def test_denasalise_nasal_assimilation(self):
self.assertEqual(transcribe('mąka'), 'mɔŋka')
self.assertEqual(transcribe('gęsty'), 'ɡɛ̃w̃stɨ')
self.assertEqual(transcribe('rąbać'), 'rɔmbatɕ')
self.assertEqual(transcribe('podjęli'), 'pɔdʲjɛlʲi')
self.assertEqual(transcribe('stanął'), 'stanɔw')
self.assertEqual(transcribe('pieniądz'), 'pʲjɛɲɔɳts')
self.assertEqual(transcribe('pędzić'), 'pɛɲdʑitɕ')
self.assertEqual(transcribe('pójdą'), 'pujdɔ̃w̃')
self.assertEqual(transcribe('kręty'), 'krɛntɨ')
self.assertEqual(transcribe('pójdę'), 'pujdɛ')
def transcribe_text_Cracow(text):
# lowercase all letters
text = text.lower()
# extract alphabetic substrings
words = re.findall(r'[a-ząćęłńóśźż]+', text)
# phonemic transcription of words taken individually (regional processes not applied)
ph_words = []
if words:
for word in words:
ph_words.append(transcribe(word))
# pre-vocalic voicing and pre-voiced-obstruent voicing
for i in range(len(ph_words) - 1):
if (ph_words[i][-2:] in voi_dict_rev and
(ph_words[i + 1][0] in vocs or ph_words[i + 1][0] in voicing_dict
or ph_words[i + 1][0] in ipa_vowels)):
ph_words[i] = (ph_words[i][:-2]) + (voi_dict_rev[
(ph_words[i][-2:])])
elif ((ph_words[i][-1]) in voi_dict_rev and
(ph_words[i + 1][0] in vocs or ph_words[i + 1][0] in voicing_dict
or ph_words[i + 1][0] in ipa_vowels)):
ph_words[i] = (ph_words[i][:-1]) + (voi_dict_rev[
(ph_words[i][-1])])
# extract non-alphabetic substrings
non_words = re.findall(r'[^a-ząćęłńóśźż]+', text)
# put transcriptions of alphabetic strings and non-alphabetic substrings back together
res = []
maxi = max([len(ph_words), len(non_words)])
if maxi == len(ph_words) and len(ph_words) != len(non_words):
for i in range(len(ph_words) - 1):
res.append(ph_words[i])
res.append(non_words[i])
res.append(ph_words[-1])
elif len(ph_words) == len(non_words):
if text[0].isalpha():
for i in range(len(ph_words)):
res.append(ph_words[i])
res.append(non_words[i])
else:
for i in range(len(ph_words)):
res.append(non_words[i])
res.append(ph_words[i])
elif maxi == len(non_words) and len(ph_words) != len(non_words):
for i in range(len(non_words) - 1):
res.append(non_words[i])
res.append(ph_words[i])
res.append(non_words[-1])
result = ''.join(res)
return result
def test_final_devoicing(self):
self.assertEqual(transcribe('standard'), 'standart')
self.assertEqual(transcribe('wąż'), 'vɔ̃w̃ʂ')
self.assertEqual(transcribe('bug'), 'buk')
def surface_palatalisation(self):
self.assertEqual(transcribe('kibitki'), 'kʲibʲitkʲi')
self.assertEqual(transcribe('kwiatki'), 'kfʲjatkʲi')
self.assertEqual(transcribe('kolie'), 'kɔlʲjɛ')
self.assertEqual(transcribe('robi'), 'rɔbʲi')
def test_regressive_devoicing(self):
self.assertEqual(transcribe('babka'), 'bapka')
self.assertEqual(transcribe('rozpacz'), 'rɔspatʂ')
def test_progressive_voicing(self):
self.assertEqual(transcribe('poczdam'), 'pɔdʐdam')
self.assertEqual(transcribe('prośba'), 'prɔʑba')
def test_progressive_devoicing(self):
self.assertEqual(transcribe('świat'), 'ɕfʲjat')
self.assertEqual(transcribe('przetrzyma'), 'pʂɛtʂɨma')
self.assertEqual(transcribe('wycwanić'), 'vɨtsfaɲitɕ')
def test_progressive_voicing_morphological(self):
self.assertEqual(transcribe('jakże'), 'jaɡʐɛ')
self.assertEqual(transcribe('wszakżeż'), 'fʂaɡʐɛʂ')
self.assertEqual(transcribe('paśże'), 'paʑʐɛ')