mozilla
/
marytts
зеркало из https://github.com/mozilla/marytts.git
1
0
Форкнуть 0
Код Задачи Пакеты Проекты Релизы Вики Активность

format java code in marytts-lang-en 

incantation:
$ mvn com.googlecode.maven-java-formatter-plugin:maven-java-formatter-plugin:format -pl marytts-languages/marytts-lang-en
This commit is contained in:
Ingmar Steiner 2014-12-19 13:10:47 +01:00
Родитель 6b091740df
Коммит 7c6ec7933d
33 изменённых файлов: 1134 добавлений и 1525 удалений
Показать статистику Скачать Patch файл Скачать Diff файл Показать все файлы Свернуть все файлы

1
marytts-languages/marytts-lang-en/src/main/java/marytts/language/en/EnglishConfig.java
Просмотреть файл

@ -22,7 +22,6 @@ package marytts.language.en;
import marytts.config.LanguageConfig;
import marytts.exceptions.MaryConfigurationException;
public class EnglishConfig extends LanguageConfig {
public EnglishConfig() throws MaryConfigurationException {
super(EnglishConfig.class.getResourceAsStream("en.config"));

64
marytts-languages/marytts-lang-en/src/main/java/marytts/language/en/FreeTTSContourGenerator.java
Просмотреть файл

@ -32,54 +32,38 @@ import com.sun.speech.freetts.Utterance;
import com.sun.speech.freetts.UtteranceProcessor;
import com.sun.speech.freetts.en.ContourGenerator;
/**
* Use an individual FreeTTS module for English synthesis.
*
*
* @author Marc Schröder
*/
public class FreeTTSContourGenerator extends InternalModule
{
private UtteranceProcessor processor;
public class FreeTTSContourGenerator extends InternalModule {
private UtteranceProcessor processor;
public FreeTTSContourGenerator()
{
super("ContourGenerator",
USEnglishDataTypes.FREETTS_DURATIONS,
USEnglishDataTypes.FREETTS_CONTOUR,
Locale.ENGLISH
);
}
public void startup() throws Exception
{
super.startup();
// Initialise FreeTTS
FreeTTSVoices.load();
processor = new ContourGenerator
(com.sun.speech.freetts.en.us.CMUVoice.class.
getResource("f0_lr_terms.txt"), 170.0f, 34.0f);
}
public MaryData process(MaryData d)
throws Exception
{
List utterances = d.getUtterances();
Iterator it = utterances.iterator();
while (it.hasNext()) {
Utterance utterance = (Utterance) it.next();
processor.processUtterance(utterance);
}
MaryData output = new MaryData(outputType(), d.getLocale());
output.setUtterances(utterances);
return output;
}
public FreeTTSContourGenerator() {
super("ContourGenerator", USEnglishDataTypes.FREETTS_DURATIONS, USEnglishDataTypes.FREETTS_CONTOUR, Locale.ENGLISH);
}
public void startup() throws Exception {
super.startup();
// Initialise FreeTTS
FreeTTSVoices.load();
processor = new ContourGenerator(com.sun.speech.freetts.en.us.CMUVoice.class.getResource("f0_lr_terms.txt"), 170.0f,
34.0f);
}
public MaryData process(MaryData d) throws Exception {
List utterances = d.getUtterances();
Iterator it = utterances.iterator();
while (it.hasNext()) {
Utterance utterance = (Utterance) it.next();
processor.processUtterance(utterance);
}
MaryData output = new MaryData(outputType(), d.getLocale());
output.setUtterances(utterances);
return output;
}
}

67
marytts-languages/marytts-lang-en/src/main/java/marytts/language/en/FreeTTSIntonator.java
Просмотреть файл

@ -33,56 +33,39 @@ import com.sun.speech.freetts.UtteranceProcessor;
import com.sun.speech.freetts.cart.CARTImpl;
import com.sun.speech.freetts.cart.Intonator;
/**
* Use an individual FreeTTS module for English synthesis.
*
*
* @author Marc Schröder
*/
public class FreeTTSIntonator extends InternalModule
{
private UtteranceProcessor processor;
public class FreeTTSIntonator extends InternalModule {
private UtteranceProcessor processor;
public FreeTTSIntonator()
{
super("Intonator",
USEnglishDataTypes.FREETTS_PAUSES,
USEnglishDataTypes.FREETTS_INTONATION,
Locale.ENGLISH
);
}
public void startup() throws Exception
{
super.startup();
// Initialise FreeTTS
FreeTTSVoices.load();
CARTImpl accentCart = new CARTImpl(
com.sun.speech.freetts.en.us.CMUVoice.class.getResource("int_accent_cart.txt"));
CARTImpl toneCart = new CARTImpl(
com.sun.speech.freetts.en.us.CMUVoice.class.getResource("int_tone_cart.txt"));
processor = new Intonator(accentCart, toneCart);
}
public MaryData process(MaryData d)
throws Exception
{
List utterances = d.getUtterances();
Iterator it = utterances.iterator();
while (it.hasNext()) {
Utterance utterance = (Utterance) it.next();
processor.processUtterance(utterance);
}
MaryData output = new MaryData(outputType(), d.getLocale());
output.setUtterances(utterances);
return output;
}
public FreeTTSIntonator() {
super("Intonator", USEnglishDataTypes.FREETTS_PAUSES, USEnglishDataTypes.FREETTS_INTONATION, Locale.ENGLISH);
}
public void startup() throws Exception {
super.startup();
// Initialise FreeTTS
FreeTTSVoices.load();
CARTImpl accentCart = new CARTImpl(com.sun.speech.freetts.en.us.CMUVoice.class.getResource("int_accent_cart.txt"));
CARTImpl toneCart = new CARTImpl(com.sun.speech.freetts.en.us.CMUVoice.class.getResource("int_tone_cart.txt"));
processor = new Intonator(accentCart, toneCart);
}
public MaryData process(MaryData d) throws Exception {
List utterances = d.getUtterances();
Iterator it = utterances.iterator();
while (it.hasNext()) {
Utterance utterance = (Utterance) it.next();
processor.processUtterance(utterance);
}
MaryData output = new MaryData(outputType(), d.getLocale());
output.setUtterances(utterances);
return output;
}
}

150
marytts-languages/marytts-lang-en/src/main/java/marytts/language/en/FreeTTSPartOfSpeechTagger.java
Просмотреть файл

@ -42,99 +42,81 @@ import com.sun.speech.freetts.Item;
import com.sun.speech.freetts.Relation;
import com.sun.speech.freetts.Utterance;
/**
* Use an individual FreeTTS module for English synthesis.
*
*
* @author Marc Schröder
*/
public class FreeTTSPartOfSpeechTagger extends InternalModule
{
//private UtteranceProcessor processor;
private Logger logger;
private Map posMap;
public FreeTTSPartOfSpeechTagger()
{
super("PartOfSpeechTagger",
USEnglishDataTypes.FREETTS_WORDS,
USEnglishDataTypes.FREETTS_POS,
Locale.ENGLISH
);
}
public class FreeTTSPartOfSpeechTagger extends InternalModule {
// private UtteranceProcessor processor;
private Logger logger;
public void startup() throws Exception
{
super.startup();
this.logger = MaryUtils.getLogger("FreeTTSPOSTagger");
buildPosMap();
// Initialise FreeTTS
FreeTTSVoices.load();
//processor = new PartOfSpeechTagger();
}
private Map posMap;
public MaryData process(MaryData d)
throws Exception
{
List utterances = d.getUtterances();
Iterator it = utterances.iterator();
while (it.hasNext()) {
Utterance utterance = (Utterance) it.next();
//processor.processUtterance(utterance);
processUtterance(utterance);
}
MaryData output = new MaryData(outputType(), d.getLocale());
output.setUtterances(utterances);
return output;
}
public FreeTTSPartOfSpeechTagger() {
super("PartOfSpeechTagger", USEnglishDataTypes.FREETTS_WORDS, USEnglishDataTypes.FREETTS_POS, Locale.ENGLISH);
private void buildPosMap(){
posMap = new HashMap();
try{
String posFile =
MaryProperties.getFilename("english.freetts.posfile");
BufferedReader reader =
new BufferedReader(new FileReader(new File (posFile)));
String line = reader.readLine();
while (line!=null){
if(!(line.startsWith("***"))){
//System.out.println(line);
StringTokenizer st =
new StringTokenizer(line," ");
String word = st.nextToken();
String pos = st.nextToken();
posMap.put(word,pos);}
line = reader.readLine();
}
}catch(Exception e){
e.printStackTrace();
throw new Error("Unable to build PoS-Map");
}
}
}
private void processUtterance(Utterance utt){
logger.debug("Tagging part of speech...");
for(Item word = utt.getRelation(Relation.WORD).getHead();
word != null; word = word.getNext()){
String pos = null;
if (posMap.containsKey(word.toString())){
pos = (String) posMap.get(word.toString());
logger.debug("Assigning pos \""+pos+"\" to word \""
+word.toString()+"\"");
}
else {pos = "content";}
word.getFeatures().setString("pos",pos);
}
}
public void startup() throws Exception {
super.startup();
this.logger = MaryUtils.getLogger("FreeTTSPOSTagger");
buildPosMap();
// Initialise FreeTTS
FreeTTSVoices.load();
// processor = new PartOfSpeechTagger();
}
public MaryData process(MaryData d) throws Exception {
List utterances = d.getUtterances();
Iterator it = utterances.iterator();
while (it.hasNext()) {
Utterance utterance = (Utterance) it.next();
// processor.processUtterance(utterance);
processUtterance(utterance);
}
MaryData output = new MaryData(outputType(), d.getLocale());
output.setUtterances(utterances);
return output;
}
private void buildPosMap() {
posMap = new HashMap();
try {
String posFile = MaryProperties.getFilename("english.freetts.posfile");
BufferedReader reader = new BufferedReader(new FileReader(new File(posFile)));
String line = reader.readLine();
while (line != null) {
if (!(line.startsWith("***"))) {
// System.out.println(line);
StringTokenizer st = new StringTokenizer(line, " ");
String word = st.nextToken();
String pos = st.nextToken();
posMap.put(word, pos);
}
line = reader.readLine();
}
} catch (Exception e) {
e.printStackTrace();
throw new Error("Unable to build PoS-Map");
}
}
private void processUtterance(Utterance utt) {
logger.debug("Tagging part of speech...");
for (Item word = utt.getRelation(Relation.WORD).getHead(); word != null; word = word.getNext()) {
String pos = null;
if (posMap.containsKey(word.toString())) {
pos = (String) posMap.get(word.toString());
logger.debug("Assigning pos \"" + pos + "\" to word \"" + word.toString() + "\"");
} else {
pos = "content";
}
word.getFeatures().setString("pos", pos);
}
}
}

61
marytts-languages/marytts-lang-en/src/main/java/marytts/language/en/FreeTTSPauseGenerator.java
Просмотреть файл

@ -32,52 +32,37 @@ import com.sun.speech.freetts.Utterance;
import com.sun.speech.freetts.UtteranceProcessor;
import com.sun.speech.freetts.en.PauseGenerator;
/**
* Use an individual FreeTTS module for English synthesis.
*
*
* @author Marc Schröder
*/
public class FreeTTSPauseGenerator extends InternalModule
{
private UtteranceProcessor processor;
public class FreeTTSPauseGenerator extends InternalModule {
private UtteranceProcessor processor;
public FreeTTSPauseGenerator()
{
super("PauseGenerator",
USEnglishDataTypes.FREETTS_SEGMENTS,
USEnglishDataTypes.FREETTS_PAUSES,
Locale.ENGLISH
);
}
public void startup() throws Exception
{
super.startup();
// Initialise FreeTTS
FreeTTSVoices.load();
processor = new PauseGenerator();
}
public MaryData process(MaryData d)
throws Exception
{
List utterances = d.getUtterances();
Iterator it = utterances.iterator();
while (it.hasNext()) {
Utterance utterance = (Utterance) it.next();
processor.processUtterance(utterance);
}
MaryData output = new MaryData(outputType(), d.getLocale());
output.setUtterances(utterances);
return output;
}
public FreeTTSPauseGenerator() {
super("PauseGenerator", USEnglishDataTypes.FREETTS_SEGMENTS, USEnglishDataTypes.FREETTS_PAUSES, Locale.ENGLISH);
}
public void startup() throws Exception {
super.startup();
// Initialise FreeTTS
FreeTTSVoices.load();
processor = new PauseGenerator();
}
public MaryData process(MaryData d) throws Exception {
List utterances = d.getUtterances();
Iterator it = utterances.iterator();
while (it.hasNext()) {
Utterance utterance = (Utterance) it.next();
processor.processUtterance(utterance);
}
MaryData output = new MaryData(outputType(), d.getLocale());
output.setUtterances(utterances);
return output;
}
}

