marytts.tools.newlanguage.en_US
Class CMUDict2MaryFST
java.lang.Object
marytts.tools.newlanguage.LexiconCreator
marytts.tools.newlanguage.en_US.CMUDict2MaryFST
public class CMUDict2MaryFST
- extends LexiconCreator
This class does a one-time, offline conversion from the CMUDict in Festival format
(cmudict-0.4.scm and cmudict_extensions.scm) into MARY format.
Specifically, the following steps are performed:
- conversion to a text format without brackets, using '|' as the delimiter between three fields:
graphemes | allophones | part-of-speech(optional)
- conversion of the phonetic alphabet used from MRPA to SAMPA
- creation of a compact FST representing the lexicon
- training of Letter-to-sound rules from the data
- Author:
- marc
|
Method Summary |
static void |
main(java.lang.String[] args)
|
protected void |
prepareLexicon()
This base implementation does nothing. |
| Methods inherited from class java.lang.Object |
clone, equals, finalize, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait |
CMUDict2MaryFST
public CMUDict2MaryFST()
throws java.lang.Exception
- Throws:
java.lang.Exception
prepareLexicon
protected void prepareLexicon()
throws java.io.IOException
- Description copied from class:
LexiconCreator
- This base implementation does nothing. Subclasses can override this method
to prepare a lexicon in the expected format, which should then be found at
lexiconFilename.
- Overrides:
prepareLexicon in class LexiconCreator
- Throws:
java.io.IOException
main
public static void main(java.lang.String[] args)
throws java.lang.Exception
- Parameters:
args -
- Throws:
java.lang.Exception