def test_more_complicated_example(self):
sequence_align = SequenceAlignment(self.x8, self.y8)
editsteps, _ = sequence_align.alignment()
self.assertEqual(self.correct_editstep8, editsteps)
def test_findsolution_simplecase(self):
sequence_align = SequenceAlignment(self.x9, self.y9)
_, solution = sequence_align.alignment()
self.assertEqual(self.correct_solution9, solution)
def test_x_longer_than_y(self):
sequence_align = SequenceAlignment(self.x6, self.y6)
editsteps, _ = sequence_align.alignment()
self.assertEqual(self.correct_editstep6, editsteps)
def test_y_longer_than_x(self):
sequence_align = SequenceAlignment(self.x7, self.y7)
editsteps, _ = sequence_align.alignment()
self.assertEqual(self.correct_editstep7, editsteps)
def test_insert_elements(self):
sequence_align = SequenceAlignment(self.x4, self.y4)
editsteps, _ = sequence_align.alignment()
self.assertEqual(self.correct_editstep4, editsteps)
def test_remove_insert_align(self):
sequence_align = SequenceAlignment(self.x5, self.y5)
editsteps, _ = sequence_align.alignment()
self.assertEqual(self.correct_editstep5, editsteps)
def test_remove_to_empty(self):
sequence_align = SequenceAlignment(self.x3, self.y3)
editsteps, _ = sequence_align.alignment()
self.assertEqual(self.correct_editstep3, editsteps)
def test_remove(self):
sequence_align = SequenceAlignment(self.x2, self.y2)
editsteps, _ = sequence_align.alignment()
self.assertEqual(self.correct_editstep2, editsteps)
def test_simplecase(self):
sequence_align = SequenceAlignment(self.x1, self.y1)
editsteps, _ = sequence_align.alignment()
self.assertEqual(self.correct_editstep1, editsteps)
def test_findsolution_empty_x(self):
sequence_align = SequenceAlignment(self.x11, self.y11)
_, solution = sequence_align.alignment()
self.assertEqual(self.correct_solution11, solution)
]
return stressed_phoneme_chunks, stressed_phoneme
# Load the aligned grapheme/phoneme pairs
with open(
'../data/g2p_alignment/m2m_preprocessed_cmudict.txt.m-mAlign.2-2.delX.1-best.conYX.align'
) as infile:
aligned_grapheme_phoneme_pairs = [
line.strip().split('\t') for line in infile.readlines()
]
# Use the aligned grapheme/phoneme pairs to create a PronunciationDictionary
word_list = []
for m2m_grapheme, m2m_phoneme in aligned_grapheme_phoneme_pairs:
grapheme_chunks = m2m_grapheme_to_grapheme_chunks(m2m_grapheme)
grapheme = grapheme_chunks_to_grapheme_string(grapheme_chunks)
phoneme_chunks = m2m_phoneme_to_phoneme_chunks(m2m_phoneme)
stressed_phoneme_chunks, stressed_phoneme = phoneme_chunks_to_stressed_phoneme_chunks(
phoneme_chunks, grapheme)
grapheme_phoneme_aligment = SequenceAlignment(grapheme_chunks,
stressed_phoneme_chunks)
new_word = Word(grapheme, stressed_phoneme, grapheme_phoneme_aligment)
word_list.append(new_word)
# Save the PronunciationDictionary
PronunciationDictionary(word_list).save('../data/pronunciation_dictionary.pkl')
