摘要 |
<P>PROBLEM TO BE SOLVED: To allow for determining conversion candidates for a character string, that has been recognized in error by speech recognition processing, on a basis other than the speech recognition process. <P>SOLUTION: A speech recognition device includes; a speech data acquisition unit for acquiring speech data; a speech recognition unit which performs speech recognition processing on the speech data in reference to an acoustic model and a first language model and generates a recognized character string representing a recognition result; a conversion target character string designation unit which designates some of the character strings included in the recognized character string as conversion target character strings; a reference character string designation unit which designates the character strings preceding or following the designated conversion target character strings in the recognized character string as reference character strings; a conversion candidate character string designation unit which designates the character strings having conjunctive relations with the designated reference character strings as conversion candidate character strings for converting the conversion target character strings in reference to a second language model; and an output unit for outputting the designated conversion candidate character strings. <P>COPYRIGHT: (C)2013,JPO&INPIT |