Voice recognition device, speech recognition method and program

    公开(公告)号:JP4296290B2

    公开(公告)日:2009-07-15

    申请号:JP2003361646

    申请日:2003-10-22

    Abstract: PROBLEM TO BE SOLVED: To improve an input rate of speech uttered separately for each syllable. SOLUTION: When reliability of a recognition result is high, a display device displays a character image of the single syllable as it is, and when the reliability of the recognition result is somewhat low, the display device displays the character image of the vowel, and also displays a predetermined image "?", at the side, representing consonant recognition failure. Further, When the reliability of the recognition result is low, a predetermined picture "*" representing to prompt inputting the next syllable is displayed. On the other hand, a loudspeaker outputs synthesized speech of the single syllable when the reliability of the recognition result is high, but outputs only synthesized speech of the vowel when the reliability of the recognition result is somewhat low. Further, the loudspeaker outputs synthesized speech "yes" representing to prompt inputting the next syllable when the reliability of the recognition result is low. COPYRIGHT: (C)2005,JPO&NCIPI

Patent Agency Ranking