Skip to content

Commit

Permalink
Better ToBI rules for punctuation (Issue marytts-it#21)
Browse files Browse the repository at this point in the history
  • Loading branch information
@ftesser
ftesser authored and alize committed Mar 4, 2013
1 parent b799d4e commit bd649ea
Showing 1 changed file with 221 additions and 26 deletions.
247 changes: 221 additions & 26 deletions marytts-lang-it/src/main/resources/marytts/language/it/prosody/tobipredparams_italian.xml
View file
Original file line number Diff line number Diff line change
Expand Up @@ -114,11 +114,12 @@ Possible values are only: "enfofpar"(end of paragraph) and "endofvorfeld"(end of
<list name="all_interrog" items="interrog:interrogW"/>
-->


<!-- R1: è, ho, ha, TOCHECK ma anche coniugazioni: erano sono avevano-->
<!-- R1: è, ho, ha, TOCHECK ma anche coniugazioni: erano sono avevano sono:sei:siamo:siete:ero:eri risposta: si?-->
<list name="verbs0" items="è:ho:ha"/>

<list name="openFB" items="«:(:{:["/>
<list name="FBopen" items="«:(:{:["/>

<list name="FFelipses" items="...:.."/>

<!-- R2: DE, DQ, PQ ad inizio frase ricevono accento -->
<list name="det0" items="DE:DQ:PQ"/>
Expand Down Expand Up @@ -148,17 +149,15 @@ Possible values are only: "enfofpar"(end of paragraph) and "endofvorfeld"(end of
</definitions>

<!-- the accentposition rules determine if a token gets a tone accent or if it doesn't receive any accent (no force accents in English) -->
<accentposition>

<accentposition>

<!--
R1: se non una tra
V+S
V+SP
V+A+S
V+A+SP
Allora pitch accent su verbo (solo sul verbo?)
*
Expand All @@ -168,6 +167,7 @@ Possible values are only: "enfofpar"(end of paragraph) and "endofvorfeld"(end of
Corri grande Fabio.
Corri Fabio.
-->

<rule>
<previousAttributes pos="V"/>
<attributes pos="INLIST:nouns0"/>
Expand Down Expand Up @@ -200,15 +200,16 @@ Possible values are only: "enfofpar"(end of paragraph) and "endofvorfeld"(end of
<!-- R3 pos = BN + è ho ha
R3:
BN + V se il verbo è {è, ho, ha}
non riceve PA (chi? il verbo o BN, il verbo direi...TOCHECK chiedi conferma cinzia)
non riceve PA (chi? il verbo o BN, il verbo direi...TOCHECK chiedi conferma cinzia)
-->
<!-- non usata: da indagare
<rule>
<previousAttributes pos="BN"/>
<text word="INLIST:verbs0"/>
<action accent=""/>
</rule>
<!-- -->
-->

<rule> <!-- list of words that usually receive an accent(content words) -->
<attributes pos="INLIST:pos_tonal_accent"/>
Expand Down Expand Up @@ -394,37 +395,233 @@ Possible values are only: "enfofpar"(end of paragraph) and "endofvorfeld"(end of
<!-- Le precedenti regole: se la prima parola di una interrogativa inizia con
(chi, che, cosa, perché, quale, quali, quanto, come, dove, ...) (PR,PQ,DQ,E,B,CS) allora -> L-L% altrimenti (non inizia con: ...) -> L-H% -->

<!-- dinstinzione tra , e ...
comma -> L-
comma in lista H- o L- penultima H-
ellipsis -> H-
:
ellipsis test sentence:
ciao... sono io che mi chiamo fabio... tu come ti chiami?
-->

<rule> <!-- FF ellipsis -->
<text word="INLIST:FFelipses"/>
<folTokens num="1+"/>
<prevTokens num="4+"/>
<nextAttributes pos="!INLIST:pos_punctuation"/>
<previousAttributes pos="!INLIST:pos_punctuation"/>
<previousMinus1Attributes pos="!INLIST:pos_punctuation"/>
<previousMinus2Attributes pos="!INLIST:pos_punctuation"/>
<previousMinus3Attributes pos="!INLIST:pos_punctuation"/>
<action bi="4" tone="L-H%"/>
</rule>

<rule> <!-- FF ellipsis -->
<text word="INLIST:FFelipses"/>
<folTokens num="1+"/>
<prevTokens num="1+"/>
<nextAttributes pos="!INLIST:pos_punctuation"/>
<previousAttributes pos="!INLIST:pos_punctuation"/>
<action bi="3" tone="H-"/>
</rule>



