Language Supplement for German Petra-ml Embedded Premium (front end configuration cfg4)
Nuance® Vocalizer Expressive 2.0
This language supplement describes issues that are particular to the German text-to-speech system. It complements the Vocalizer Expressive User's Guide and Programmer's Reference, which describes the API and the language independent properties of the Vocalizer text-to-speech system.
This document includes these sections:
- Input format
- Text normalization
- Phonetic alphabets
- Voice specific information
- Comments and technical support
This document uses the font Arial Unicode MS to display IPA characters and other special characters in the Phonetic alphabets section. Arial Unicode MS is normally distributed with Microsoft Office. If the German characters don't appear when you view this document, make sure you choose View→Encoding→Unicode (UTF-8) in your browser.
Input format
The German text-to-speech system is designed to pronounce any input written according to the rules of German orthography.
The Vocalizer engine processes text as UTF-16 (transcoded to that if the original input used a different encoding) and thus the input may contain any character from the Unicode Basic Multilingual Plane (BMP) used for any written language. However, the German TTS system only reads the pieces of text that consist of characters that are meaningful for German, discarding all other characters.
For more information on Unicode and UTF-16, see the Unicode Consortium web site.
Text normalization
Vocalizer supports a range of special formats for German. The formats are described in Text Normalization in Vocalizer Expressive for German and German Token Documentation. The supported abbreviations are listed in German Abbreviations
Phonetic alphabets
In the Vocalizer text-to-speech system, there are several ways to intervene directly in the pronunciation of text as described in the Vocalizer Expressive User's Guide and Programmer's Reference, including the ability to specify phonetic input using markup or user dictionaries. German phonetic input strings should be composed of symbols of the L&H+ phonetic alphabet shown in the tables below (L&H+ subset for German).
Table version 2.2Vowel
IPA L&H+ Description IPA Unicode Example IPA Transcription L&H+ Transcription a a open front 0061 St(a)dt ˈʃtat 'Stat ɛ E open-mid front 025B B(e)tt ˈbɛt 'bEt ɪ I mid-high mid-front 026A M(i)lch ˈmɪlç 'mIlC ɔ O open-mid back rounded 0254 v(o)ll ˈfɔl 'fOl ʏ Y open-mid back 028F K(ü)ste ˈkʏs.tə 'kYs.t$ ɛː E: open-mid front long 025B +
02D0F(ä)hre ˈfɛː.ʀə 'fE:.R$ ə $ central (schwa) 0259 Tast(e) ˈtas.tə 'tas.t$ ø e+ close-mid front rounded 00F8 sch(ö)n ˈʃøn 'Se+n ɑ̃ A%~ open back nasalized 0251 +
0303Restaur(ant) ʀɛs.toː.ˈʀɑ̃ REs.to:.'RA%~ ɔ̃ O%~ open-mid back rounded nasalized 0254 +
0303Balk(on) bal.ˈkɔ̃ bal.'kO%~ ɛ̃ E%~ open-mid front nasalized 025B +
0303T(eint) ˈtɛ̃ 'tE%~ œ̃ E+%~ open-mid front rounded nasalized 0153 +
0303Parf(um) paʀ.ˈfœ̃ paR.'fE+%~ œ E+ open-mid front rounded 0153 l(ö)schen ˈlœ.ʃən 'lE+.S$n ʊ U mid-high back rounded 028A K(u)nst ˈkʊnst 'kUnst iː i: close front long 0069 +
02D0R(ie)se ˈʀiː.zə 'Ri:.z$ eː e: close-mid front long 0065 +
02D0K(e)hle ˈkeː.lə 'ke:.l$ aː a: open front long 0061 +
02D0W(a)gen ˈvaː.gən 'va:.g$n yː y: close front rounded long 0079 +
02D0gr(ü)n ˈgʀyːn 'gRy:n uː u: close back rounded long 0075 +
02D0F(u)ß ˈfuːs 'fu:s oː o: close-mid back rounded long 006F +
02D0gr(o)ß ˈgʀoːs 'gRo:s Diphthong
IPA L&H+ Description IPA Unicode Example IPA Transcription L&H+ Transcription a͡u a&u from open front to close back rounded 0061 +
0361 +
0075B(au)m ˈba͡um 'ba&um ɔ͡y O&y from open-mid back rounded to close front rounded 0254 +