64
marytts-languages/marytts-lang-en/src/main/java/marytts/language/en/FreeTTSPhraser.java
Просмотреть файл

@ -33,54 +33,38 @@ import com.sun.speech.freetts.UtteranceProcessor;
import com.sun.speech.freetts.cart.CARTImpl;
import com.sun.speech.freetts.cart.Phraser;
/**
* Use an individual FreeTTS module for English synthesis.
*
*
* @author Marc Schröder
*/
public class FreeTTSPhraser extends InternalModule
{
private UtteranceProcessor processor;
public class FreeTTSPhraser extends InternalModule {
private UtteranceProcessor processor;
public FreeTTSPhraser()
{
super("Phraser",
USEnglishDataTypes.FREETTS_POS,
USEnglishDataTypes.FREETTS_PHRASES,
Locale.ENGLISH
);
}
public void startup() throws Exception
{
super.startup();
// Initialise FreeTTS
FreeTTSVoices.load();
CARTImpl phrasingCart = new CARTImpl(
com.sun.speech.freetts.en.us.CMUVoice.class.getResource("phrasing_cart.txt"));
processor = new Phraser(phrasingCart);
}
public MaryData process(MaryData d)
throws Exception
{
List utterances = d.getUtterances();
Iterator it = utterances.iterator();
while (it.hasNext()) {
Utterance utterance = (Utterance) it.next();
processor.processUtterance(utterance);
}
MaryData output = new MaryData(outputType(), d.getLocale());
output.setUtterances(utterances);
return output;
}
public FreeTTSPhraser() {
super("Phraser", USEnglishDataTypes.FREETTS_POS, USEnglishDataTypes.FREETTS_PHRASES, Locale.ENGLISH);
}
public void startup() throws Exception {
super.startup();
// Initialise FreeTTS
FreeTTSVoices.load();
CARTImpl phrasingCart = new CARTImpl(com.sun.speech.freetts.en.us.CMUVoice.class.getResource("phrasing_cart.txt"));
processor = new Phraser(phrasingCart);
}
public MaryData process(MaryData d) throws Exception {
List utterances = d.getUtterances();
Iterator it = utterances.iterator();
while (it.hasNext()) {
Utterance utterance = (Utterance) it.next();
processor.processUtterance(utterance);
}
MaryData output = new MaryData(outputType(), d.getLocale());
output.setUtterances(utterances);
return output;
}
}

126
marytts-languages/marytts-lang-en/src/main/java/marytts/language/en/FreeTTSTextToTokens.java
Просмотреть файл

@ -44,87 +44,67 @@ import com.sun.speech.freetts.Tokenizer;
import com.sun.speech.freetts.Utterance;
import com.sun.speech.freetts.en.us.USEnglish;
/**
* Use an individual FreeTTS module for English synthesis.
*
*
* @author Marc Schröder
*/
public class FreeTTSTextToTokens extends InternalModule
{
public FreeTTSTextToTokens()
{
super("TextToTokens",
MaryDataType.TEXT,
USEnglishDataTypes.FREETTS_TOKENS,
Locale.ENGLISH
);
}
public class FreeTTSTextToTokens extends InternalModule {
public FreeTTSTextToTokens() {
super("TextToTokens", MaryDataType.TEXT, USEnglishDataTypes.FREETTS_TOKENS, Locale.ENGLISH);
}
public void startup() throws Exception
{
super.startup();
// Initialise FreeTTS
FreeTTSVoices.load();
}
public MaryData process(MaryData d)
throws Exception
{
String text = d.getPlainText();
// create a basic utterance from text
Tokenizer tokenizer = new com.sun.speech.freetts.en.TokenizerImpl();
tokenizer.setWhitespaceSymbols(USEnglish.WHITESPACE_SYMBOLS);
tokenizer.setSingleCharSymbols(USEnglish.SINGLE_CHAR_SYMBOLS);
tokenizer.setPrepunctuationSymbols(USEnglish.PREPUNCTUATION_SYMBOLS);
tokenizer.setPostpunctuationSymbols(USEnglish.PUNCTUATION_SYMBOLS);
tokenizer.setInputText(text);
ArrayList utteranceList = new ArrayList();
Token savedToken = null;
boolean first = true;
while (tokenizer.hasMoreTokens()) {
// Fill a new Utterance:
ArrayList tokenList = new ArrayList();
Utterance utterance = null;
if (savedToken != null) {
tokenList.add(savedToken);
savedToken = null;
}
while (tokenizer.hasMoreTokens()) {
Token token = tokenizer.getNextToken();
if ((token.getWord().length() == 0) ||
(tokenList.size() > 500) ||
tokenizer.isBreak()) {
savedToken = token;
break;
}
tokenList.add(token);
}
marytts.modules.synthesis.Voice maryVoice =
d.getDefaultVoice();
if (maryVoice == null ||
!maryVoice.getLocale().equals(Locale.US)) {
maryVoice = marytts.modules.synthesis.Voice.
getDefaultVoice(Locale.US);
}
utterance = new Utterance
(FreeTTSVoices.getFreeTTSVoice(maryVoice), tokenList);
//utterance.setSpeakable(speakable);
utterance.setFirst(first);
first = false;
utterance.setLast(!tokenizer.hasMoreTokens());
utteranceList.add(utterance);
}
MaryData output = new MaryData(outputType(), d.getLocale());
output.setUtterances(utteranceList);
return output;
}
public void startup() throws Exception {
super.startup();
// Initialise FreeTTS
FreeTTSVoices.load();
}
public MaryData process(MaryData d) throws Exception {
String text = d.getPlainText();
// create a basic utterance from text
Tokenizer tokenizer = new com.sun.speech.freetts.en.TokenizerImpl();
tokenizer.setWhitespaceSymbols(USEnglish.WHITESPACE_SYMBOLS);
tokenizer.setSingleCharSymbols(USEnglish.SINGLE_CHAR_SYMBOLS);
tokenizer.setPrepunctuationSymbols(USEnglish.PREPUNCTUATION_SYMBOLS);
tokenizer.setPostpunctuationSymbols(USEnglish.PUNCTUATION_SYMBOLS);
tokenizer.setInputText(text);
ArrayList utteranceList = new ArrayList();
Token savedToken = null;
boolean first = true;
while (tokenizer.hasMoreTokens()) {
// Fill a new Utterance:
ArrayList tokenList = new ArrayList();
Utterance utterance = null;
if (savedToken != null) {
tokenList.add(savedToken);
savedToken = null;
}
while (tokenizer.hasMoreTokens()) {
Token token = tokenizer.getNextToken();
if ((token.getWord().length() == 0) || (tokenList.size() > 500) || tokenizer.isBreak()) {
savedToken = token;
break;
}
tokenList.add(token);
}
marytts.modules.synthesis.Voice maryVoice = d.getDefaultVoice();
if (maryVoice == null || !maryVoice.getLocale().equals(Locale.US)) {
maryVoice = marytts.modules.synthesis.Voice.getDefaultVoice(Locale.US);
}
utterance = new Utterance(FreeTTSVoices.getFreeTTSVoice(maryVoice), tokenList);
// utterance.setSpeakable(speakable);
utterance.setFirst(first);
first = false;
utterance.setLast(!tokenizer.hasMoreTokens());
utteranceList.add(utterance);
}
MaryData output = new MaryData(outputType(), d.getLocale());
output.setUtterances(utteranceList);
return output;
}
}

69
marytts-languages/marytts-lang-en/src/main/java/marytts/language/en/FreeTTSTokenToWords.java
Просмотреть файл

@ -46,57 +46,40 @@ import com.sun.speech.freetts.en.us.PrefixFSM;
import com.sun.speech.freetts.en.us.PronounceableFSM;
import com.sun.speech.freetts.en.us.SuffixFSM;
/**
* Use an individual FreeTTS module for English synthesis.
*
*
* @author Marc Schröder
*/
public class FreeTTSTokenToWords extends InternalModule
{
private UtteranceProcessor processor;
public class FreeTTSTokenToWords extends InternalModule {
private UtteranceProcessor processor;
public FreeTTSTokenToWords()
{
super("TokenToWords",
USEnglishDataTypes.FREETTS_TOKENS,
USEnglishDataTypes.FREETTS_WORDS,
Locale.ENGLISH
);
}
public void startup() throws Exception
{
super.startup();
// Initialise FreeTTS
FreeTTSVoices.load();
CARTImpl numbersCart = new CARTImpl
(com.sun.speech.freetts.en.us.CMUVoice.class.getResource("nums_cart.txt"));
PronounceableFSM prefixFSM = new PrefixFSM
(com.sun.speech.freetts.en.us.CMUVoice.class.getResource("prefix_fsm.txt"));
PronounceableFSM suffixFSM = new SuffixFSM
(com.sun.speech.freetts.en.us.CMUVoice.class.getResource("suffix_fsm.txt"));
processor = new TokenToWords(numbersCart, prefixFSM, suffixFSM);
}
public MaryData process(MaryData d)
throws Exception
{
List utterances = d.getUtterances();
Iterator it = utterances.iterator();
while (it.hasNext()) {
Utterance utterance = (Utterance) it.next();
processor.processUtterance(utterance);
}
MaryData output = new MaryData(outputType(), d.getLocale());
output.setUtterances(utterances);
return output;
}
public FreeTTSTokenToWords() {
super("TokenToWords", USEnglishDataTypes.FREETTS_TOKENS, USEnglishDataTypes.FREETTS_WORDS, Locale.ENGLISH);
}
public void startup() throws Exception {
super.startup();
// Initialise FreeTTS
FreeTTSVoices.load();
CARTImpl numbersCart = new CARTImpl(com.sun.speech.freetts.en.us.CMUVoice.class.getResource("nums_cart.txt"));
PronounceableFSM prefixFSM = new PrefixFSM(com.sun.speech.freetts.en.us.CMUVoice.class.getResource("prefix_fsm.txt"));
PronounceableFSM suffixFSM = new SuffixFSM(com.sun.speech.freetts.en.us.CMUVoice.class.getResource("suffix_fsm.txt"));
processor = new TokenToWords(numbersCart, prefixFSM, suffixFSM);
}
public MaryData process(MaryData d) throws Exception {
List utterances = d.getUtterances();
Iterator it = utterances.iterator();
while (it.hasNext()) {
Utterance utterance = (Utterance) it.next();
processor.processUtterance(utterance);
}
MaryData output = new MaryData(outputType(), d.getLocale());
output.setUtterances(utterances);
return output;
}
}

117
marytts-languages/marytts-lang-en/src/main/java/marytts/language/en/JTokeniser.java
Просмотреть файл