<rule> <!-- comma or ellipsis (FF , ...) after al leats 5 token of not punctuation
type major boundary after a punctuation mark in the middle of the sentence -->
<rule> <!-- FF not ellipsis (comma) -->
<attributes pos="FF"/>
<folTokens num="1+"/>
<prevTokens num="1+"/>
<prevTokens num="7+"/>
<nextAttributes pos="!INLIST:pos_punctuation"/>
<previousAttributes pos="!INLIST:pos_punctuation"/>
<previousMinus1Attributes pos="!INLIST:pos_punctuation"/>
<previousMinus2Attributes pos="!INLIST:pos_punctuation"/>
<previousMinus3Attributes pos="!INLIST:pos_punctuation"/>
<previousMinus4Attributes pos="!INLIST:pos_punctuation"/>
<previousMinus5Attributes pos="!INLIST:pos_punctuation"/>
<previousMinus6Attributes pos="!INLIST:pos_punctuation"/>
<action bi="4" tone="L-L%"/>
</rule>

<rule> <!-- FF not ellipsis (comma) -->
<attributes pos="FF"/>
<folTokens num="1+"/>
<prevTokens num="1+"/>
<nextAttributes pos="!INLIST:pos_punctuation"/>
<previousAttributes pos="!INLIST:pos_punctuation"/>
<action bi="3" tone="L-"/>
</rule>



<rule> <!-- FC colon-semicolon -->
<text word="INLIST:FCcolon-semicolon"/>
<folTokens num="1+"/>
<prevTokens num="4+"/>
<nextAttributes pos="!INLIST:pos_punctuation"/>
<previousAttributes pos="!INLIST:pos_punctuation"/>
<previousMinus1Attributes pos="!INLIST:pos_punctuation"/>
<previousMinus2Attributes pos="!INLIST:pos_punctuation"/>
<previousMinus3Attributes pos="!INLIST:pos_punctuation"/>
<action bi="4" tone="L-L%"/>
</rule>

<rule> <!-- FC colon-semicolon -->
<text word="INLIST:FCcolon-semicolon"/>
<folTokens num="1+"/>
<prevTokens num="1+"/>
<nextAttributes pos="!INLIST:pos_punctuation"/>
<previousAttributes pos="!INLIST:pos_punctuation"/>
<action bi="3" tone="L-"/>
</rule>


<rule> <!-- FC not colon-semicolon -->
<attributes pos="FC"/>
<folTokens num="1+"/>
<prevTokens num="4+"/>
<nextAttributes pos="!INLIST:pos_punctuation"/>
<previousAttributes pos="!INLIST:pos_punctuation"/>
<previousMinus1Attributes pos="!INLIST:pos_punctuation"/>
<previousMinus2Attributes pos="!INLIST:pos_punctuation"/>
<previousMinus3Attributes pos="!INLIST:pos_punctuation"/>
<action bi="4" tone="L-L%"/>
</rule>

<rule> <!-- FC not colon-semicolon -->
<attributes pos="FC"/>
<folTokens num="1+"/>
<prevTokens num="1+"/>
<nextAttributes pos="!INLIST:pos_punctuation"/>
<previousAttributes pos="!INLIST:pos_punctuation"/>
<action bi="3" tone="L-"/>
</rule>


<rule> <!-- FB open brackets -->
<text word="INLIST:INLIST:FBopen"/>
<folTokens num="1+"/>
<prevTokens num="4+"/>
<nextAttributes pos="!INLIST:pos_punctuation"/>
<previousAttributes pos="!INLIST:pos_punctuation"/>
<previousMinus1Attributes pos="!INLIST:pos_punctuation"/>
<previousMinus2Attributes pos="!INLIST:pos_punctuation"/>
<previousMinus3Attributes pos="!INLIST:pos_punctuation"/>
<action bi="4" tone="L-L%"/>
</rule>

<rule> <!-- FB open brackets -->
<text word="INLIST:INLIST:FBopen"/>
<folTokens num="1+"/>
<prevTokens num="1+"/>
<nextAttributes pos="!INLIST:pos_punctuation"/>
<previousAttributes pos="!INLIST:pos_punctuation"/>
<action bi="3" tone="L-"/>
</rule>


<rule> <!-- FB closed brackets -->
<text word="INLIST:INLIST:FBclosed"/>
<folTokens num="1+"/>
<prevTokens num="4+"/>
<nextAttributes pos="!INLIST:pos_punctuation"/>
<previousAttributes pos="!INLIST:pos_punctuation"/>
<previousMinus1Attributes pos="!INLIST:pos_punctuation"/>
<previousMinus2Attributes pos="!INLIST:pos_punctuation"/>
<previousMinus3Attributes pos="!INLIST:pos_punctuation"/>
<action bi="4" tone="H-L%"/>
</rule>

<rule> <!-- FB closed brackets -->
<text word="INLIST:INLIST:FBclosed"/>
<folTokens num="1+"/>
<prevTokens num="1+"/>
<nextAttributes pos="!INLIST:pos_punctuation"/>
<previousAttributes pos="!INLIST:pos_punctuation"/>
<action bi="3" tone="H-"/>
</rule>

