Wu 2001 Speech-Communication
19
Automatic generation of synthesis units and prosodic information for Chinese concatenative synthesis Chung-Hsien Wu * , Jau-Hung Chen Department of Computer Science and Information Engineering, National Cheng Kung University, 1 Ta-Hsueh Road, Tainan, Taiwan, ROC Received 17 August 1999; accepted 25 July 2000 Abstract In this paper, some approaches to the generation of synthesis units and prosodic information are proposed for Mandarin Chinese text-to-speech TTS) conversion. The monosyllables are adopted as the basic synthesis units. A set of synthesis units is selected from a large continuous speech database based on two cost functions, which minimize the inter- and intra-syllable distortion. The speech database is also employed to establish a word-prosody-based template tree according to the linguistic features: tone combination, word length, part-of-speech POS) of the word, and word position in a phrase. This template tree stores the prosodic features including pitch contour, average energy, and syllable duration of a word for possible combinations of linguistic features. Two modules for sentence intonation and template selection are proposed to generate the target prosodic templates. The experimental results showed that the synthesized prosodic features matched quite well with their original counterparts. Evaluation by subjective experiments also con®rmed the satisfactory performance of these approaches. Ó 2001 Elsevier Science B.V. All rights reserved. Keywords: Chinese text-to-speech conversion; Synthesis units; Prosodic information; Concatenative synthesis; Pitch contour; Syllable duration 1. Introduction In past years, many studies have focused on TTS systems for dierent languages Klatt, 1987; Bigorgne et al., 1993; Kawai et al., 1995). Also, TTS systems and synthesis technology for the Chinese language have been developed in the last two decades Lee et al., 1989, 1993; Chan and Chan 1992; Chen et al., 1998; Shih and Sproat, 1996; Chou and Tseng, 1998). A detailed overview of TTS systems for English and Chinese were in- troduced by Klatt 1987) and Shih and Sproat 1996), respectively. Potential applications include aids for the handicapped, teaching aids, speech-to- speech translation, and any applications for text reading, such as email reader, news reader, and so on. General speaking, a TTS system could be logi- cally composed of three main parts: text/linguistic analysis, prosodic information generation, and speech synthesis. Text analysis is ®rst invoked to analyze the input text. It carries out two main tasks: 1) transcription from letter to phoneme is executed, which forms the basis for the following acoustic unit selection and 2) morphologic and syntactic analyses are performed to extract con- textual information and output linguistic features. Speech Communication 35 2001) 219±237 www.elsevier.com/locate/specom * Corresponding author. Tel.: +86-2757575; fax: +86-2747076. 0167-6393/01/$ - see front matter Ó 2001 Elsevier Science B.V. All rights reserved. PII: S 0 1 6 7 - 6 3 9 3 0 0 ) 0 0 0 7 5 - 3
-
Upload
amanda-martinez -
Category
Documents
-
view
228 -
download
0
description
hkjh
Transcript of Wu 2001 Speech-Communication
-
! " !#
! " ###
$
% &
C40(()* + ( ""," , +
" "
, , - - .
""," + (
, " " ," -, "
" / , - "
C4001)* - -
+ ( "
"
""," - ," , " + (- "
" " "+ ( &" " -
. 2 -"" - " + 3" , ,4 &
" 5 + ## 3" ) 6+7+ "" +
"$ &
) 0 0 )"","
$ (() 8 " C409" :6 "+ ; 9- "+ !*+ "(() "
" ,
" "- C40" ? (() " , """
/ &@""
+ (& " 5 ? ". &+ $ - ?/ C40* "
& - , ""- " C40 * " " & &" " +
)
;! C40 ##* A ;---+"+@"@
+ ("+/ B:= !!! &/ B:= C#=+
#==;;@#@D
## 3" ) 6+7+ "" +0$$/ ) # = = ; ; C40 # # * # # # ! ;
-
>"" & " " (() + ( " " " + (
"" @"
"
2 C40#* + ( " " " ? " ?
" "+ E""
""," "
, , + (
/ " C40) ) =*+ (
" 8
- "
,? + ( , (() + 6
" C400)1 ) * 4 @2+
$
"" " . "","" 2"
+ $ ? ,
C409" :*+ ( ,
, + ) ? "" + -
""," " " ,-
4+ 2" " ""," ""," " 8 " C409" : 6 "+;*+ " "
"
, ,
," 2"
C40 (:*+ ( 5
" 2 +
1 ", ,
5 C409" : ; C !
E"
> = : #
E"
% % & '% % ( )* C40,--. ,/0,)1
-
C+ & + $ " -"" ""- + E &" # -"" " " + ( " - -"" 8
+
$ - "
, , " +( " - ++ , - " 01) - - - ++ ""," + ( ," - "
,
- + I
& " " ;C:
- C40" C""," -* - ,+ (
, - ,
, - (," !+
9%% "
( - "
- E+ !+ ( "
- ""/ -" "" , ""+ $ -" "" """," - ""," - ;"","
- C""," -+ 3 "
-" "" . , , , ""+E , -" ," " + E ""," - ""-/ 3 + ( ,# ; +
+ ( " 5" "+
+ ( " 5" "+
" "/ 01) - - "","+ ( , " 5"+
( " 5" + 6! +$ , ?