@ -34,75 +34,68 @@ import org.w3c.dom.traversal.DocumentTraversal;
import org.w3c.dom.traversal.NodeFilter;
import org.w3c.dom.traversal.NodeIterator;
/**
*
*
* @author Marc Schröder
*/
public class JTokeniser extends marytts.modules.JTokeniser {
/**
/**
*
*/
public JTokeniser() {
super(MaryDataType.RAWMARYXML,
MaryDataType.TOKENS,
Locale.ENGLISH);
}
public JTokeniser() {
super(MaryDataType.RAWMARYXML, MaryDataType.TOKENS, Locale.ENGLISH);
}
public MaryData process(MaryData d) throws Exception
{
MaryData result = super.process(d);
normaliseToAscii(result);
propagateForceAccent(result);
return result;
}
protected void normaliseToAscii(MaryData d)
{
Document doc = d.getDocument();
NodeIterator ni = ((DocumentTraversal) doc).createNodeIterator(doc,
NodeFilter.SHOW_ELEMENT, new NameNodeFilter(MaryXML.TOKEN), false);
Element t = null;
while ((t = (Element) ni.nextNode()) != null) {
String s = MaryDomUtils.tokenText(t);
String normalised = MaryUtils.normaliseUnicodeLetters(s, Locale.ENGLISH);
if (!s.equals(normalised)) {
MaryDomUtils.setTokenText(t, normalised);
}
}
}
public MaryData process(MaryData d) throws Exception {
MaryData result = super.process(d);
normaliseToAscii(result);
propagateForceAccent(result);
return result;
}
/**
* In current FreeTTS code, prosody elements get lost. So remember
* at least the force-accent element on individual tokens:
* @param d
*/
protected void propagateForceAccent(MaryData d)
{
Document doc = d.getDocument();
NodeIterator prosodyNI = ((DocumentTraversal)doc).createNodeIterator(doc,
NodeFilter.SHOW_ELEMENT, new NameNodeFilter(MaryXML.PROSODY), false);
Element prosody = null;
while ((prosody = (Element) prosodyNI.nextNode()) != null) {
if (prosody.hasAttribute("force-accent")) {
String forceAccent = prosody.getAttribute("force-accent");
String accent = null;
if (forceAccent.equals("none")) {
accent = "none";
} else {
accent = "unknown";
}
NodeIterator tNI = ((DocumentTraversal)doc).createNodeIterator(prosody,
NodeFilter.SHOW_ELEMENT, new NameNodeFilter(MaryXML.TOKEN), false);
Element t = null;
while ((t = (Element) tNI.nextNode()) != null) {
if (!t.hasAttribute("accent")) {
t.setAttribute("accent", accent);
}
} // while t
}
} // while prosody
}
protected void normaliseToAscii(MaryData d) {
Document doc = d.getDocument();
NodeIterator ni = ((DocumentTraversal) doc).createNodeIterator(doc, NodeFilter.SHOW_ELEMENT, new NameNodeFilter(
MaryXML.TOKEN), false);
Element t = null;
while ((t = (Element) ni.nextNode()) != null) {
String s = MaryDomUtils.tokenText(t);
String normalised = MaryUtils.normaliseUnicodeLetters(s, Locale.ENGLISH);
if (!s.equals(normalised)) {
MaryDomUtils.setTokenText(t, normalised);
}
}
}
/**
* In current FreeTTS code, prosody elements get lost. So remember at least the force-accent element on individual tokens:
*
* @param d
*/
protected void propagateForceAccent(MaryData d) {
Document doc = d.getDocument();
NodeIterator prosodyNI = ((DocumentTraversal) doc).createNodeIterator(doc, NodeFilter.SHOW_ELEMENT, new NameNodeFilter(
MaryXML.PROSODY), false);
Element prosody = null;
while ((prosody = (Element) prosodyNI.nextNode()) != null) {
if (prosody.hasAttribute("force-accent")) {
String forceAccent = prosody.getAttribute("force-accent");
String accent = null;
if (forceAccent.equals("none")) {
accent = "none";
} else {
accent = "unknown";
}
NodeIterator tNI = ((DocumentTraversal) doc).createNodeIterator(prosody, NodeFilter.SHOW_ELEMENT,
new NameNodeFilter(MaryXML.TOKEN), false);
Element t = null;
while ((t = (Element) tNI.nextNode()) != null) {
if (!t.hasAttribute("accent")) {
t.setAttribute("accent", accent);
}
} // while t
}
} // while prosody
}
}

915
marytts-languages/marytts-lang-en/src/main/java/marytts/language/en/NumberExpander.java
Просмотреть файл

