《語(yǔ)音線性預(yù)測(cè)編碼》ppt課件-文庫(kù)吧
2025-04-18 08:59 本頁(yè)面
【正文】 0 , 1 , 2 , ...,? 2 ( ) [ ( ) ( ) ] 0 , 12 ( ) ( ) 0 , 1?knnipn n k nmknnmkEEips m i s m s m k i ps m i e m i p???????????? ? ? ? ? ?? ? ? ? ????? ca n f in d v a lu e s o f t h a t m in im iz e b y s e t t in g : g iv in g t h e se t o f e q u a t io n s w h e r e a r e t?)knkknnEe m s m iip????h e v a lu e s o f t h a t m in im iz e ( f r o m n o w o n j u st u se r a t h e r t h a n f o r t h e o p t im u m v a lu e s) p r e d ict io n e r r o r ( ( ) ) is o r t h o g o n a l t o sig n a l ( ( ) ) f o r d e la y s ( o f 1 t o 9 Solution for {αk} 1( , ) ( ) ( )( , ) ( , 0), 1 , 2 , .. .,n n nmpk n nkki k s m i s m ki k i i ppp?? ? ????? ? ??????? d e f in in g w e g e t le a d in g t o a se t o f e q u a t io n s in u n k n o w n s t h a t ca n b eso lv e d in a n e f f icie n t m a n n e r f o r t h e { }10 Solution for {αk} 211( ) ( ) ( )( 0 , 0 ) ( 0 , )( , ) 1 , 0pn n k n nm k mpn n k nknE s m s m s m kEki k i p?? ? ?????? ? ??????? ? ?? m i n i m u m m e a n s q u a r e d p r e d i c ti o n e r r o r h a s t h e f o r m w h i c h can b e w r i tt e n i n t h e f o r m Pr o c e s s : 1 . c o m p u te f o r ( , )()knnkpm i ksm??????2 . s o l v e m a tr i x e q u a ti o n f o r n e e d t o spe c i fy r a n g e o f t o p u te n e e d t o spe c i fy 11 Autocorrelation Method ( ) 0 1( ) ( ) ( )()() a ss u m e e x ist s f o r a n d is e x a ct ly z e r o e v e r y w h e r e e lse ( . , w in d o w o f le n g t h sa m p le s) w h e r e is a f in ite le n g t h w in d o w o f le n g t h sa m p le s if is nnns m m NNs m s m n w mw m Nsm? ? ? ?????112200 1 ( ) ( ) ( )0 1 ,( ) ( )0 , 1...., 1 )n o n z e r o o n ly f o r t h e n is n o n z e r o o n ly o v e r t h e in t e r v a l g iv i ng a t v a lu e s o f n e a r 0 ( . , w e a rpn n k nkNpn n nmmm N e m s m s m km N pE e m e mm m p?????? ? ? ?? ? ? ? ? ?? ? ? ???? ? ????(), 1 , ..., 1 )e p r e d ict in gsig n a l f r o m z e r o v a lu e d sa m p le s o u t sid e t h e w in d o w r a n g e= w ill b e ( r e la t iv e ly ) la r g e a t v a lu e s n e a r ( . , w e a r ep r e d ict in g z e r o v a lu e d sa m p le s ( o u t sid e w in dnemm N m N N N p? ? ? ? ? ?()o w r a n g e ) f r o mn o n z e r o sa m p le s = w ill b e ( r e la t iv e ly ) la r g e f o r t h e se r e a so n s, n o r m a lly u se w in d o w s t h a t t a p e r t h e se g m e n tt o z e r o ( e . g . , Ha m m in g w in d o w )nem?12 Autocorrelation Method 101(0( , ) ( ) 0 0 1( , ) ( ) ( ) , 1 , 0( , ) ( ) ( ) , f o r ca lcula tio n o f sin ce o u tsid e t h e r a n g e , th e n w h ich is e q u iv a len t to t h e f o r m nnNpn n nmN i kn n nmi k s m m Ni k s m i s m k i p k pi k s m s m i k??????? ? ??? ? ? ? ?? ? ? ? ? ? ??? ? ??)1 , 0| | ( , )。( , ) ( ) ( )() t h e r e a r e n o n z e r o t e r m s in t h e co m p u t a tio n o f f o r e a ch v a lue o f a n d ca n e a sil y sh o w t h a t w h e r e is t h e sh o r t tim e a u to co r r e latnnnni p k pN i k i kiki k f i k R i kR i k??? ? ? ?? ? ?? ? ? ????10()( ) ( ) ( )ion o f e v a lua te d a t w h e r e nNkn n nms m i kR k s m s m k???????13 Autocorrelation Method 14 Covariance Method 120()() t h e r e is a se con d b a si c a p p r o a ch to d e f i n i n g t h e sp e e ch seg m e n t a n d t h e li m i ts o n t h e su m s, n a m e l y f i x th e int e r v a l o v e r w h i chth e m e a n squ a r e d e r r o r is co m p u te d , g i v i ng nNnnmsmE e m??????21011011( ) ( )( ,
教學(xué)課件相關(guān)推薦
鄂ICP備17016276號(hào)-1