marytts.modules.phonemiser.Syllabifier Maven / Gradle / Ivy
The newest version!
/**
* Copyright 2000-2006 DFKI GmbH.
* All Rights Reserved. Use is subject to license terms.
*
* This file is part of MARY TTS.
*
* MARY TTS is free software: you can redistribute it and/or modify
* it under the terms of the GNU Lesser General Public License as published by
* the Free Software Foundation, version 3 of the License.
*
* This program is distributed in the hope that it will be useful,
* but WITHOUT ANY WARRANTY; without even the implied warranty of
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
* GNU Lesser General Public License for more details.
*
* You should have received a copy of the GNU Lesser General Public License
* along with this program. If not, see phoneList = splitIntoAllophones(phoneString);
syllabify(phoneList);
StringBuilder sb = new StringBuilder();
for (String p : phoneList) {
if (sb.length() > 0)
sb.append(" ");
sb.append(p);
}
return sb.toString();
}
/**
* Syllabify a LinkedList of phones. This is an implementation of the syllabification rules by Jürgen Trouvain.
*
* @param phoneList
* phoneList
* @return a LinkedList of phone strings with inserted "-" strings at syllable boundaries.
*/
public LinkedList syllabify(LinkedList phoneList) {
// Regel(1a)
// Jede Grenze einer morphologischen Wurzel stellt eine
// Silbengrenze dar.
// Regel(1b)
// Jede Grenze eines Präfixes stellt eine Silbengrenze dar.
// Dort, wo ein Fugensuffix bzw ein Suffix beginnt,
// gibt es keine morphologische Silbengrenze.
// Bsp: Lebens-gefaehrte und nicht Leben-s-gefaehrte
// Bsp: Mei-nung und nicht Mein-ung
// Note: We don't have morpheme boundaries! (=> so ignore rule 1)
// 2.: finde nicht-morphologische Silbengrenzen
// teile Woerter, die eine morphologisch bedingte Silbengrenze haben
// in ihre Silbenteile, um dort spaeter nach weiteren,
// nicht-morphologisch bedingten, Silbengrenzen zu suchen
// Only one such component as long as we don't have morpheme boundaries
if (phoneList == null) {
throw new IllegalArgumentException("Cannot syllabify null string");
}
ListIterator it = phoneList.listIterator(0);
if (!it.hasNext()) {
return phoneList;
}
Allophone previous = getAllophone(it.next());
boolean previousIsVowel = (previous != null && previous.sonority() >= 4);
while (it.hasNext()) {
Allophone next = getAllophone(it.next());
boolean nextIsVowel = (next != null && next.sonority() >= 4);
// Regel(5)
// Wenn zwischen zwei Vokalen keine weiteren Phone sind,
// dann setze die Silbengrenze vor den zweiten Vokal.
if (previousIsVowel && nextIsVowel && !next.name().equals("6")) {
// Insert a syllable boundary between the two.
it.previous(); // one step back
it.add("-"); // insert syllable boundary
it.next(); // and forward again
}
previousIsVowel = nextIsVowel;
}
// Regel(4)
// Suche das "Tal" (kleinster Level < 4) zwischen zwei benachbarten
// Vokalen, sofern die Vokale nicht durch eine morphologisch bedingte
// Grenze getrennt werden.
it = phoneList.listIterator(0);
int minSonority = 7; // one higher than possible maximum.
int minIndex = -1; // position of the sonority minimum
int syllableStart = -1;
while (it.hasNext()) {
String s = it.next();
if (s.equals("-")) {
// Forget about all valleys:
minSonority = 7;
minIndex = -1;
syllableStart = it.previousIndex();
} else {
Allophone ph = getAllophone(s);
if (ph != null && ph.sonority() < minSonority) {
minSonority = ph.sonority();
minIndex = it.previousIndex();
} else if (ph != null && ph.sonority() >= 4) {
// Found a vowel. Now, if there is a (non-initial) sonority
// valley before this vowel, insert a valley marker:
if (minIndex > syllableStart + 1) {
int steps = 0;
while (it.nextIndex() > minIndex) {
steps++;
it.previous();
}
it.add(".");
while (steps > 0) {
it.next();
steps--;
}
}
minSonority = 7;
minIndex = -1;
}
}
}
// Regel(6a)
// Steht zwischen einem ungespannten Vokal (Level 5) und dem
// darauffolgenden Vokal (Level 4, 5 oder 6) nur *ein* Konsonant des
// Levels 2 oder 3, so ersetze die Talmarkierung durch eine
// ambisilbische Silbengrenze (Symbol "_").
// halbformal:
// ([v5]).([k2,3])([v4,5,6])
// --> ([v5])_([k2,3])([v4,5,6])
// Regel(6b)
// Steht zwischen einem ungespannten Vokal (Level 5) und dem
// darauffolgenden Vokal (Level 4, 5 oder 6) mehr als ein Konsonant
// (Levels 1,2 oder 3), und folgt gleichzeitig dem 5er-Vokal eine
// Talmarkierung, so versetze die "Talmarkierung" ein Phonem weiter und
// ersetze sie durch eine normale Silbengrenze.