@ -15,514 +15,445 @@ import com.sun.speech.freetts.Item;
import com.sun.speech.freetts.util.Utilities;
/**
* Expands Strings containing digits characters into
* a list of words representing those digits.
*
* It translates the following code from flite:
* <code>lang/usEnglish/us_expand.c</code>
* Expands Strings containing digits characters into a list of words representing those digits.
*
* It translates the following code from flite: <code>lang/usEnglish/us_expand.c</code>
*/
public class NumberExpander {
private static final String[] digit2num = {
"zero",
"one",
"two",
"three",
"four",
"five",
"six",
"seven",
"eight",
"nine" };
private static final String[] digit2teen = {
"ten", /* shouldn't get called */
"eleven",
"twelve",
"thirteen",
"fourteen",
"fifteen",
"sixteen",
"seventeen",
"eighteen",
"nineteen" };
private static final String[] digit2enty = {
"zero", /* shouldn't get called */
"ten",
"twenty",
"thirty",
"forty",
"fifty",
"sixty",
"seventy",
"eighty",
"ninety" };
private static final String[] ord2num = {
"zeroth",
"first",
"second",
"third",
"fourth",
"fifth",
"sixth",
"seventh",
"eighth",
"ninth" };
private static final String[] ord2teen = {
"tenth", /* shouldn't get called */
"eleventh",
"twelfth",
"thirteenth",
"fourteenth",
"fifteenth",
"sixteenth",
"seventeenth",
"eighteenth",
"nineteenth" };
private static final String[] ord2enty = {
"zeroth", /* shouldn't get called */
"tenth",
"twentieth",
"thirtieth",
"fortieth",
"fiftieth",
"sixtieth",
"seventieth",
"eightieth",
"ninetieth" };
private static final String[] digit2num = { "zero", "one", "two", "three", "four", "five", "six", "seven", "eight", "nine" };
/**
* Unconstructable
*/
private NumberExpander() {
}
private static final String[] digit2teen = { "ten", /* shouldn't get called */
"eleven", "twelve", "thirteen", "fourteen", "fifteen", "sixteen", "seventeen", "eighteen", "nineteen" };
private static final String[] digit2enty = { "zero", /* shouldn't get called */
"ten", "twenty", "thirty", "forty", "fifty", "sixty", "seventy", "eighty", "ninety" };
/**
* Expands a digit string into a list of English words of those digits.
* For example, "1234" expands to "one two three four"
*
* @param numberString the digit string to expand.
* @param wordRelation words are added to this Relation
*/
public static void expandNumber(String numberString,
WordRelation wordRelation, Item tokenItem) {
int numDigits = numberString.length();
if (numDigits == 0) {
// wordRelation = null;
} else if (numDigits == 1) {
expandDigits(numberString, wordRelation, tokenItem);
} else if (numDigits == 2) {
expand2DigitNumber(numberString, wordRelation, tokenItem);
} else if (numDigits == 3) {
expand3DigitNumber(numberString, wordRelation, tokenItem);
} else if (numDigits < 7) {
expandBelow7DigitNumber(numberString, wordRelation, tokenItem);
} else if (numDigits < 10) {
expandBelow10DigitNumber(numberString, wordRelation, tokenItem);
} else if (numDigits < 13) {
expandBelow13DigitNumber(numberString, wordRelation, tokenItem);
} else {
expandDigits(numberString, wordRelation, tokenItem);
private static final String[] ord2num = { "zeroth", "first", "second", "third", "fourth", "fifth", "sixth", "seventh",
"eighth", "ninth" };
private static final String[] ord2teen = { "tenth", /* shouldn't get called */
"eleventh", "twelfth", "thirteenth", "fourteenth", "fifteenth", "sixteenth", "seventeenth", "eighteenth", "nineteenth" };
private static final String[] ord2enty = { "zeroth", /* shouldn't get called */
"tenth", "twentieth", "thirtieth", "fortieth", "fiftieth", "sixtieth", "seventieth", "eightieth", "ninetieth" };
/**
* Unconstructable
*/
private NumberExpander() {
}
}
/**
* Expands a digit string into a list of English words of those digits. For example, "1234" expands to "one two three four"
*
* @param numberString
* the digit string to expand.
* @param wordRelation
* words are added to this Relation
*/
public static void expandNumber(String numberString, WordRelation wordRelation, Item tokenItem) {
int numDigits = numberString.length();
/**
* Expands a two-digit string into a list of English words.
*
* @param numberString the string which is the number to expand
* @param wordRelation words are added to this Relation
*/
private static void expand2DigitNumber(String numberString,
WordRelation wordRelation, Item tokenItem) {
if (numberString.charAt(0) == '0') {
// numberString is "0X"
if (numberString.charAt(1) == '0') {
// numberString is "00", do nothing
} else {
// numberString is "01", "02" ...
String number = digit2num[numberString.charAt(1)-'0'];
wordRelation.addWord(tokenItem,number);
}
} else if (numberString.charAt(1) == '0') {
// numberString is "10", "20", ...
String number = digit2enty[numberString.charAt(0)-'0'];
wordRelation.addWord(tokenItem,number);
} else if (numberString.charAt(0) == '1') {
// numberString is "11", "12", ..., "19"
String number = digit2teen[numberString.charAt(1)-'0'];
wordRelation.addWord(tokenItem,number);
} else {
// numberString is "2X", "3X", ...
String enty = digit2enty[numberString.charAt(0)-'0'];
wordRelation.addWord(tokenItem,enty);
expandDigits(numberString.substring(1,numberString.length()),
wordRelation, tokenItem);
}
}
/**
* Expands a three-digit string into a list of English words.
*
* @param numberString the string which is the number to expand
* @param wordRelation words are added to this Relation
*/
private static void expand3DigitNumber(String numberString,
WordRelation wordRelation,Item tokenItem) {
if (numberString.charAt(0) == '0') {
expandNumberAt(numberString, 1, wordRelation, tokenItem);
} else {
String hundredDigit = digit2num[numberString.charAt(0)-'0'];
wordRelation.addWord(tokenItem,hundredDigit);
wordRelation.addWord(tokenItem,"hundred");
expandNumberAt(numberString, 1, wordRelation, tokenItem);
}
}
/**
* Expands a string that is a 4 to 6 digits number into a list
* of English words. For example, "333000" into "three hundred
* and thirty-three thousand".
*
* @param numberString the string which is the number to expand
* @param wordRelation words are added to this Relation
*/
private static void expandBelow7DigitNumber(String numberString,
WordRelation wordRelation,Item tokenItem) {
expandLargeNumber(numberString, "thousand", 3, wordRelation, tokenItem);
}
/**
* Expands a string that is a 7 to 9 digits number into a list
* of English words. For example, "19000000" into nineteen million.
*
* @param numberString the string which is the number to expand
* @param wordRelation words are added to this Relation
*/
private static void expandBelow10DigitNumber(String numberString,
WordRelation wordRelation,Item tokenItem) {
expandLargeNumber(numberString, "million", 6, wordRelation, tokenItem);
}
/**
* Expands a string that is a 10 to 12 digits number into a list
* of English words. For example, "27000000000" into twenty-seven
* billion.
*
* @param numberString the string which is the number to expand
* @param wordRelation words are added to this Relation
*/
private static void expandBelow13DigitNumber(String numberString,
WordRelation wordRelation,Item tokenItem) {
expandLargeNumber(numberString, "billion", 9, wordRelation, tokenItem);
}
/**
* Expands a string that is a number longer than 3 digits into a list
* of English words. For example, "1000" into one thousand.
*
* @param numberString the string which is the number to expand
* @param order either "thousand", "million", or "billion"
* @param numberZeroes the number of zeroes, depending on the order, so
* its either 3, 6, or 9
* @param wordRelation words are added to this Relation
*/
private static void expandLargeNumber(String numberString,
String order,
int numberZeroes,
WordRelation wordRelation,Item tokenItem) {
int numberDigits = numberString.length();
// parse out the prefix, e.g., "113" in "113,000"
int i = numberDigits - numberZeroes;
String part = numberString.substring(0, i);
// get how many thousands/millions/billions
Item oldTail = wordRelation.getTail();
expandNumber(part, wordRelation, tokenItem);
if (wordRelation.getTail() == oldTail) {
expandNumberAt(numberString, i, wordRelation,tokenItem);
} else {
wordRelation.addWord(tokenItem,order);
expandNumberAt(numberString, i, wordRelation,tokenItem);
}
}
/**
* Returns the number string list of the given string starting at
* the given index. E.g., expandNumberAt("1100", 1) gives "one hundred"
*
* @param numberString the string which is the number to expand
* @param startIndex the starting position
* @param wordRelation words are added to this Relation
*/
private static void expandNumberAt(String numberString,
int startIndex,
WordRelation wordRelation,Item tokenItem) {
expandNumber(numberString.substring(startIndex,numberString.length()),
wordRelation,tokenItem);
}
/**
* Expands given token to list of words pronouncing it as digits
*
* @param numberString the string which is the number to expand
* @param wordRelation words are added to this Relation
*/
public static void expandDigits(String numberString,
WordRelation wordRelation,Item tokenItem) {
int numberDigits = numberString.length();
for (int i = 0; i < numberDigits; i++) {
char digit = numberString.charAt(i);
if (isDigit(digit)) {
wordRelation.addWord(tokenItem,digit2num[numberString.charAt(i)-'0']);
} else {
wordRelation.addWord(tokenItem,"umpty");
}
}
}
/**
* Expands the digit string of an ordinal number.
*
* @param rawNumberString the string which is the number to expand
* @param wordRelation words are added to this Relation
*/
public static void expandOrdinal(String rawNumberString,
WordRelation wordRelation,Item tokenItem) {
// remove all ','s from the raw number string
String numberString = Utilities.deleteChar(rawNumberString, ',');
expandNumber(numberString, wordRelation,tokenItem);
// get the last in the list of number strings
Item lastItem = wordRelation.getTail();
if (lastItem != null) {
FeatureSet featureSet = lastItem.getFeatures();
String lastNumber = featureSet.getString("name");
String ordinal = findMatchInArray(lastNumber, digit2num, ord2num);
if (ordinal == null) {
ordinal = findMatchInArray(lastNumber, digit2teen, ord2teen);
}
if (ordinal == null) {
ordinal = findMatchInArray(lastNumber, digit2enty, ord2enty);
}
if (lastNumber.equals("hundred")) {
ordinal = "hundredth";
} else if (lastNumber.equals("thousand")) {
ordinal = "thousandth";
} else if (lastNumber.equals("billion")) {
ordinal = "billionth";
}
// if there was an ordinal, set the last element of the list
// to that ordinal; otherwise, don't do anything
if (ordinal != null) {
wordRelation.setLastWord(ordinal);
}
}
}
/**
* Finds a match of the given string in the given array,
* and returns the element at the same index in the returnInArray
*
* @param strToMatch the string to match
* @param matchInArray the source array
* @param returnInArray the return array
*
* @return an element in returnInArray, or <code>null</code>
* if a match is not found
*/
private static String findMatchInArray(String strToMatch,
String[] matchInArray,
String[] returnInArray) {
for (int i = 0; i < matchInArray.length; i++) {
if (strToMatch.equals(matchInArray[i])) {
if (i < returnInArray.length) {
return returnInArray[i];
if (numDigits == 0) {
// wordRelation = null;
} else if (numDigits == 1) {
expandDigits(numberString, wordRelation, tokenItem);
} else if (numDigits == 2) {
expand2DigitNumber(numberString, wordRelation, tokenItem);
} else if (numDigits == 3) {
expand3DigitNumber(numberString, wordRelation, tokenItem);
} else if (numDigits < 7) {
expandBelow7DigitNumber(numberString, wordRelation, tokenItem);
} else if (numDigits < 10) {
expandBelow10DigitNumber(numberString, wordRelation, tokenItem);
} else if (numDigits < 13) {
expandBelow13DigitNumber(numberString, wordRelation, tokenItem);
} else {
return null;
expandDigits(numberString, wordRelation, tokenItem);
}
}
}
return null;
}
/**
* Expands the given number string as pairs as in years or IDs
*
* @param numberString the string which is the number to expand
* @param wordRelation words are added to this Relation
*/
public static void expandID(String numberString, WordRelation wordRelation, Item tokenItem) {
int numberDigits = numberString.length();
if ((numberDigits == 4) &&
(numberString.charAt(2) == '0') &&
(numberString.charAt(3) == '0')) {
if (numberString.charAt(1) == '0') { // e.g. 2000, 3000
expandNumber(numberString, wordRelation,tokenItem);
} else {
expandNumber(numberString.substring(0,2), wordRelation,tokenItem);
wordRelation.addWord(tokenItem,"hundred");
}
} else if ((numberDigits == 2) && (numberString.charAt(0) == '0')) {
wordRelation.addWord(tokenItem,"oh");
expandDigits(numberString.substring(1,2), wordRelation,tokenItem);
} else if ((numberDigits == 4 &&
numberString.charAt(1) == '0') ||
numberDigits < 3) {
expandNumber(numberString, wordRelation,tokenItem);
} else if (numberDigits % 2 == 1) {
String firstDigit = digit2num[numberString.charAt(0)-'0'];
wordRelation.addWord(tokenItem,firstDigit);
expandID(numberString.substring(1,numberDigits), wordRelation,tokenItem);
} else {
expandNumber(numberString.substring(0,2), wordRelation,tokenItem);
expandID(numberString.substring(2,numberDigits), wordRelation,tokenItem);
}
}
/**
* Expands the given number string as a real number.
*
* @param numberString the string which is the real number to expand
* @param wordRelation words are added to this Relation
*/
public static void expandReal(String numberString, WordRelation wordRelation,Item tokenItem) {
int stringLength = numberString.length();
int position;
if (numberString.charAt(0) == '-') {
// negative real numbers
wordRelation.addWord(tokenItem,"minus");
expandReal(numberString.substring(1, stringLength), wordRelation,tokenItem);
} else if (numberString.charAt(0) == '+') {
// prefixed with a '+'
wordRelation.addWord(tokenItem,"plus");
expandReal(numberString.substring(1, stringLength), wordRelation,tokenItem);
} else if ((position = numberString.indexOf('e')) != -1 ||
(position = numberString.indexOf('E')) != -1) {
// numbers with 'E' or 'e'
expandReal(numberString.substring(0, position), wordRelation,tokenItem);
wordRelation.addWord(tokenItem,"e");
expandReal(numberString.substring(position + 1), wordRelation,tokenItem);
} else if ((position = numberString.indexOf('.')) != -1) {
// numbers with '.'
String beforeDot = numberString.substring(0, position);
if (beforeDot.length() > 0) {
expandReal(beforeDot, wordRelation,tokenItem);
}
wordRelation.addWord(tokenItem,"point");
String afterDot = numberString.substring(position + 1);
if (afterDot.length() > 0) {
expandDigits(afterDot, wordRelation,tokenItem);
}
} else {
// everything else
expandNumber(numberString, wordRelation,tokenItem);
}
}
/**
* Expands the given string of letters as a list of single char symbols.
*
* @param letters the string of letters to expand
* @param wordRelation words are added to this Relation
*/
public static void expandLetters(String letters,
WordRelation wordRelation,Item tokenItem) {
letters = letters.toLowerCase();
char c;
for (int i = 0; i < letters.length(); i++) {
// if this is a number
c = letters.charAt(i);
if (isDigit(c)) {
wordRelation.addWord(tokenItem,digit2num[c-'0']);
} else if (letters.equals("a")) {
wordRelation.addWord(tokenItem,"_a");
} else {
wordRelation.addWord(tokenItem,String.valueOf(c));
}
}
}
/**
* Returns the integer value of the given string of Roman numerals.
*
* @param roman the string of Roman numbers
*
* @return the integer value
*/
public static int expandRoman(String roman) {
int value = 0;
for (int p = 0; p < roman.length(); p++) {
char c = roman.charAt(p);
if (c == 'X') {
value += 10;
} else if (c == 'V') {
value += 5;
} else if (c == 'I') {
if (p+1 < roman.length()) {
char p1 = roman.charAt(p+1);
if (p1 == 'V') {
value += 4;
p++;
} else if (p1 == 'X') {
value += 9;
p++;
} else {
value += 1;
}
/**
* Expands a two-digit string into a list of English words.
*
* @param numberString
* the string which is the number to expand
* @param wordRelation
* words are added to this Relation
*/
private static void expand2DigitNumber(String numberString, WordRelation wordRelation, Item tokenItem) {
if (numberString.charAt(0) == '0') {
// numberString is "0X"
if (numberString.charAt(1) == '0') {
// numberString is "00", do nothing
} else {
// numberString is "01", "02" ...
String number = digit2num[numberString.charAt(1) - '0'];
wordRelation.addWord(tokenItem, number);
}
} else if (numberString.charAt(1) == '0') {
// numberString is "10", "20", ...
String number = digit2enty[numberString.charAt(0) - '0'];
wordRelation.addWord(tokenItem, number);
} else if (numberString.charAt(0) == '1') {
// numberString is "11", "12", ..., "19"
String number = digit2teen[numberString.charAt(1) - '0'];
wordRelation.addWord(tokenItem, number);
} else {
value += 1;
// numberString is "2X", "3X", ...
String enty = digit2enty[numberString.charAt(0) - '0'];
wordRelation.addWord(tokenItem, enty);
expandDigits(numberString.substring(1, numberString.length()), wordRelation, tokenItem);
}
}
}
return value;
}
/**
* Expands a three-digit string into a list of English words.
*
* @param numberString
* the string which is the number to expand
* @param wordRelation
* words are added to this Relation
*/
private static void expand3DigitNumber(String numberString, WordRelation wordRelation, Item tokenItem) {
if (numberString.charAt(0) == '0') {
expandNumberAt(numberString, 1, wordRelation, tokenItem);
} else {
String hundredDigit = digit2num[numberString.charAt(0) - '0'];
wordRelation.addWord(tokenItem, hundredDigit);
wordRelation.addWord(tokenItem, "hundred");
expandNumberAt(numberString, 1, wordRelation, tokenItem);
}
}
/**
* Returns true if the given character is a digit (0-9 only).
*
* @param ch the character to test
*
* @return true or false
*/
public static boolean isDigit(char ch) {
return ('0' <= ch && ch <= '9');
}
/**
* Expands a string that is a 4 to 6 digits number into a list of English words. For example, "333000" into "three hundred and
* thirty-three thousand".
*
* @param numberString
* the string which is the number to expand
* @param wordRelation
* words are added to this Relation
*/
private static void expandBelow7DigitNumber(String numberString, WordRelation wordRelation, Item tokenItem) {
expandLargeNumber(numberString, "thousand", 3, wordRelation, tokenItem);
}
/**
* Expands a string that is a 7 to 9 digits number into a list of English words. For example, "19000000" into nineteen
* million.
*
* @param numberString
* the string which is the number to expand
* @param wordRelation
* words are added to this Relation
*/
private static void expandBelow10DigitNumber(String numberString, WordRelation wordRelation, Item tokenItem) {
expandLargeNumber(numberString, "million", 6, wordRelation, tokenItem);
}
/**
* Expands a string that is a 10 to 12 digits number into a list of English words. For example, "27000000000" into
* twenty-seven billion.
*
* @param numberString
* the string which is the number to expand
* @param wordRelation
* words are added to this Relation
*/
private static void expandBelow13DigitNumber(String numberString, WordRelation wordRelation, Item tokenItem) {
expandLargeNumber(numberString, "billion", 9, wordRelation, tokenItem);
}
/**
* Expands a string that is a number longer than 3 digits into a list of English words. For example, "1000" into one thousand.
*
* @param numberString
* the string which is the number to expand
* @param order
* either "thousand", "million", or "billion"
* @param numberZeroes
* the number of zeroes, depending on the order, so its either 3, 6, or 9
* @param wordRelation
* words are added to this Relation
*/
private static void expandLargeNumber(String numberString, String order, int numberZeroes, WordRelation wordRelation,
Item tokenItem) {
int numberDigits = numberString.length();
// parse out the prefix, e.g., "113" in "113,000"
int i = numberDigits - numberZeroes;
String part = numberString.substring(0, i);
// get how many thousands/millions/billions
Item oldTail = wordRelation.getTail();
expandNumber(part, wordRelation, tokenItem);
if (wordRelation.getTail() == oldTail) {
expandNumberAt(numberString, i, wordRelation, tokenItem);
} else {
wordRelation.addWord(tokenItem, order);
expandNumberAt(numberString, i, wordRelation, tokenItem);
}
}
/**
* Returns the number string list of the given string starting at the given index. E.g., expandNumberAt("1100", 1) gives
* "one hundred"
*
* @param numberString
* the string which is the number to expand
* @param startIndex
* the starting position
* @param wordRelation
* words are added to this Relation
*/
private static void expandNumberAt(String numberString, int startIndex, WordRelation wordRelation, Item tokenItem) {
expandNumber(numberString.substring(startIndex, numberString.length()), wordRelation, tokenItem);
}
/**
* Expands given token to list of words pronouncing it as digits
*
* @param numberString
* the string which is the number to expand
* @param wordRelation
* words are added to this Relation
*/
public static void expandDigits(String numberString, WordRelation wordRelation, Item tokenItem) {
int numberDigits = numberString.length();
for (int i = 0; i < numberDigits; i++) {
char digit = numberString.charAt(i);
if (isDigit(digit)) {
wordRelation.addWord(tokenItem, digit2num[numberString.charAt(i) - '0']);
} else {
wordRelation.addWord(tokenItem, "umpty");
}
}
}
/**
* Expands the digit string of an ordinal number.
*
* @param rawNumberString
* the string which is the number to expand
* @param wordRelation
* words are added to this Relation
*/
public static void expandOrdinal(String rawNumberString, WordRelation wordRelation, Item tokenItem) {
// remove all ','s from the raw number string
String numberString = Utilities.deleteChar(rawNumberString, ',');
expandNumber(numberString, wordRelation, tokenItem);
// get the last in the list of number strings
Item lastItem = wordRelation.getTail();
if (lastItem != null) {
FeatureSet featureSet = lastItem.getFeatures();
String lastNumber = featureSet.getString("name");
String ordinal = findMatchInArray(lastNumber, digit2num, ord2num);
if (ordinal == null) {
ordinal = findMatchInArray(lastNumber, digit2teen, ord2teen);
}
if (ordinal == null) {
ordinal = findMatchInArray(lastNumber, digit2enty, ord2enty);
}
if (lastNumber.equals("hundred")) {
ordinal = "hundredth";
} else if (lastNumber.equals("thousand")) {
ordinal = "thousandth";
} else if (lastNumber.equals("billion")) {
ordinal = "billionth";
}
// if there was an ordinal, set the last element of the list
// to that ordinal; otherwise, don't do anything
if (ordinal != null) {
wordRelation.setLastWord(ordinal);
}
}
}
/**
* Finds a match of the given string in the given array, and returns the element at the same index in the returnInArray
*
* @param strToMatch
* the string to match
* @param matchInArray
* the source array
* @param returnInArray
* the return array
*
* @return an element in returnInArray, or <code>null</code> if a match is not found
*/
private static String findMatchInArray(String strToMatch, String[] matchInArray, String[] returnInArray) {
for (int i = 0; i < matchInArray.length; i++) {
if (strToMatch.equals(matchInArray[i])) {
if (i < returnInArray.length) {
return returnInArray[i];
} else {
return null;
}
}
}
return null;
}
/**
* Expands the given number string as pairs as in years or IDs
*
* @param numberString
* the string which is the number to expand
* @param wordRelation
* words are added to this Relation
*/
public static void expandID(String numberString, WordRelation wordRelation, Item tokenItem) {
int numberDigits = numberString.length();
if ((numberDigits == 4) && (numberString.charAt(2) == '0') && (numberString.charAt(3) == '0')) {
if (numberString.charAt(1) == '0') { // e.g. 2000, 3000
expandNumber(numberString, wordRelation, tokenItem);
} else {
expandNumber(numberString.substring(0, 2), wordRelation, tokenItem);
wordRelation.addWord(tokenItem, "hundred");
}
} else if ((numberDigits == 2) && (numberString.charAt(0) == '0')) {
wordRelation.addWord(tokenItem, "oh");
expandDigits(numberString.substring(1, 2), wordRelation, tokenItem);
} else if ((numberDigits == 4 && numberString.charAt(1) == '0') || numberDigits < 3) {
expandNumber(numberString, wordRelation, tokenItem);
} else if (numberDigits % 2 == 1) {
String firstDigit = digit2num[numberString.charAt(0) - '0'];
wordRelation.addWord(tokenItem, firstDigit);
expandID(numberString.substring(1, numberDigits), wordRelation, tokenItem);
} else {
expandNumber(numberString.substring(0, 2), wordRelation, tokenItem);
expandID(numberString.substring(2, numberDigits), wordRelation, tokenItem);
}
}
/**
* Expands the given number string as a real number.
*
* @param numberString
* the string which is the real number to expand
* @param wordRelation
* words are added to this Relation
*/
public static void expandReal(String numberString, WordRelation wordRelation, Item tokenItem) {
int stringLength = numberString.length();
int position;
if (numberString.charAt(0) == '-') {
// negative real numbers
wordRelation.addWord(tokenItem, "minus");
expandReal(numberString.substring(1, stringLength), wordRelation, tokenItem);
} else if (numberString.charAt(0) == '+') {
// prefixed with a '+'
wordRelation.addWord(tokenItem, "plus");
expandReal(numberString.substring(1, stringLength), wordRelation, tokenItem);
} else if ((position = numberString.indexOf('e')) != -1 || (position = numberString.indexOf('E')) != -1) {
// numbers with 'E' or 'e'
expandReal(numberString.substring(0, position), wordRelation, tokenItem);
wordRelation.addWord(tokenItem, "e");
expandReal(numberString.substring(position + 1), wordRelation, tokenItem);
} else if ((position = numberString.indexOf('.')) != -1) {
// numbers with '.'
String beforeDot = numberString.substring(0, position);
if (beforeDot.length() > 0) {
expandReal(beforeDot, wordRelation, tokenItem);
}
wordRelation.addWord(tokenItem, "point");
String afterDot = numberString.substring(position + 1);
if (afterDot.length() > 0) {
expandDigits(afterDot, wordRelation, tokenItem);
}
} else {
// everything else
expandNumber(numberString, wordRelation, tokenItem);
}
}
/**
* Expands the given string of letters as a list of single char symbols.
*
* @param letters
* the string of letters to expand
* @param wordRelation
* words are added to this Relation
*/
public static void expandLetters(String letters, WordRelation wordRelation, Item tokenItem) {
letters = letters.toLowerCase();
char c;
for (int i = 0; i < letters.length(); i++) {
// if this is a number
c = letters.charAt(i);
if (isDigit(c)) {
wordRelation.addWord(tokenItem, digit2num[c - '0']);
} else if (letters.equals("a")) {
wordRelation.addWord(tokenItem, "_a");
} else {
wordRelation.addWord(tokenItem, String.valueOf(c));
}
}
}
/**
* Returns the integer value of the given string of Roman numerals.
*
* @param roman
* the string of Roman numbers
*
* @return the integer value
*/
public static int expandRoman(String roman) {
int value = 0;
for (int p = 0; p < roman.length(); p++) {
char c = roman.charAt(p);
if (c == 'X') {
value += 10;
} else if (c == 'V') {
value += 5;
} else if (c == 'I') {
if (p + 1 < roman.length()) {
char p1 = roman.charAt(p + 1);
if (p1 == 'V') {
value += 4;
p++;
} else if (p1 == 'X') {
value += 9;
p++;
} else {
value += 1;
}
} else {
value += 1;
}
}
}
return value;
}
/**
* Returns true if the given character is a digit (0-9 only).
*
* @param ch
* the character to test
*
* @return true or false
*/
public static boolean isDigit(char ch) {
return ('0' <= ch && ch <= '9');
}
}

