public static void main(String[] args) throws IOException, MaryConfigurationException {
if (args.length < 2) {
System.out.println("Usage:");
System.out.println(
"java marytts.modules.phonemiser.TrainedLTS allophones.xml lts-model.lts [removeTrailingOneFromPhones]");
System.exit(0);
}
String allophoneFile = args[0];
String ltsFile = args[1];
boolean myRemoveTrailingOneFromPhones = true;
if (args.length > 2) {
myRemoveTrailingOneFromPhones = Boolean.getBoolean(args[2]);
}
TrainedLTS lts =
new TrainedLTS(
AllophoneSet.getAllophoneSet(allophoneFile),
new FileInputStream(ltsFile),
myRemoveTrailingOneFromPhones,
new Syllabifier(
AllophoneSet.getAllophoneSet(allophoneFile), myRemoveTrailingOneFromPhones));
BufferedReader br = new BufferedReader(new InputStreamReader(System.in));
String line;
while ((line = br.readLine()) != null) {
line = line.trim();
String pron = lts.predictPronunciation(line);
String syl = lts.syllabify(pron);
String sylStripped = syl.replaceAll("[-' ]+", "");
System.out.println(sylStripped);
}
}
/**
* Phonemise the word text. This starts with a simple lexicon lookup, followed by some heuristics,
* and finally applies letter-to-sound rules if nothing else was successful.
*
* @param text the textual (graphemic) form of a word.
* @param pos the part-of-speech of the word
* @param g2pMethod This is an awkward way to return a second String parameter via a
* StringBuilder. If a phonemisation of the text is found, this parameter will be filled with
* the method of phonemisation ("lexicon", ... "rules").
* @return a phonemisation of the text if one can be generated, or null if no phonemisation method
* was successful.
*/
public String phonemise(String text, String pos, StringBuilder g2pMethod) {
// First, try a simple userdict and lexicon lookup:
String result = userdictLookup(text, pos);
if (result != null) {
g2pMethod.append("userdict");
return result;
}
result = lexiconLookup(text, pos);
if (result != null) {
g2pMethod.append("lexicon");
return result;
}
// HB 150915 adding secondary lexicon
result = secondary_lexiconLookup(text, pos);
if (result != null) {
g2pMethod.append("lexicon");
return result;
}
// Lookup attempts failed. Try normalising exotic letters
// (diacritics on vowels, etc.), look up again:
String normalised = MaryUtils.normaliseUnicodeLetters(text, getLocale());
if (!normalised.equals(text)) {
result = userdictLookup(normalised, pos);
if (result != null) {
g2pMethod.append("userdict");
return result;
}
result = lexiconLookup(normalised, pos);
if (result != null) {
g2pMethod.append("lexicon");
return result;
}
// HB 150915 adding secondary lexicon
result = secondary_lexiconLookup(text, pos);
if (result != null) {
g2pMethod.append("lexicon");
return result;
}
}
// Cannot find it in the lexicon -- apply letter-to-sound rules
// to the normalised form
String phones = lts.predictPronunciation(text);
try {
result = lts.syllabify(phones);
} catch (IllegalArgumentException e) {
logger.error(String.format("Problem with token <%s> [%s]: %s", text, phones, e.getMessage()));
}
if (result != null) {
g2pMethod.append("rules");
return result;
}
return null;
}