This　paper　proposes　a　new　approach　for　generating　realistic　three-dimensional　speech　animation.　The　basic　idea　is　to　synthesize　the　animated　faces　using　prosodic　information　edited　by　user　with　a　kind　of　text　markup　language.　By　capturing　characteristic　trajectories　of　utterances　from　video　clips,　our　technique　builds　up　a　parametric　model　based　on　the　exponential　formula.　Based　on　this　formula　the　static　viseme　is　extended　to　dynamic　one.　To　relate　the　prosody　text　with　the　3D　animation,　the　input　attribute　is　mapped　to　be　the　value　of　formula　parameter.　Experimental　results　show　that　the　proposed　technique　synthesizes　animation　of　different　effects　depending　on　the　availability　with　the　prosodic　information.