// halbformal:
// ([v5]).([k1,2,3])([k1,2,3]+)([v4,5,6])
// --> ([v5]).([k1,2,3])-([k1,2,3]+)([v4,5,6])
// Regel(6c)
// In allen anderen Faellen ersetze die "Talmarkierung" mit einer
// normalen Silbengrenze.
it = phoneList.listIterator(0);
while (it.hasNext()) {
String s = it.next();
if (s.equals(".")) {
it.previous(); // skip . backwards
Allophone ph = getAllophone(it.previous());
it.next();
it.next(); // skip ph and . forwards
if (ph != null && ph.sonority() == 5) {
// The phone just after the marker:
ph = getAllophone(it.next());
if (ph != null && ph.sonority() <= 3) {
// Now the big question: another consonant or not?
ph = getAllophone(it.next());
if (ph != null && ph.sonority() <= 3) {
// (6b) remove ., go one further, insert -
// two ph back, and the .:
it.previous();
it.previous();
it.previous();
it.remove(); // remove the .
it.next(); // skip one ph
it.add("-");
} else {
// (6a) replace . with _
// two ph back, and the .:
it.previous();
it.previous();
it.previous();
// only use minuses, because underscores denote also pauses
// it.set("_"); // replace . with _
it.set("-"); // replace . with -
}
} else {
// unlikely case: no consonant after a 5
it.previous();
it.previous();
it.set("-");
}
} else {
// (6c) simply replace . with -
it.set("-");
}
}
}
// Regel(7)
// Folgt einem Phonem /N/, vor dem unmittelbar eine ambisilbische
// Silbengrenze steht, ein Vollvokal (Level 5 oder 6), so verschiebe
// die Silbengrenze um ein Phonem (naemlich hinter das /N/) und
// ersetze es durch eine normale Silbengrenze.
// halbformal:
// _N([v5,6])
// --> N-([v5,6])
it = phoneList.listIterator(0);
while (it.hasNext()) {
String s = it.next();
// only use minuses, because underscores denote also pauses
// if (s.equals("_")) {
if (s.equals("-")) {
Allophone ph = getAllophone(it.next());
if (ph != null && ph.name().equals("N")) {
ph = getAllophone(it.next());
if (ph != null && ph.sonority() >= 5) {
// (7) remove _, put a - after the N
// skip vowel, N, and _ backwards:
it.previous();
it.previous();
it.previous();
it.remove(); // remove _
it.next(); // skip N forwards
it.add("-"); // insert -
} // else, just leave it
}
}
}
correctStressSymbol(phoneList);
return phoneList;
}
/**
* For those syllables containing a "1" character, remove that "1" character and add a stress marker ' at the beginning of the
* syllable.
*
* @param phoneList
* phoneList
*/
protected void correctStressSymbol(LinkedList phoneList) {
boolean stressFound = false;
ListIterator it = phoneList.listIterator(0);
while (it.hasNext()) {
String s = it.next();
if (s.endsWith("1")) {
if (this.removeTrailingOneFromPhones) {
it.set(s.substring(0, s.length() - 1)); // delete "1"
}
if (!stressFound) {
// Only add a stress marker for first occurrence of "1":
// Search backwards for syllable boundary or beginning of word:
int steps = 0;
while (it.hasPrevious()) {
steps++;
String t = it.previous();
if (t.equals("-") || t.equals("_")) { // syllable boundary
it.next();
steps--;
break;
}
}
it.add("'");
while (steps > 0) {
it.next();
steps--;
}
stressFound = true;
}
}
}
// No stressed vowel in word?
if (!stressFound) {
// Stress first non-schwa syllable
it = phoneList.listIterator(0);
while (it.hasNext()) {
String s = it.next();
Allophone ph = allophoneSet.getAllophone(s);
if (ph.sonority() >= 5) { // non-schwa vowel
// Search backwards for syllable boundary or beginning of word:
int steps = 0;
while (it.hasPrevious()) {
steps++;
String t = it.previous();
if (t.equals("-") || t.equals("_")) { // syllable boundary
it.next();
steps--;
break;
}
}
it.add("'");
while (steps > 0) {
it.next();
steps--;
}
break; // OK, that's it.
}
}
}
}
/**
* Convert a phone string into a list of string representations of individual phones. The input can use the suffix "1" to
* indicate stressed vowels.
*
* @param phoneString
* the phone string to split
* @return a linked list of strings, each string representing an individual phone
* @deprecated This duplicates (badly) {@link AllophoneSet#splitAllophoneString(String)}; use that method instead.
*/
@Deprecated
protected LinkedList splitIntoAllophones(String phoneString) {
LinkedList phoneList = new LinkedList();
for (int i = 0; i < phoneString.length(); i++) {
// Try to cut off individual segments,
// starting with the longest prefixes,
// and allowing for a suffix "1" marking stress:
String name = null;
for (int j = 3; j >= 1; j--) {
if (i + j <= phoneString.length()) {
String candidate = phoneString.substring(i, i + j);
try {
allophoneSet.getAllophone(candidate);
name = candidate;
i += j - 1; // so that the next i++ goes beyond current phone
break;
} catch (IllegalArgumentException e) {
// ignore
}
}
}
if (name != null) {
phoneList.add(name);
}
}
return phoneList;
}
/**
* Get the Allophone object named phone; if phone ends with "1", discard the "1" and use the rest of the string as the phone
* symbol.
*
* @param phone
* phone
* @deprecated Use {@link AllophoneSet#getAllophone(String)} instead
* @return allophoneset.getAllophone(phonesubstring(0, phone.length() - 1)) if this.removeTrailingOneFromPhones and
* phone.endsWith("1"), allophoneset.getAllophone(phonesubstring(phone) otherwise
*/
@Deprecated
protected Allophone getAllophone(String phone) {
if (this.removeTrailingOneFromPhones && phone.endsWith("1"))
return allophoneSet.getAllophone(phone.substring(0, phone.length() - 1));
else
return allophoneSet.getAllophone(phone);
}
}
© 2015 - 2025 Weber Informatics LLC | Privacy Policy