作者: Yamada Keiichi
DOI:
关键词:
摘要: PROBLEM TO BE SOLVED: To attain highly reliable voice recognition based on a rhythm feature of input by effectively detecting pitch frequency, even when influence noise etc. is large, and also whisper low voice. SOLUTION: Input from an section 11 sent to acoustic analysis 12, amount required for identification extracted 13. In the 13, processing performed unknown data using parameter storing 14. conversion frequency characteristic in 21, obtained correlation change calculation 22, calculated mutual matrix output COPYRIGHT: (C)2008,JPO&INPIT