19
marytts-languages/marytts-lang-en/src/main/java/marytts/language/en/Prosody.java
Просмотреть файл

@ -25,16 +25,11 @@ import marytts.datatypes.MaryDataType;
import marytts.modules.ProsodyGeneric;
import marytts.server.MaryProperties;
public class Prosody extends ProsodyGeneric
{
public Prosody()
{
super(MaryDataType.PHONEMES,
MaryDataType.INTONATION,
Locale.ENGLISH,
MaryProperties.localePrefix(Locale.ENGLISH)+".prosody.tobipredparams",
MaryProperties.localePrefix(Locale.ENGLISH)+".prosody.accentPriorities",
MaryProperties.localePrefix(Locale.ENGLISH)+".prosody.syllableaccents",
MaryProperties.localePrefix(Locale.ENGLISH)+".prosody.paragraphdeclination");
}
public class Prosody extends ProsodyGeneric {
public Prosody() {
super(MaryDataType.PHONEMES, MaryDataType.INTONATION, Locale.ENGLISH, MaryProperties.localePrefix(Locale.ENGLISH)
+ ".prosody.tobipredparams", MaryProperties.localePrefix(Locale.ENGLISH) + ".prosody.accentPriorities",
MaryProperties.localePrefix(Locale.ENGLISH) + ".prosody.syllableaccents", MaryProperties
.localePrefix(Locale.ENGLISH) + ".prosody.paragraphdeclination");
}
}

187
marytts-languages/marytts-lang-en/src/main/java/marytts/language/en/Segmenter.java
Просмотреть файл

@ -31,40 +31,35 @@ import com.sun.speech.freetts.Utterance;
import com.sun.speech.freetts.UtteranceProcessor;
import com.sun.speech.freetts.lexicon.Lexicon;
/**
* Annotates an utterance with <code>Relation.SYLLABLE</code>,
* <code>Relation.SYLLABLE_STRUCTURE</code>, and
* <code>Relation.SEGMENT</code>.
* To determine stress, the <code>isStressed</code> method relies upon
* a phone ending in the number "1". Subclasses should override
* <code>isStressed</code> and <code>deStress</code> if stresses are
* determined in other ways.
*
* Annotates an utterance with <code>Relation.SYLLABLE</code>, <code>Relation.SYLLABLE_STRUCTURE</code>, and
* <code>Relation.SEGMENT</code>. To determine stress, the <code>isStressed</code> method relies upon a phone ending in the number
* "1". Subclasses should override <code>isStressed</code> and <code>deStress</code> if stresses are determined in other ways.
*
* @see Relation#SEGMENT
* @see Relation#SYLLABLE
* @see Relation#SYLLABLE_STRUCTURE
*/
public class Segmenter implements UtteranceProcessor {
private final static String STRESS = "1";
private final static String NO_STRESS = "0";
private Map addenda;
private final static String STRESS = "1";
private final static String NO_STRESS = "0";
private Map addenda;
/**
* Annotates an utterance with <code>Relation.SYLLABLE</code>,
* <code>Relation.SYLLABLE_STRUCTURE</code>, and
* <code>Relation.SEGMENT</code>.
*
* @param utterance the utterance to process/tokenize
*
* @see Relation#SEGMENT
* @see Relation#SYLLABLE
* @see Relation#SYLLABLE_STRUCTURE
*
* @throws ProcessException if an IOException is thrown during the
* processing of the utterance
*/
public void processUtterance(Utterance utterance) throws ProcessException {
/**
* Annotates an utterance with <code>Relation.SYLLABLE</code>, <code>Relation.SYLLABLE_STRUCTURE</code>, and
* <code>Relation.SEGMENT</code>.
*
* @param utterance
* the utterance to process/tokenize
*
* @see Relation#SEGMENT
* @see Relation#SYLLABLE
* @see Relation#SYLLABLE_STRUCTURE
*
* @throws ProcessException
* if an IOException is thrown during the processing of the utterance
*/
public void processUtterance(Utterance utterance) throws ProcessException {
// preconditions
if (utterance.getRelation(Relation.WORD) == null) {
@ -183,80 +178,70 @@ public class Segmenter implements UtteranceProcessor {
assert utterance.getRelation(Relation.SEGMENT) != null;
}
/**
* Determines if the given phonene is stressed.
* To determine stress, this method relies upon
* a phone ending in the number "1". Subclasses should override this
* method if stresses are determined in other ways.
*
* @param phone the phone to check
*
* @return true if the phone is stressed, otherwise false
*/
protected boolean isStressed(String phone) {
return phone.endsWith("1");
}
/**
* Converts stressed phone to regular phone. This method
* merely removes the last character of the phone. Subclasses
* should override this if another method is to be used.
*
* @param phone the phone to convert
*
* @return de-stressed phone
*/
protected String deStress(String phone) {
String retPhone = phone;
if (isStressed(phone)) {
retPhone = phone.substring(0, phone.length() - 1);
/**
* Determines if the given phonene is stressed. To determine stress, this method relies upon a phone ending in the number "1".
* Subclasses should override this method if stresses are determined in other ways.
*
* @param phone
* the phone to check
*
* @return true if the phone is stressed, otherwise false
*/
protected boolean isStressed(String phone) {
return phone.endsWith("1");
}
return retPhone;
}
public void saveAddenda() throws IOException{
String path = MaryProperties.maryBase()+"/log/addenda.log";
if (addenda != null && path != null){
try{
String line;
StringBuilder addendaBuf = new StringBuilder();
addendaBuf.append("# Logfile contains words that are not in the lexicon"
+"\n# and their transcriptions predicted by the LTS-rules"
+"\n# (for English)\n");
//add new entries
Set words = addenda.keySet();
for (Iterator it = words.iterator();it.hasNext();){
String nextWord = (String) it.next();
addendaBuf.append(nextWord+" "
+addenda.get(nextWord)+"\n");
}
//open addenda file
PrintWriter addendaOut = new PrintWriter(
new FileOutputStream(new File(path)));
//print addendaBuf and close
addendaOut.print(addendaBuf.toString());
addendaOut.flush();
addendaOut.close();
} catch (Exception e){
throw new IOException("Could not save addenda because : "
+e.getMessage());
}
}
}
/**
* Returns the simple name of this class.
*
* @return the simple name of this class
*/
public String toString() {
return "Segmenter";
}
/**
* Converts stressed phone to regular phone. This method merely removes the last character of the phone. Subclasses should
* override this if another method is to be used.
*
* @param phone
* the phone to convert
*
* @return de-stressed phone
*/
protected String deStress(String phone) {
String retPhone = phone;
if (isStressed(phone)) {
retPhone = phone.substring(0, phone.length() - 1);
}
return retPhone;
}
public void saveAddenda() throws IOException {
String path = MaryProperties.maryBase() + "/log/addenda.log";
if (addenda != null && path != null) {
try {
String line;
StringBuilder addendaBuf = new StringBuilder();
addendaBuf.append("# Logfile contains words that are not in the lexicon"
+ "\n# and their transcriptions predicted by the LTS-rules" + "\n# (for English)\n");
// add new entries
Set words = addenda.keySet();
for (Iterator it = words.iterator(); it.hasNext();) {
String nextWord = (String) it.next();
addendaBuf.append(nextWord + " " + addenda.get(nextWord) + "\n");
}
// open addenda file
PrintWriter addendaOut = new PrintWriter(new FileOutputStream(new File(path)));
// print addendaBuf and close
addendaOut.print(addendaBuf.toString());
addendaOut.flush();
addendaOut.close();
} catch (Exception e) {
throw new IOException("Could not save addenda because : " + e.getMessage());
}
}
}
/**
* Returns the simple name of this class.
*
* @return the simple name of this class
*/
public String toString() {
return "Segmenter";
}
}

