Open ftesser opened 12 years ago
This is the English TEXT to PHONEMES of TEXT=punto.
<?xml version="1.0" encoding="UTF-8"?> <maryxml xmlns="http://mary.dfki.de/2002/MaryXML" xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance" version="0.5" xml:lang="en-US"> <p> <voice name="cmu-slt-hsmm"> <s> <t g2p_method="lexicon" ph="' p u n - t @U" pos="VBN"> punto </t> <t g2p_method="" ph="null" pos="."> . </t> </s> </voice> </p> </maryxml>
The Italian and germ,an ones are different:
<?xml version="1.0" encoding="UTF-8"?> <maryxml xmlns="http://mary.dfki.de/2002/MaryXML" xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance" version="0.5" xml:lang="it"> <p> <s> <t g2p_method="lexicon" ph="' p u1 n - t o" pos="S"> punto </t> <t pos="FS"> . </t> </s> </p> </maryxml>
The result is that all the English sentences end with the phones "n u l l"
Moreover the English LTS rules are not applied because of the syllabifier.
This is the English TEXT to PHONEMES of TEXT=punto.
The Italian and germ,an ones are different:
The result is that all the English sentences end with the phones "n u l l"
Moreover the English LTS rules are not applied because of the syllabifier.