| 摘要 |
<p>Even if the user makes an utterance containing a word satisfying a predetermined condition such as an unknown word, the fact is presented to the user so that the user can easily confirm the fact. A speech recognition system comprises a word speech recognizing section for converting an inputted speech into a recognition result word sequence by using a predetermined recognizing word dictionary, a syllable recognizing section for converting the inputted speech into a recognition result syllable sequence, a segment judging section for judging a segment meeting a predetermined condition to serve as a basis for presuming a word in the recognition result word sequence as an unknown word, and an output section for determining a partial syllable sequence in the recognition result syllable sequence corresponding to the judged segment and outputting both one or more word entries near the place where the partial syllable sequence is arranged in the recognition word dictionary where the word entries are arranged in order of definition in advance and the recognition result word sequence.</p> |