22
marytts-languages/marytts-lang-en/src/main/java/marytts/language/en/Utt2XMLDurationsEn.java
Просмотреть файл

@ -25,25 +25,15 @@ import marytts.datatypes.MaryDataType;
import marytts.language.en_US.datatypes.USEnglishDataTypes;
import marytts.modules.Utt2XMLBase;
/**
* Convert FreeTTS utterances into MaryXML format
* (Durations, English).
*
* Convert FreeTTS utterances into MaryXML format (Durations, English).
*
* @author Marc Schr&ouml;der
*/
public class Utt2XMLDurationsEn extends Utt2XMLBase
{
public Utt2XMLDurationsEn()
{
super("Utt2XML DurationsEn",
USEnglishDataTypes.FREETTS_MBROLISED_DURATIONS,
MaryDataType.DURATIONS,
Locale.ENGLISH);
}
public class Utt2XMLDurationsEn extends Utt2XMLBase {
public Utt2XMLDurationsEn() {
super("Utt2XML DurationsEn", USEnglishDataTypes.FREETTS_MBROLISED_DURATIONS, MaryDataType.DURATIONS, Locale.ENGLISH);
}
}

22
marytts-languages/marytts-lang-en/src/main/java/marytts/language/en/Utt2XMLIntonationEn.java
Просмотреть файл

@ -25,25 +25,15 @@ import marytts.datatypes.MaryDataType;
import marytts.language.en_US.datatypes.USEnglishDataTypes;
import marytts.modules.Utt2XMLBase;
/**
* Convert FreeTTS utterances into MaryXML format
* (Intonation, English).
*
* Convert FreeTTS utterances into MaryXML format (Intonation, English).
*
* @author Marc Schr&ouml;der
*/
public class Utt2XMLIntonationEn extends Utt2XMLBase
{
public Utt2XMLIntonationEn()
{
super("Utt2XML IntonationEn",
USEnglishDataTypes.FREETTS_INTONATION,
MaryDataType.INTONATION,
Locale.ENGLISH);
}
public class Utt2XMLIntonationEn extends Utt2XMLBase {
public Utt2XMLIntonationEn() {
super("Utt2XML IntonationEn", USEnglishDataTypes.FREETTS_INTONATION, MaryDataType.INTONATION, Locale.ENGLISH);
}
}

22
marytts-languages/marytts-lang-en/src/main/java/marytts/language/en/Utt2XMLPausesEn.java
Просмотреть файл

@ -24,25 +24,15 @@ import java.util.Locale;
import marytts.language.en_US.datatypes.USEnglishDataTypes;
import marytts.modules.Utt2XMLBase;
/**
* Convert FreeTTS utterances into MaryXML format
* (Pauses, English).
*
* Convert FreeTTS utterances into MaryXML format (Pauses, English).
*
* @author Marc Schr&ouml;der
*/
public class Utt2XMLPausesEn extends Utt2XMLBase
{
public Utt2XMLPausesEn()
{
super("Utt2XML PausesEn",
USEnglishDataTypes.FREETTS_PAUSES,
USEnglishDataTypes.PAUSES_US,
Locale.ENGLISH);
}
public class Utt2XMLPausesEn extends Utt2XMLBase {
public Utt2XMLPausesEn() {
super("Utt2XML PausesEn", USEnglishDataTypes.FREETTS_PAUSES, USEnglishDataTypes.PAUSES_US, Locale.ENGLISH);
}
}

22
marytts-languages/marytts-lang-en/src/main/java/marytts/language/en/Utt2XMLPhrasesEn.java
Просмотреть файл

@ -24,25 +24,15 @@ import java.util.Locale;
import marytts.language.en_US.datatypes.USEnglishDataTypes;
import marytts.modules.Utt2XMLBase;
/**
* Convert FreeTTS utterances into MaryXML format
* (Phrases, English).
*
* Convert FreeTTS utterances into MaryXML format (Phrases, English).
*
* @author Marc Schr&ouml;der
*/
public class Utt2XMLPhrasesEn extends Utt2XMLBase
{
public Utt2XMLPhrasesEn()
{
super("Utt2XML PhrasesEn",
USEnglishDataTypes.FREETTS_PHRASES,
USEnglishDataTypes.PHRASES_US,
Locale.ENGLISH);
}
public class Utt2XMLPhrasesEn extends Utt2XMLBase {
public Utt2XMLPhrasesEn() {
super("Utt2XML PhrasesEn", USEnglishDataTypes.FREETTS_PHRASES, USEnglishDataTypes.PHRASES_US, Locale.ENGLISH);
}
}

21
marytts-languages/marytts-lang-en/src/main/java/marytts/language/en/Utt2XMLPosEn.java
Просмотреть файл

@ -25,24 +25,15 @@ import marytts.datatypes.MaryDataType;
import marytts.language.en_US.datatypes.USEnglishDataTypes;
import marytts.modules.Utt2XMLBase;
/**
* Convert FreeTTS utterances into MaryXML format
* (parts of speech, English).
*
* Convert FreeTTS utterances into MaryXML format (parts of speech, English).
*
* @author Marc Schr&ouml;der
*/
public class Utt2XMLPosEn extends Utt2XMLBase
{
public Utt2XMLPosEn()
{
super("Utt2XML PosEn",
USEnglishDataTypes.FREETTS_POS,
MaryDataType.PARTSOFSPEECH,
Locale.ENGLISH);
}
public class Utt2XMLPosEn extends Utt2XMLBase {
public Utt2XMLPosEn() {
super("Utt2XML PosEn", USEnglishDataTypes.FREETTS_POS, MaryDataType.PARTSOFSPEECH, Locale.ENGLISH);
}
}

22
marytts-languages/marytts-lang-en/src/main/java/marytts/language/en/Utt2XMLPostprocessedEn.java
Просмотреть файл

@ -25,25 +25,15 @@ import marytts.datatypes.MaryDataType;
import marytts.language.en_US.datatypes.USEnglishDataTypes;
import marytts.modules.Utt2XMLBase;
/**
* Convert FreeTTS utterances into MaryXML format
* (Postprocessed, English).
*
* Convert FreeTTS utterances into MaryXML format (Postprocessed, English).
*
* @author Marc Schr&ouml;der
*/
public class Utt2XMLPostprocessedEn extends Utt2XMLBase
{
public Utt2XMLPostprocessedEn()
{
super("Utt2XML PostprocessedEn",
USEnglishDataTypes.FREETTS_POSTPROCESSED,
MaryDataType.ALLOPHONES,
Locale.ENGLISH);
}
public class Utt2XMLPostprocessedEn extends Utt2XMLBase {
public Utt2XMLPostprocessedEn() {
super("Utt2XML PostprocessedEn", USEnglishDataTypes.FREETTS_POSTPROCESSED, MaryDataType.ALLOPHONES, Locale.ENGLISH);
}
}

21
marytts-languages/marytts-lang-en/src/main/java/marytts/language/en/Utt2XMLSegmentsEn.java
Просмотреть файл

@ -25,24 +25,15 @@ import marytts.datatypes.MaryDataType;
import marytts.language.en_US.datatypes.USEnglishDataTypes;
import marytts.modules.Utt2XMLBase;
/**
* Convert FreeTTS utterances into MaryXML format
* (Segments, English).
*
* Convert FreeTTS utterances into MaryXML format (Segments, English).
*
* @author Marc Schr&ouml;der
*/
public class Utt2XMLSegmentsEn extends Utt2XMLBase
{
public Utt2XMLSegmentsEn()
{
super("Utt2XML SegmentsEn",
USEnglishDataTypes.FREETTS_SEGMENTS,
MaryDataType.PHONEMES,
Locale.ENGLISH);
}
public class Utt2XMLSegmentsEn extends Utt2XMLBase {
public Utt2XMLSegmentsEn() {
super("Utt2XML SegmentsEn", USEnglishDataTypes.FREETTS_SEGMENTS, MaryDataType.PHONEMES, Locale.ENGLISH);
}
}

22
marytts-languages/marytts-lang-en/src/main/java/marytts/language/en/Utt2XMLTokensEn.java
Просмотреть файл

@ -25,25 +25,15 @@ import marytts.datatypes.MaryDataType;
import marytts.language.en_US.datatypes.USEnglishDataTypes;
import marytts.modules.Utt2XMLBase;
/**
* Convert FreeTTS utterances into MaryXML format
* (Tokens, English).
*
* Convert FreeTTS utterances into MaryXML format (Tokens, English).
*
* @author Marc Schr&ouml;der
*/
public class Utt2XMLTokensEn extends Utt2XMLBase
{
public Utt2XMLTokensEn()
{
super("Utt2XML TokensEn",
USEnglishDataTypes.FREETTS_TOKENS,
MaryDataType.TOKENS,
Locale.ENGLISH);
}
public class Utt2XMLTokensEn extends Utt2XMLBase {
public Utt2XMLTokensEn() {
super("Utt2XML TokensEn", USEnglishDataTypes.FREETTS_TOKENS, MaryDataType.TOKENS, Locale.ENGLISH);
}
}

21
marytts-languages/marytts-lang-en/src/main/java/marytts/language/en/Utt2XMLWordsEn.java
Просмотреть файл

@ -25,24 +25,15 @@ import marytts.datatypes.MaryDataType;
import marytts.language.en_US.datatypes.USEnglishDataTypes;
import marytts.modules.Utt2XMLBase;
/**
* Convert FreeTTS utterances into MaryXML format
* (Words, English).
*
* Convert FreeTTS utterances into MaryXML format (Words, English).
*
* @author Marc Schr&ouml;der
*/
public class Utt2XMLWordsEn extends Utt2XMLBase
{
public Utt2XMLWordsEn()
{
super("Utt2XML WordsEn",
USEnglishDataTypes.FREETTS_WORDS,
MaryDataType.WORDS,
Locale.ENGLISH);
}
public class Utt2XMLWordsEn extends Utt2XMLBase {
public Utt2XMLWordsEn() {
super("Utt2XML WordsEn", USEnglishDataTypes.FREETTS_WORDS, MaryDataType.WORDS, Locale.ENGLISH);
}
}

126
marytts-languages/marytts-lang-en/src/main/java/marytts/language/en/WordRelation.java
Просмотреть файл