0361 +
0079h(eu)te ˈhɔ͡y.tə 'hO&y.t$ a͡i a&i from open front to close front 0061 +
0361 +
0069T(ei)l ˈta͡il 'ta&il Consonant (Pulmonic)
IPA L&H+ Description IPA Unicode Example IPA Transcription L&H+ Transcription j j palatal approximant voiced 006A (j)emand ˈjeː.mant 'je:.mant p p bilabial plosive voiceless 0070 (P)ost ˈpɔst 'pOst b b bilabial plosive voiced 0062 (B)ein ˈba͡in 'ba&in t t alveolar plosive voiceless 0074 (T)inte ˈtɪn.tə 'tIn.t$ d d alveolar plosive voiced 0064 (d)ich ˈdɪç 'dIC k k velar plosive voiceless 006B (k)lein ˈkla͡in 'kla&in g g velar plosive voiced 0067 lie(g)en ˈliː.gən 'li:.g$n f f labiodental fricative voiceless 0066 (F)eld ˈfɛlt 'fElt v v labiodental fricative voiced 0076 (w)ach ˈvax 'vax s s alveolar fricative voiceless 0073 Fel(s) ˈfɛls 'fEls z z alveolar fricative voiced 007A (S)aal ˈzaːl 'za:l ʃ S postalveolar fricative voiceless 0283 (Sch)nee ˈʃneː 'Sne: ʒ Z postalveolar fricative voiced 0292 Gara(g)e ga.ˈʀaː.ʒə ga.'Ra:.Z$ x x velar fricative voiceless 0078 Ba(ch) ˈbax 'bax h h glottal fricative voiceless 0068 (H)and ˈhant 'hant m m bilabial nasal voiced 006D (M)ann ˈman 'man n n alveolar nasal voiced 006E (N)orden ˈnɔʀ.dən 'nOR.d$n ŋ nK velar nasal voiced 014B Ri(ng) ˈʀɪŋ 'RInK l l alveolar lateral approximant 006C (L)icht ˈlɪçt 'lICt ʔ ? glottal plosive 0294 be()achten bə.ˈʔax.tən b$.'?ax.t$n ʀ R uvular trill 0280 (R)eise ˈʀa͡i.zə 'Ra&i.z$ ç C palatal fricative voiceless 00E7 Mil(ch) ˈmɪlç 'mIlC Affricate
IPA L&H+ Description IPA Unicode Example IPA Transcription L&H+ Transcription t͡ʃ t&S affricate voiceless from alveolar to postalveolar 0074 +
0361 +
0283Ru(tsch) ˈʀʊt͡ʃ 'RUt&S d͡ʒ d&Z affricate voiced from alveolar to postalveolar 0064 +
0361 +
0292(J)azz ˈd͡ʒɛs 'd&ZEs t͡s t&s affricate voiceless from alveolar to alveolar 0074 +
0361 +
0073(Z)ug ˈt͡suːk 't&su:k p͡f p&f affricate voiceless from bilabial to labiodental 0070 +
0361 +
0066(Pf)erd ˈp͡feːʀt 'p&fe:Rt Suprasegmental
IPA L&H+ Description IPA Unicode . . syllable break 002E ˈ ' primary stress 02C8 ˌ '2 secondary stress 02CC Other symbol
IPA L&H+ Description None " sentence accent None # silence (pause) None _ word delimitor None *. end of declarative None *, comma None *! end of exclamation None *? end of question None *; semicolon None *: colon Microsoft SAPI 5 API Phoneme Mappings
SAPI Symbol L&H+ SAPI Phone ID - . 0001 ! *! 0002 & _ 0003 , *, 0004 . *. 0005 ? *? 0006 _ # 0007 ^ ? 0008 1 ' 0009 2 '2 000a ~ 000b : : 000c a A 000d a : A: 000d+000c a : ax a:R 000d+000c+000f a ax aR 000d+000f aw a&u 000e ax $ 000f ay a&I 0010 b B 0011 d D 0012 ch t&S 0013 eh E 0014 eh : e: 0014+000c eh : ax e:R 0014+000c+000f eh ax ER 0014+000f eu e+ 0015 eu ax e+R 0015+000f ey E: 0016 ey ax e:R 0016+000f f F 0017 g G 0018 h H 0019 ih I 001a ih ax IR 001a+000f iy i: 001b iy ax i:R 001b+000f jh d&Z 001c k K 001d l L 001e m M 001f n N 0020 ng NK 0021 oe E+ 0022 oe ax E+R 0022+000f oh O 0023 oh ax OR 0023+000f ow O: 0024 ow ax o:R 0024+000f oy O&y 0025 p P 0026 pf p&f 0027 r R 0028 s S 0029 sh S 002a t T 002b ts t&s 002c ue Y: 002d ue ax y:R 002d+000f uh U 002e uh ax UR 002e+000f uw U: 002f uw ax u:R 002f+000f uy Y 0030 uy ax YR 0030+000f v V 0031 x C 0032 y J 0033 z Z 0034 zh Z 0035
Voice specific information
Multi-lingual skills Multi-lingual skills of Petra-ml embedded-premium
Linguistic knowledge Pronunciation orthographic: (none) near native orthographic: eng,frf,iti,spe accented orthographic: (none) approximate phonetic: all other languages approximate
Comments and technical support
Send comments on Nuance documentation to techdoc@nuance.com .
Technical support is provided online at the Nuance Network .
© 2012-2015 Nuance Communications, Inc. All Rights Reserved.