" JJ- KK+ ( G" , & "+ - 8 - " - 4 # + " " # -"" " " + 2" - " " C40#* ,+ ( " 8 - "" ," , "+
1
, "
& ," + ( " " , " - # " ? ? ,+ ( - - "
(," !
H, -
,
" (
,
N+ -
"
! :;
# #;
; :# =
C ; # !:#
(," C
( : 01)
N 4 C40* N C40H* 0 C400*
>" , C40H7* % C40%* H C40H(*
" , C40)0* $4 C40$* >" N C40N*
4 C40*
-
,# ; " ,"+( #
"
+
9%,%
-
5 - , 5 ""," -+ $%
" 8 , "
+ )
+
E - - 2 E+= " " - 5 ""," - 5 "","
- = "","+ ( 5""," - , , ""- 2/
(0
-
, !+ (
"" # - 5
#
# #
$ 2 # 8
" -# # + ( ," + $ 8" , ,-
+! ;+ "
- ""," - E+ - " " " " + " ,
"","@- , , "" " "+ " , + (
32+ C40* -""
, - 2 " +
E+ =+ > - " +
;# % % & '% % ( )* C40,--. ,/0,)1
-
9%,%,% E - - 2
" " " 5
- "
- " ,+ ) (0 , " "" - "C40* "+ ( - " , " " - " + $ ,
""-/
( (0 # ( , "" "
- # 5 ""," - "+ ( " # " + (
" ,-
-
(+ ( "
&
(
#01)01)
#$> $>
#E>E> # # #6!6!
( 2 ,
""-/ - " ( #
,-
5
# #
;
01) 01) - - " $> E> , " 5" ""," " " 5" ""," " 6! 6! ""," + ( ," - & "
+
E"" - " - "
+
)
, - -"" , ; C"","
" - + $
- " ""," "","
- , ""-/ E ;""," - , ""," - ""- , ""," -+
E C""," - , - ""," -+
E !""," " - , 5 ""," -+
E+ + " - ""," ! ! # #=!+
% % & '% % ( )* C40,--. ,/0,)1 ;
-
$
,
," , ("
-
""," ""," ##+ ( ""," ;! ! ""," JJ!KK JJCKK "+
*%,%
$ & ;; - 5 ""
"" " ,
& - "" " C40%I)*+ ( %I) - "" , - ""- ? / C40* " -
2" - ? C40 * - "-@ # -
C40;* - @" 5" - C40C* -""@" - + (
; C+ N&
8 ""," + ( " - (," =+( JJ KK 5
," %I)+ E ,"
-
1" =; ""," - "" " " C+:O - ,+ ( " ""," "" - @"
,+
*%)% 7 4
$
, ,
C40 "* 5 - (," + E " ? " C40* ( C "" " C40 #* -"
( ; ! " " C40"- #*+ 5 ( " + E " " ,
5 ""
8 ,-
+ ( , - " C40,*+ ( - -
+ " 5
- # " "" "
""-/ C40++ ##### #####* ! C= ;: . ( A! "+ ( " " C40;: .* - " " ( C+
1 "
"
JJ KK +
(," C40* " - " ""-/
C ( =: ( ; :; ( ;C C ( C! : ( ! C
*%9%
$ & & -
+ E+# - &" "
. 2
" 5" + E ""," & "C40* " . " +$ ,
- " "","+ ( "
" 5" - " C40,* C40* C40* "+ $ ,
""," 2 -"" - +- , "","+ " &" ""," -
= . 5"
(,"
H, C40 "* 5 + 0" C40* " +
0" C40,* " C40O*
C40* ( )+H+
C ;
=: C
; :! :
C C =
!
C40,* (
C40 "* C40O*
# #A;# ;#A!# !#A# #A# #A # #A ;# ;#B
+C =+C ;+ +! #+: + # #
#+C C+: =+: +# +! =+# ;+; #+
; +C +; C+ +! ;+ !+; #+; !+:
C C+ +# :+# !+= !+ =+# +C #+
! #+: C+; :+# :+ ;+: !+; :+ +
;C % % & '% % ( )* C40,--. ,/0,)1
-
E+ #+ 3&" " C40" "* . C40 "* 2 C40*
C40,* C40* " C40* 5" + ( & "","
6!+
% % & '% % ( )* C40,--. ,/0,)1 ;!
-
," " + ( ,
- - "& - ""," -+ 1
" "","
- "
+ " ""," +
) " " - + # ,4 - ? ,4" " (()
""- /+ 5+ $ ,4 -
? " " .
- ?-"
+ ( ,4 - -- + 6 " -" & - ,+
+ + E ,4 - ? " -
" , , (() - 5+ ( .
- 5 (() - "+E .
,4
C40%1)* " A!++ ! &"" "" C "" ;
"" "" ""+
( " ""," - (," #+ ( - =+O+ ," - - " "- "" ""," " + 1 - - " - "," &" " J4K JK JK " J.K JK JK "+ ( ""
+ (," " %1) - - 8 "+ ," %1) - ;+= "+ 1 "","
" ? " , %1) " " -
+ E %1) & - "-
%1)+ ( "? -
+
!
$ " ,
%
(() "
,+ " < "" - + E -
" 2" + ( /
5" "
& " " &+ E - 5" 5
+ )""," - , + (
"
-
+ ( "
2 , + ( " " " " "
" + 3&" " - . 2
-"" - " + 3", ,4 & " 5 +
"#
6 H+ 68 1+ ," 6+ 3 E+