@ -20,76 +20,72 @@ import com.sun.speech.freetts.Utterance;
*/
public class WordRelation {
private Relation relation;
private TokenToWords tokenToWords;
private Relation relation;
private TokenToWords tokenToWords;
private WordRelation(Relation parentRelation, TokenToWords tokenToWords) {
this.relation = parentRelation;
this.tokenToWords = tokenToWords;
}
/**
* Creates a WordRelation object with the given utterance and
* TokenToWords.
*
* @param utterance the Utterance from which to create a Relation
* @param tokenToWords the TokenToWords object to use
*
* @return a WordRelation object
*/
public static WordRelation createWordRelation(Utterance utterance,
TokenToWords tokenToWords) {
Relation relation = utterance.createRelation(Relation.WORD);
return new WordRelation(relation, tokenToWords);
}
/**
* Adds a break as a feature to the last item in the list.
*/
public void addBreak() {
Item wordItem = relation.getTail();
if (wordItem != null) {
FeatureSet featureSet = wordItem.getFeatures();
featureSet.setString("break", "1");
private WordRelation(Relation parentRelation, TokenToWords tokenToWords) {
this.relation = parentRelation;
this.tokenToWords = tokenToWords;
}
}
/**
* Creates a WordRelation object with the given utterance and TokenToWords.
*
* @param utterance
* the Utterance from which to create a Relation
* @param tokenToWords
* the TokenToWords object to use
*
* @return a WordRelation object
*/
public static WordRelation createWordRelation(Utterance utterance, TokenToWords tokenToWords) {
Relation relation = utterance.createRelation(Relation.WORD);
return new WordRelation(relation, tokenToWords);
}
/**
* Adds a word as an Item to this WordRelation object.
*
* @param word the word to add
*/
public void addWord(Item tokenItem, String word) {
assert (tokenItem != null);
Item wordItem = tokenItem.createDaughter();
FeatureSet featureSet = wordItem.getFeatures();
featureSet.setString("name", word);
relation.appendItem(wordItem);
}
/**
* Adds a break as a feature to the last item in the list.
*/
public void addBreak() {
Item wordItem = relation.getTail();
if (wordItem != null) {
FeatureSet featureSet = wordItem.getFeatures();
featureSet.setString("break", "1");
}
}
/**
* Adds a word as an Item to this WordRelation object.
*
* @param word
* the word to add
*/
public void addWord(Item tokenItem, String word) {
assert(tokenItem != null);
Item wordItem = tokenItem.createDaughter();
FeatureSet featureSet = wordItem.getFeatures();
featureSet.setString("name", word);
relation.appendItem(wordItem);
}
/**
* Sets the last Item in this WordRelation to the given word.
*
* @param word the word to set
*/
public void setLastWord(String word) {
Item lastItem = relation.getTail();
FeatureSet featureSet = lastItem.getFeatures();
featureSet.setString("name", word);
}
/**
* Sets the last Item in this WordRelation to the given word.
*
* @param word
* the word to set
*/
public void setLastWord(String word) {
Item lastItem = relation.getTail();
FeatureSet featureSet = lastItem.getFeatures();
featureSet.setString("name", word);
}
/**
* Returns the last item in this WordRelation.
*
* @return the last item
*/
public Item getTail() {
return relation.getTail();
}
/**
* Returns the last item in this WordRelation.
*
* @return the last item
*/
public Item getTail() {
return relation.getTail();
}
}

39
marytts-languages/marytts-lang-en/src/main/java/marytts/language/en/XML2UttDurationsEn.java
Просмотреть файл

@ -29,35 +29,24 @@ import org.w3c.dom.Element;
import com.sun.speech.freetts.Utterance;
/**
* Convert a MaryXML DOM tree into FreeTTS utterances
* (durations, English).
*
* Convert a MaryXML DOM tree into FreeTTS utterances (durations, English).
*
* @author Marc Schr&ouml;der
*/
public class XML2UttDurationsEn extends XML2UttBase
{
public XML2UttDurationsEn() {
super("XML2Utt DurationsEn",
MaryDataType.DURATIONS,
USEnglishDataTypes.FREETTS_DURATIONS,
Locale.ENGLISH);
}
public class XML2UttDurationsEn extends XML2UttBase {
public XML2UttDurationsEn() {
super("XML2Utt DurationsEn", MaryDataType.DURATIONS, USEnglishDataTypes.FREETTS_DURATIONS, Locale.ENGLISH);
}
/**
* Depending on the data type, find the right information in the sentence
* and insert it into the utterance.
*/
protected void fillUtterance(Utterance utterance, Element sentence)
{
fillUtterance(utterance, sentence,
true, // create word relation
true, // create sylstruct relation
false); // don't create target relation
}
/**
* Depending on the data type, find the right information in the sentence and insert it into the utterance.
*/
protected void fillUtterance(Utterance utterance, Element sentence) {
fillUtterance(utterance, sentence, true, // create word relation
true, // create sylstruct relation
false); // don't create target relation
}
}

39
marytts-languages/marytts-lang-en/src/main/java/marytts/language/en/XML2UttIntonationEn.java
Просмотреть файл

@ -29,35 +29,24 @@ import org.w3c.dom.Element;
import com.sun.speech.freetts.Utterance;
/**
* Convert a MaryXML DOM tree into FreeTTS utterances
* (intonation, English).
*
* Convert a MaryXML DOM tree into FreeTTS utterances (intonation, English).
*
* @author Marc Schr&ouml;der
*/
public class XML2UttIntonationEn extends XML2UttBase
{
public XML2UttIntonationEn() {
super("XML2Utt IntonationEn",
MaryDataType.INTONATION,
USEnglishDataTypes.FREETTS_INTONATION,
Locale.ENGLISH);
}
public class XML2UttIntonationEn extends XML2UttBase {
public XML2UttIntonationEn() {
super("XML2Utt IntonationEn", MaryDataType.INTONATION, USEnglishDataTypes.FREETTS_INTONATION, Locale.ENGLISH);
}
/**
* Depending on the data type, find the right information in the sentence
* and insert it into the utterance.
*/
protected void fillUtterance(Utterance utterance, Element sentence)
{
fillUtterance(utterance, sentence,
true, // create word relation
true, // create sylstruct relation
false); // don't create target relation
}
/**
* Depending on the data type, find the right information in the sentence and insert it into the utterance.
*/
protected void fillUtterance(Utterance utterance, Element sentence) {
fillUtterance(utterance, sentence, true, // create word relation
true, // create sylstruct relation
false); // don't create target relation
}
}

39
marytts-languages/marytts-lang-en/src/main/java/marytts/language/en/XML2UttPausesEn.java
Просмотреть файл

@ -28,35 +28,24 @@ import org.w3c.dom.Element;
import com.sun.speech.freetts.Utterance;
/**
* Convert a MaryXML DOM tree into FreeTTS utterances
* (pauses, English).
*
* Convert a MaryXML DOM tree into FreeTTS utterances (pauses, English).
*
* @author Marc Schr&ouml;der
*/
public class XML2UttPausesEn extends XML2UttBase
{
public XML2UttPausesEn() {
super("XML2Utt PausesEn",
USEnglishDataTypes.PAUSES_US,
USEnglishDataTypes.FREETTS_PAUSES,
Locale.ENGLISH);
}
public class XML2UttPausesEn extends XML2UttBase {
public XML2UttPausesEn() {
super("XML2Utt PausesEn", USEnglishDataTypes.PAUSES_US, USEnglishDataTypes.FREETTS_PAUSES, Locale.ENGLISH);
}
/**
* Depending on the data type, find the right information in the sentence
* and insert it into the utterance.
*/
protected void fillUtterance(Utterance utterance, Element sentence)
{
fillUtterance(utterance, sentence,
true, // create word relation
true, // create sylstruct relation
false); // don't create target relation
}
/**
* Depending on the data type, find the right information in the sentence and insert it into the utterance.
*/
protected void fillUtterance(Utterance utterance, Element sentence) {
fillUtterance(utterance, sentence, true, // create word relation
true, // create sylstruct relation
false); // don't create target relation
}
}

39
marytts-languages/marytts-lang-en/src/main/java/marytts/language/en/XML2UttPhrasesEn.java
Просмотреть файл

@ -28,36 +28,25 @@ import org.w3c.dom.Element;
import com.sun.speech.freetts.Utterance;
/**
* Convert a MaryXML DOM tree into FreeTTS utterances
* (phrases, English).
*
* Convert a MaryXML DOM tree into FreeTTS utterances (phrases, English).
*
* @author Marc Schr&ouml;der
*/
public class XML2UttPhrasesEn extends XML2UttBase
{
public XML2UttPhrasesEn() {
super("XML2Utt PhrasesEn",
USEnglishDataTypes.PHRASES_US,
USEnglishDataTypes.FREETTS_PHRASES,
Locale.ENGLISH);
}
public class XML2UttPhrasesEn extends XML2UttBase {
public XML2UttPhrasesEn() {
super("XML2Utt PhrasesEn", USEnglishDataTypes.PHRASES_US, USEnglishDataTypes.FREETTS_PHRASES, Locale.ENGLISH);
}
/**
* Depending on the data type, find the right information in the sentence
* and insert it into the utterance.
*/
protected void fillUtterance(Utterance utterance, Element sentence)
{
fillUtterance(utterance, sentence,
true, // create word relation
false, // do not create sylstruct relation
false); // don't create target relation
/**
* Depending on the data type, find the right information in the sentence and insert it into the utterance.
*/
protected void fillUtterance(Utterance utterance, Element sentence) {
fillUtterance(utterance, sentence, true, // create word relation
false, // do not create sylstruct relation
false); // don't create target relation
}
}
}

39
marytts-languages/marytts-lang-en/src/main/java/marytts/language/en/XML2UttPosEn.java
Просмотреть файл

@ -29,36 +29,25 @@ import org.w3c.dom.Element;
import com.sun.speech.freetts.Utterance;
/**
* Convert a MaryXML DOM tree into FreeTTS utterances
* (parts of speech, English).
*
* Convert a MaryXML DOM tree into FreeTTS utterances (parts of speech, English).
*
* @author Marc Schr&ouml;der
*/
public class XML2UttPosEn extends XML2UttBase
{
public XML2UttPosEn() {
super("XML2Utt PosEn",
MaryDataType.PARTSOFSPEECH,
USEnglishDataTypes.FREETTS_POS,
Locale.ENGLISH);
}
public class XML2UttPosEn extends XML2UttBase {
public XML2UttPosEn() {
super("XML2Utt PosEn", MaryDataType.PARTSOFSPEECH, USEnglishDataTypes.FREETTS_POS, Locale.ENGLISH);
}
/**
* Depending on the data type, find the right information in the sentence
* and insert it into the utterance.
*/
protected void fillUtterance(Utterance utterance, Element sentence)
{
fillUtterance(utterance, sentence,
true, // create word relation
false, // do not create sylstruct relation
false); // don't create target relation
/**
* Depending on the data type, find the right information in the sentence and insert it into the utterance.
*/
protected void fillUtterance(Utterance utterance, Element sentence) {
fillUtterance(utterance, sentence, true, // create word relation
false, // do not create sylstruct relation
false); // don't create target relation
}
}
}

39
marytts-languages/marytts-lang-en/src/main/java/marytts/language/en/XML2UttPostprocessedEn.java
Просмотреть файл

@ -29,35 +29,24 @@ import org.w3c.dom.Element;
import com.sun.speech.freetts.Utterance;
/**
* Convert a MaryXML DOM tree into FreeTTS utterances
* (postprocessed, English).
*
* Convert a MaryXML DOM tree into FreeTTS utterances (postprocessed, English).
*
* @author Marc Schr&ouml;der
*/
public class XML2UttPostprocessedEn extends XML2UttBase
{
public XML2UttPostprocessedEn() {
super("XML2Utt PostprocessedEn",
MaryDataType.ALLOPHONES,
USEnglishDataTypes.FREETTS_POSTPROCESSED,
Locale.ENGLISH);
}
public class XML2UttPostprocessedEn extends XML2UttBase {
public XML2UttPostprocessedEn() {
super("XML2Utt PostprocessedEn", MaryDataType.ALLOPHONES, USEnglishDataTypes.FREETTS_POSTPROCESSED, Locale.ENGLISH);
}
/**
* Depending on the data type, find the right information in the sentence
* and insert it into the utterance.
*/
protected void fillUtterance(Utterance utterance, Element sentence)
{
fillUtterance(utterance, sentence,
true, // create word relation
true, // create sylstruct relation
false); // don't create target relation
}
/**
* Depending on the data type, find the right information in the sentence and insert it into the utterance.
*/
protected void fillUtterance(Utterance utterance, Element sentence) {
fillUtterance(utterance, sentence, true, // create word relation
true, // create sylstruct relation
false); // don't create target relation
}
}

39
marytts-languages/marytts-lang-en/src/main/java/marytts/language/en/XML2UttSegmentsEn.java
Просмотреть файл

