if there is the similar example, acquiring the similar example and a translation example which is a result of the translation of the similar example in the second language, the similar example indicating a word string that is similar to the first language word string in the first language;
• selecting, in accordance with a user instruction, at least one of the first language word string associated with the similar example and the second language word string associated with the translation example, as a selected word string;
• and
• presenting one or more similar examples and one or more translation examples associated with the selected word string.

A non-transitory computer readable medium including computer executable instructions, wherein the instructions, when executed by a processor, cause the processor to perform a method comprising:

• acquiring speech in a first language as a speech signal;
• performing speech recognition on the speech signal to obtain a first language word string which is a result of the speech recognition successively;
• translating the first language word string into a second language to obtain a second language word string which is a result of translation;
• searching for at least one similar example for each first language word string, and, if there is the similar example, acquiring the similar example and a translation example which is a result of the translation of the similar example in the second language, the similar example indicating a word string that is similar to the first language word string in the first language;
• selecting, in accordance with a user instruction, at least one of the first language word string associated with the similar example and the second language word string associated with the translation example, as a selected word string;
• and
• presenting one or more similar examples and one or more translation examples associated with the selected word string.