def add_from_dict(self, dictfilename):
"""
Add the list of phones from a pronunciation dictionary.
@param dictfilename (str) is the name of an HTK-ASCII pronunciation dictionary
"""
d = DictPron( dictfilename ).get_dict()
for value in d.values():
variants = value.split("|")
for variant in variants:
phones = variant.split("-")
for phone in phones:
self.add( phone )
def test_save(self):
d = DictPron( DICT_FRA )
d.save_as_ascii( DICT_FRA+".copy" )
d2 = DictPron( DICT_FRA+".copy", nodump=True )
for w in d.get_keys():
self.assertEqual( d.get_pron(w), d2.get_pron(w) )
os.remove( DICT_FRA+".copy" )
def gen_dependencies(self, grammarname, dictname):
"""
Generate the dependencies (grammar, dictionary) for HVite.
@param grammarname is the file name of the tokens
@param dictname is the dictionary file name
"""
dictpron = DictPron()
with codecs.open(grammarname, 'w', encoding) as flab:
for token,pron in zip(self._tokens.split(),self._phones.split()):
# dictionary:
for variant in pron.split("|"):
dictpron.add_pron( token, variant.replace("-"," ") )
if self._infersp is True:
variant = variant + '-sil'
dictpron.add_pron( token, variant.replace("-"," ") )
# lab file (one token per line)
flab.write( token+"\n")
dictpron.save_as_ascii( dictname )
def test_dict(self):
d = DictPron( DICT_FRA )
self.assertTrue( d.is_unk('azerty') )
self.assertFalse( d.is_unk('il_y_a') )
self.assertFalse( d.is_unk(u'être') )
self.assertEqual( d.get_pron(u'sil'), "s.i.l" )
self.assertEqual( d.get_pron(u'azerty'), "UNK" )
def test_phonetizeFR(self):
dictdir = os.path.join(SPPAS, "resources", "dict")
dictfile = os.path.join(dictdir, "fra.dict")
dd = DictPron(dictfile)
grph = DictPhon(dd)
result = grph.phonetize('pas_encore', phonunk=False)
self.assertEqual(result, 'UNK')
result = grph.phonetize('pas_encore', phonunk=True)
self.assertEqual(result,
"p.a.a~.k.o.r|p.a.z.a~.k.o.r|p.a.a~.k.o.r.eu|p.a.z.a~.k.o.r.eu")
result = grph.phonetize(u'/lemot/', phonunk=True)
self.assertEqual(result, u"lemot")
result = grph.phonetize(u'/lemot/', phonunk=False)
self.assertEqual(result, u"lemot")
def gen_slm_dependencies(self, basename, N=3):
"""
Generate the dependencies (slm, dictionary) for julius.
@param basename (str - IN) the base name of the slm file and of the dictionary file
@param N (int) Language model N-gram length.
"""
dictname = basename + ".dict"
slmname = basename + ".arpa"
phoneslist = self._phones.split()
tokenslist = self._tokens.split()
dictpron = DictPron()
for token,pron in zip(tokenslist,phoneslist):
for variant in pron.split("|"):
dictpron.add_pron( token, variant.replace("-"," ") )
if dictpron.is_unk(START_SENT_SYMBOL) is True:
dictpron.add_pron( START_SENT_SYMBOL, "sil" )
if dictpron.is_unk(END_SENT_SYMBOL) is True:
dictpron.add_pron( END_SENT_SYMBOL, "sil" )
dictpron.save_as_ascii( dictname, False )
# Write the SLM
model = NgramsModel(N)
model.append_sentences( [self._tokens] )
probas = model.probabilities( method="logml" )
arpaio = ArpaIO()
arpaio.set( probas )
arpaio.save( slmname )