@ -29,35 +29,24 @@ import org.w3c.dom.Element;
import com.sun.speech.freetts.Utterance;
/**
* Convert a MaryXML DOM tree into FreeTTS utterances
* (segments, English).
*
* Convert a MaryXML DOM tree into FreeTTS utterances (segments, English).
*
* @author Marc Schr&ouml;der
*/
public class XML2UttSegmentsEn extends XML2UttBase
{
public XML2UttSegmentsEn() {
super("XML2Utt SegmentsEn",
MaryDataType.PHONEMES,
USEnglishDataTypes.FREETTS_SEGMENTS,
Locale.ENGLISH);
}
public class XML2UttSegmentsEn extends XML2UttBase {
public XML2UttSegmentsEn() {
super("XML2Utt SegmentsEn", MaryDataType.PHONEMES, USEnglishDataTypes.FREETTS_SEGMENTS, Locale.ENGLISH);
}
/**
* Depending on the data type, find the right information in the sentence
* and insert it into the utterance.
*/
protected void fillUtterance(Utterance utterance, Element sentence)
{
fillUtterance(utterance, sentence,
true, // create word relation
true, // create sylstruct relation
false); // don't create target relation
}
/**
* Depending on the data type, find the right information in the sentence and insert it into the utterance.
*/
protected void fillUtterance(Utterance utterance, Element sentence) {
fillUtterance(utterance, sentence, true, // create word relation
true, // create sylstruct relation
false); // don't create target relation
}
}

39
marytts-languages/marytts-lang-en/src/main/java/marytts/language/en/XML2UttTokensEn.java
Просмотреть файл

@ -29,36 +29,25 @@ import org.w3c.dom.Element;
import com.sun.speech.freetts.Utterance;
/**
* Convert a MaryXML DOM tree into FreeTTS utterances
* (tokens, English).
*
* Convert a MaryXML DOM tree into FreeTTS utterances (tokens, English).
*
* @author Marc Schr&ouml;der
*/
public class XML2UttTokensEn extends XML2UttBase
{
public XML2UttTokensEn() {
super("XML2Utt TokensEn",
MaryDataType.TOKENS,
USEnglishDataTypes.FREETTS_TOKENS,
Locale.ENGLISH);
}
public class XML2UttTokensEn extends XML2UttBase {
public XML2UttTokensEn() {
super("XML2Utt TokensEn", MaryDataType.TOKENS, USEnglishDataTypes.FREETTS_TOKENS, Locale.ENGLISH);
}
/**
* Depending on the data type, find the right information in the sentence
* and insert it into the utterance.
*/
protected void fillUtterance(Utterance utterance, Element sentence)
{
fillUtterance(utterance, sentence,
false, // do not create word relation
false, // do not create sylstruct relation
false); // don't create target relation
/**
* Depending on the data type, find the right information in the sentence and insert it into the utterance.
*/
protected void fillUtterance(Utterance utterance, Element sentence) {
fillUtterance(utterance, sentence, false, // do not create word relation
false, // do not create sylstruct relation
false); // don't create target relation
}
}
}

39
marytts-languages/marytts-lang-en/src/main/java/marytts/language/en/XML2UttWordsEn.java
Просмотреть файл

@ -29,36 +29,25 @@ import org.w3c.dom.Element;
import com.sun.speech.freetts.Utterance;
/**
* Convert a MaryXML DOM tree into FreeTTS utterances
* (words, English).
*
* Convert a MaryXML DOM tree into FreeTTS utterances (words, English).
*
* @author Marc Schr&ouml;der
*/
public class XML2UttWordsEn extends XML2UttBase
{
public XML2UttWordsEn() {
super("XML2Utt WordsEn",
MaryDataType.WORDS,
USEnglishDataTypes.FREETTS_WORDS,
Locale.ENGLISH);
}
public class XML2UttWordsEn extends XML2UttBase {
public XML2UttWordsEn() {
super("XML2Utt WordsEn", MaryDataType.WORDS, USEnglishDataTypes.FREETTS_WORDS, Locale.ENGLISH);
}
/**
* Depending on the data type, find the right information in the sentence
* and insert it into the utterance.
*/
protected void fillUtterance(Utterance utterance, Element sentence)
{
fillUtterance(utterance, sentence,
true, // create word relation
false, // do not create sylstruct relation
false); // don't create target relation
/**
* Depending on the data type, find the right information in the sentence and insert it into the utterance.
*/
protected void fillUtterance(Utterance utterance, Element sentence) {
fillUtterance(utterance, sentence, true, // create word relation
false, // do not create sylstruct relation
false); // don't create target relation
}
}
}

105
marytts-languages/marytts-lang-en/src/main/java/marytts/language/en/prosody/SentenceType.java
Просмотреть файл

@ -22,56 +22,67 @@ package marytts.language.en.prosody;
/**
* Information on how to recognise and how to realise different sentence types.
*/
public class SentenceType
{
public static final SentenceType declarative =
new SentenceType( "declarative", "L-L%", "H-L%", "H-", "!H*", "L+H*");
public static final SentenceType interrogative =
new SentenceType("interrogative", "H-H%", "H-L%", "H-", "L*", "H*");
public static final SentenceType exclamation =
new SentenceType("exclamation", "L-L%", "H-L%", "H-", "H*", "H*");
public static final SentenceType interrogYN =
new SentenceType("interrogYN", "H-H%", "H-L%", "H-", "L*", "H*");
public static final SentenceType interrogWH =
new SentenceType("interrogWH", "L-L%", "H-L%", "H-", "H*", "H*");
public class SentenceType {
public static final SentenceType declarative = new SentenceType("declarative", "L-L%", "H-L%", "H-", "!H*", "L+H*");
public static final SentenceType interrogative = new SentenceType("interrogative", "H-H%", "H-L%", "H-", "L*", "H*");
public static final SentenceType exclamation = new SentenceType("exclamation", "L-L%", "H-L%", "H-", "H*", "H*");
public static final SentenceType interrogYN = new SentenceType("interrogYN", "H-H%", "H-L%", "H-", "L*", "H*");
public static final SentenceType interrogWH = new SentenceType("interrogWH", "L-L%", "H-L%", "H-", "H*", "H*");
public static SentenceType punctuationType(String punct)
{
if (punct.equals(".")) return declarative;
else if (punct.equals("?")) return interrogative;
else if (punct.equals("!")) return exclamation;
else return null;
}
public static SentenceType punctuationType(String punct) {
if (punct.equals("."))
return declarative;
else if (punct.equals("?"))
return interrogative;
else if (punct.equals("!"))
return exclamation;
else
return null;
}
private String name;
private String sentenceFinalBoundary;
private String nonFinalMajorBoundary;
private String minorBoundary;
private String nuclearAccent;
private String nonNuclearAccent;
private String name;
private String sentenceFinalBoundary;
private String nonFinalMajorBoundary;
private String minorBoundary;
private String nuclearAccent;
private String nonNuclearAccent;
private SentenceType(String name,
String sentenceFinalBoundary,
String nonFinalMajorBoundary,
String minorBoundary,
String nuclearAccent,
String nonNuclearAccent)
{
this.name = name;
this.sentenceFinalBoundary = sentenceFinalBoundary;
this.nonFinalMajorBoundary = nonFinalMajorBoundary;
this.minorBoundary = minorBoundary;
this.nuclearAccent = nuclearAccent;
this.nonNuclearAccent = nonNuclearAccent;
}
private SentenceType(String name, String sentenceFinalBoundary, String nonFinalMajorBoundary, String minorBoundary,
String nuclearAccent, String nonNuclearAccent) {
this.name = name;
this.sentenceFinalBoundary = sentenceFinalBoundary;
this.nonFinalMajorBoundary = nonFinalMajorBoundary;
this.minorBoundary = minorBoundary;
this.nuclearAccent = nuclearAccent;
this.nonNuclearAccent = nonNuclearAccent;
}
public String name() { return name; }
public String toString() { return name(); }
public String sentenceFinalBoundary() { return sentenceFinalBoundary; }
public String nonFinalMajorBoundary() { return nonFinalMajorBoundary; }
public String minorBoundary() { return minorBoundary; }
public String nuclearAccent() { return nuclearAccent; }
public String nonNuclearAccent() { return nonNuclearAccent; }
public String name() {
return name;
}
public String toString() {
return name();
}
public String sentenceFinalBoundary() {
return sentenceFinalBoundary;
}
public String nonFinalMajorBoundary() {
return nonFinalMajorBoundary;
}
public String minorBoundary() {
return minorBoundary;
}
public String nuclearAccent() {
return nuclearAccent;
}
public String nonNuclearAccent() {
return nonNuclearAccent;
}
}

42
marytts-languages/marytts-lang-en/src/main/java/marytts/language/en_US/datatypes/USEnglishDataTypes.java
Просмотреть файл

@ -23,25 +23,29 @@ import marytts.datatypes.MaryDataType;
import marytts.datatypes.MaryXML;
/**
* This class will register the data types that are specific for the
* US English synthesis modules.
* This class will register the data types that are specific for the US English synthesis modules.
*
* @author marc
*
*
*/
public class USEnglishDataTypes
{
public static final MaryDataType FREETTS_CONTOUR = new MaryDataType("FREETTS_CONTOUR", false, false, MaryDataType.UTTERANCES);
public static final MaryDataType FREETTS_DURATIONS = new MaryDataType("FREETTS_DURATIONS", false, false, MaryDataType.UTTERANCES);
public static final MaryDataType FREETTS_INTONATION = new MaryDataType("FREETTS_INTONATION", false, false, MaryDataType.UTTERANCES);
public static final MaryDataType FREETTS_MBROLISED_DURATIONS = new MaryDataType("FREETTS_MBROLISED_DURATIONS", false, false, MaryDataType.UTTERANCES);
public static final MaryDataType FREETTS_PAUSES = new MaryDataType("FREETTS_PAUSES", false, false, MaryDataType.UTTERANCES);
public static final MaryDataType FREETTS_PHRASES = new MaryDataType("FREETTS_PHRASES", false, false, MaryDataType.UTTERANCES);
public static final MaryDataType FREETTS_POS = new MaryDataType("FREETTS_POS", false, false, MaryDataType.UTTERANCES);
public static final MaryDataType FREETTS_POSTPROCESSED = new MaryDataType("FREETTS_POSTPROCESSED", false, false, MaryDataType.UTTERANCES);
public static final MaryDataType FREETTS_SEGMENTS = new MaryDataType("FREETTS_SEGMENTS", false, false, MaryDataType.UTTERANCES);
public static final MaryDataType FREETTS_TOKENS = new MaryDataType("FREETTS_TOKENS", false, false, MaryDataType.UTTERANCES);
public static final MaryDataType FREETTS_WORDS = new MaryDataType("FREETTS_WORDS", false, false, MaryDataType.UTTERANCES);
public static final MaryDataType PAUSES_US = new MaryDataType("PAUSES_US", true, true, MaryDataType.MARYXML, MaryXML.MARYXML);
public static final MaryDataType PHRASES_US = new MaryDataType("PHRASES_US", true, true, MaryDataType.MARYXML, MaryXML.MARYXML);
public class USEnglishDataTypes {
public static final MaryDataType FREETTS_CONTOUR = new MaryDataType("FREETTS_CONTOUR", false, false, MaryDataType.UTTERANCES);
public static final MaryDataType FREETTS_DURATIONS = new MaryDataType("FREETTS_DURATIONS", false, false,
MaryDataType.UTTERANCES);
public static final MaryDataType FREETTS_INTONATION = new MaryDataType("FREETTS_INTONATION", false, false,
MaryDataType.UTTERANCES);
public static final MaryDataType FREETTS_MBROLISED_DURATIONS = new MaryDataType("FREETTS_MBROLISED_DURATIONS", false, false,
MaryDataType.UTTERANCES);
public static final MaryDataType FREETTS_PAUSES = new MaryDataType("FREETTS_PAUSES", false, false, MaryDataType.UTTERANCES);
public static final MaryDataType FREETTS_PHRASES = new MaryDataType("FREETTS_PHRASES", false, false, MaryDataType.UTTERANCES);
public static final MaryDataType FREETTS_POS = new MaryDataType("FREETTS_POS", false, false, MaryDataType.UTTERANCES);
public static final MaryDataType FREETTS_POSTPROCESSED = new MaryDataType("FREETTS_POSTPROCESSED", false, false,
MaryDataType.UTTERANCES);
public static final MaryDataType FREETTS_SEGMENTS = new MaryDataType("FREETTS_SEGMENTS", false, false,
MaryDataType.UTTERANCES);
public static final MaryDataType FREETTS_TOKENS = new MaryDataType("FREETTS_TOKENS", false, false, MaryDataType.UTTERANCES);
public static final MaryDataType FREETTS_WORDS = new MaryDataType("FREETTS_WORDS", false, false, MaryDataType.UTTERANCES);
public static final MaryDataType PAUSES_US = new MaryDataType("PAUSES_US", true, true, MaryDataType.MARYXML, MaryXML.MARYXML);
public static final MaryDataType PHRASES_US = new MaryDataType("PHRASES_US", true, true, MaryDataType.MARYXML,
MaryXML.MARYXML);
}