def cmu2ipa(tok):
import lexconvert
res=lexconvert.convert(tok,'cmu','unicode-ipa-syls')
## BUG FIXES
if tok.endswith(' T') and not res.endswith('t'): res=res+'t'
return res
def cmu2ipa(tok):
import lexconvert
res=lexconvert.convert(str(tok),'cmu','unicode-ipa')
## BUG FIXES
if tok.endswith(' T') and not res.endswith('t'): res=res+'t'
return res
def init_run_mary(self,text):
import lexconvert,bs4
numwords = 0
stanza=self.newchild()
line=stanza.newchild()
for stanzatext in text.split('\n\n'):
stanzatext=stanzatext.strip()
if not stanzatext: continue
for linetext in stanzatext.split('\n'):
linetext=linetext.strip()
if not linetext: continue
wordlist=linetext.split()
for i,word in enumerate(wordlist):
p0,word,p1=gleanPunc2(word)
if p0 and not line.empty(): line.finish()
if not word: continue
if stanza.finished: stanza = self.newchild()
if line.finished: line = stanza.newchild()
if self.dict.has(word):
words=self.dict.get(word)
for w in words: w.origin='cmu'
else:
## make word from openmary
wordxml=bs4.BeautifulSoup(openmary(word))
sylls=[]
for syll in wordxml.find_all('syllable'):
syllstr="'" if syll.get('stress',None) else ""
syllstr+=lexconvert.convert(syll['ph'],'x-sampa','unicode-ipa')
sylls+=[syllstr]
from Phoneme import Phoneme
if len(sylls)>1 and not True in [Phoneme(phon).isVowel() for phon in sylls[0]]:
sylls=[sylls[0]+sylls[1]]+ (sylls[2:] if len(sylls)>2 else [])
pronounc='.'.join(sylls)
words=[ self.dict.make((pronounc,[]), word) ]
for w in words: w.origin='openmary'
line.newchild(words)
if p1 and not line.empty(): line.finish()
numwords+=1
if not line.empty(): line.finish()
if not line.empty(): line.finish()
if not stanza.empty(): stanza.finish()
def sampa2ipa(sampa):
import lexconvert
if sampa == 'r=':
return u'ɛː'
return lexconvert.convert(sampa, 'x-sampa', 'unicode-ipa')
sampa = sampa.replace(u'?', u'')
if not sampa: return sampa
if ':' in sampa:
colon = u'ː'
sampa = sampa.replace(u':', u'')
else:
colon = u''
if not sampa in d_sampa2ipa and len(sampa) == 2:
#print "?!!",sampa
return sampa2ipa(sampa[0]) + sampa2ipa(sampa[1])
else:
return d_sampa2ipa[sampa] + colon
def sampa2ipa(sampa): import lexconvert if sampa == 'r=': return u'ɛː' return lexconvert.convert(sampa,'x-sampa','unicode-ipa') sampa=sampa.replace(u'?',u'') if not sampa: return sampa if ':' in sampa: colon=u'ː' sampa=sampa.replace(u':',u'') else: colon=u'' if not sampa in d_sampa2ipa and len(sampa)==2: #print "?!!",sampa return sampa2ipa(sampa[0]) + sampa2ipa(sampa[1]) else: return d_sampa2ipa[sampa] + colon
def ipa2cmu(tok): import lexconvert return lexconvert.convert(str(tok),'unicode-ipa','cmu')
def espeak2cmu(tok): import lexconvert return lexconvert.convert(str(tok),'espeak','cmu')
def cmu2ipa(tok):
import lexconvert
return lexconvert.convert(tok, 'cmu', 'unicode-ipa')
def ipa2cmu(tok): import lexconvert return lexconvert.convert(tok,'unicode-ipa','cmu')
def espeak2cmu(tok): import lexconvert return lexconvert.convert(tok,'espeak','cmu')
def cmu2ipa(tok): import lexconvert return lexconvert.convert(tok,'cmu','unicode-ipa')