<!-- -->

<rule> <!-- FB not open or closed brackets -->
<attributes pos="FB"/>
<folTokens num="1+"/>
<prevTokens num="4+"/>
<nextAttributes pos="!INLIST:pos_punctuation"/>
<previousAttributes pos="!INLIST:pos_punctuation"/>
<previousMinus1Attributes pos="!INLIST:pos_punctuation"/>
<previousMinus2Attributes pos="!INLIST:pos_punctuation"/>
<previousMinus3Attributes pos="!INLIST:pos_punctuation"/>
<action bi="4" tone="L-L%"/>
</rule>

<rule> <!-- FB not open or closed brackets -->
<attributes pos="FB"/>
<folTokens num="1+"/>
<prevTokens num="1+"/>
<nextAttributes pos="!INLIST:pos_punctuation"/>
<previousAttributes pos="!INLIST:pos_punctuation"/>
<action bi="3" tone="H-"/>
</rule>


<rule> <!-- others major boundary after a punctuation mark in the middle of the sentence -->
<attributes pos="INLIST:pos_punctuation"/>
<folTokens num="1+"/>
<prevTokens num="1+"/>
<action bi="3" tone="H-"/>
</rule>


<rule> <!-- others major boundary after a punctuation mark in the middle of the sentence -->
<attributes pos="INLIST:pos_punctuation"/>
<folTokens num="1+"/>
<prevTokens num="4+"/>
<nextAttributes pos="!INLIST:pos_punctuation"/>
<previousAttributes pos="!INLIST:pos_punctuation"/>
<previousMinus1Attributes pos="!INLIST:pos_punctuation"/>
<previousMinus2Attributes pos="!INLIST:pos_punctuation"/>
<previousMinus3Attributes pos="!INLIST:pos_punctuation"/>
<action bi="4" tone="L-L%"/>
</rule>



<!--
<rule> <!-- comma or ellipsis (FF , ...) after al leats 5 token of not punctuation
type major boundary after a punctuation mark in the middle of the sentence -->
<rule> comma or ellipsis (FF , ...) after al leats 5 token of not punctuation
type major boundary after a punctuation mark in the middle of the sentence
<attributes pos="FF"/>
<folTokens num="1+"/>
<prevTokens num="1+"/>
<nextAttributes pos="!INLIST:pos_punctuation"/>
<previousAttributes pos="!INLIST:pos_punctuation"/>
<previousMinus1Attributes pos="!INLIST:pos_punctuation"/>
<previousMinus2Attributes pos="!INLIST:pos_punctuation"/>
<previousMinus3Attributes pos="!INLIST:pos_punctuation"/>
<previousMinus4Attributes pos="!INLIST:pos_punctuation"/>
<previousMinus5Attributes pos="!INLIST:pos_punctuation"/>
<action bi="4" tone="H-L%"/>
</rule>
-->
<!--
<rule> comma or ellipsis (FF , ...) after al leats 5 token of not punctuation
type major boundary after a punctuation mark in the middle of the sentence
<attributes pos="FF"/>
<nextAttributes pos="!INLIST:pos_punctuation"/>
<previousAttributes pos="!INLIST:pos_punctuation"/>
<folTokens num="1+"/>
<prevTokens num="1+"/>
<action bi="3" tone="H-"/>
</rule>

-->

<!-- Not used
<rule> (FB « » ( ) " ') minor boundary after in the middle of the sentence
<attributes pos="FB"/>
<text word="INLIST:openFB"/>
<text word="INLIST:FBopen"/>
<nextAttributes pos="!INLIST:pos_punctuation"/>
<previousAttributes pos="!INLIST:pos_punctuation"/>
<folTokens num="1+"/>
Expand All @@ -433,15 +630,19 @@ Possible values are only: "enfofpar"(end of paragraph) and "endofvorfeld"(end of
</rule>
-->

<rule> <!-- (FB « » ( ) " ') sentence type major boundary after a punctuation mark in the middle of the sentence -->


<!-- Not used
<rule> (FB « » ( ) " ') sentence type major boundary after a punctuation mark in the middle of the sentence
<attributes pos="FB"/>
<folTokens num="1+"/>
<prevTokens num="1+"/>
<action bi="4" tone="H-L%"/>
</rule>
-->


<rule> <!-- (FC : ; -) sentence type major boundary after a punctuation mark in the middle of the sentence -->
<!-- Not used
<rule> (FC . : ; -) sentence type major boundary after a punctuation mark in the middle of the sentence
<attributes pos="FC"/>
<folTokens num="1+"/>
<prevTokens num="1+"/>
Expand All @@ -450,15 +651,9 @@ Possible values are only: "enfofpar"(end of paragraph) and "endofvorfeld"(end of
<previousMinus1Attributes pos="!INLIST:pos_punctuation"/>
<action bi="4" tone="H-L%"/>
</rule>
-->


<rule> <!-- others major boundary after a punctuation mark in the middle of the sentence -->
<attributes pos="INLIST:pos_punctuation"/>
<folTokens num="1+"/>
<prevTokens num="1+"/>
<action bi="4" tone="H-L%"/>
</rule>





Expand Down

0 comments on commit bd649ea

Please sign in to comment.