def test_misaligned(self):
misaligned_pause = Pause('', 0.50, 0.25)
assert_true(misaligned_pause.misaligned)
phone_sil = Phone('SIL')
misaligned_pause._phones = [phone_sil]
assert_true(misaligned_pause.misaligned)
def test_multiple_medial_pauses(self):
words = self.words[:]
words[5:] = [
Pause('<SIL>', 0.91, 1.0),
Pause('<SIL>', 1.0, 1.42),
Word('mat', 1.42, 1.7, ['m', 'ae', 't'], ['m', 'ae', 't'])
]
utterances = list(words_to_utterances(words))
assert_equal(len(utterances), 2)
for i, word in enumerate(utterances[0]):
assert_equal(words[i].beg, word.beg)
assert_equal(words[i].end, word.end)
assert_equal(words[i].orthography, word.orthography)
assert_equal(words[i].phonemic, word.phonemic)
assert_equal(words[i].phonetic, word.phonetic)
for i, word in enumerate(utterances[1], 7):
assert_equal(words[i].beg, word.beg)
assert_equal(words[i].end, word.end)
assert_equal(words[i].orthography, word.orthography)
assert_equal(words[i].phonemic, word.phonemic)
assert_equal(words[i].phonetic, word.phonetic)
def test_speech_rate_pause_at_beg_and_middle(self):
words = [Pause(beg=0, end=0.39)] + self.words[2:]
words[2] = Pause(beg=0.55, end=0.73)
utt = Utterance(words)
# raises
assert_raises(ValueError, utt.speech_rate)
assert_raises(ValueError, utt.speech_rate, False, 'raises')
# zero
assert_equal(utt.speech_rate(no_syllables='zero'), 2.4)
# squeeze
assert_equal(utt.speech_rate(no_syllables='squeeze'), 3 / 0.86)
def test_speech_rate_pause_at_end_and_middle(self):
words = self.words + [Pause(beg=1.25, end=1.50)]
words[-3] = Pause(beg=0.73, end=0.80)
utt = Utterance(words)
# raises
assert_raises(ValueError, utt.speech_rate)
assert_raises(ValueError, utt.speech_rate, False, 'raises')
# zero
assert_equal(utt.speech_rate(no_syllables='zero'), 8 / 3)
# squeeze
assert_equal(utt.speech_rate(no_syllables='squeeze'), 4 / 1.25)
def test_strip_pause_only(self):
pause = Pause(beg=1.25, end=1.95)
utt_strip = Utterance([pause])
utt_strip.strip()
assert_equal(utt_strip.words(), [])
def test_sep_arg(self):
words = self.words[:]
words[5:] = [
Pause('<SIL>', 0.91, 1.0),
Word('mat', 1.0, 1.28, ['m', 'ae', 't'], ['m', 'ae', 't'])
]
utterances = list(words_to_utterances(words, sep=0.08))
assert_equal(len(utterances), 2)
assert_equal(len(utterances[0]), 5)
assert_equal(len(utterances[1]), 1)
for i, word in enumerate(utterances[0]):
assert_equal(words[i].beg, word.beg)
assert_equal(words[i].end, word.end)
assert_equal(words[i].orthography, word.orthography)
assert_equal(words[i].phonemic, word.phonemic)
assert_equal(words[i].phonetic, word.phonetic)
assert_equal(words[6].beg, utterances[1][0].beg)
assert_equal(words[6].end, utterances[1][0].end)
assert_equal(words[6].orthography, utterances[1][0].orthography)
assert_equal(words[6].phonemic, utterances[1][0].phonemic)
assert_equal(words[6].phonetic, utterances[1][0].phonetic)
def test_strip_beg_pause(self):
pause = Pause(beg=0, end=0.55)
utt_strip = Utterance([pause] + self.words[3:])
utt_strip.strip()
assert_not_in(pause, utt_strip)
assert_equal(utt_strip.words(), self.words[3:])
def test_strip_end_pause(self):
pause = Pause(beg=1.25, end=1.85)
utt_strip = Utterance(self.words + [pause])
utt_strip.strip()
assert_not_in(pause, utt_strip)
assert_equal(utt_strip.words(), self.words)
def test_multiple_short_medial_pauses(self):
words = self.words[:]
words[3:] = [
Pause('<VOCNOISE>', 0.59, 0.65),
Word('on', 0.65, 0.77, ['aa', 'n'], ['aa', 'n']), self.words[4],
Pause('<SIL>', 0.91, 1.35),
Word('mat', 1.35, 1.63, ['m', 'ae', 't'], ['m', 'ae', 't'])
]
utterances = list(words_to_utterances(words))
assert_equal(len(utterances), 1)
assert_equal(words[3].entry, utterances[0][3].entry)
assert_equal(words[6].entry, utterances[0][6].entry)
for i, word in enumerate(utterances[0]):
assert_equal(words[i].beg, word.beg)
assert_equal(words[i].end, word.end)
def test_initial_pause_no_strip(self):
initial_pause = Pause('<SIL>', 0.0, 0.05)
words = [initial_pause] + self.words
utterances = list(words_to_utterances(words, strip_pauses=False))
assert_equal(len(utterances), 1)
assert_equal(len(utterances[0]), 7)
for i, word in enumerate(utterances[0]):
assert_equal(words[i].beg, word.beg)
assert_equal(words[i].end, word.end)
def test_multiple_medial_pauses_no_strip(self):
words = self.words[:]
words[5:] = [
Pause('<SIL>', 0.91, 1.0),
Pause('<SIL>', 1.0, 1.42),
Word('mat', 1.42, 1.7, ['m', 'ae', 't'], ['m', 'ae', 't'])
]
utterances = list(words_to_utterances(words, strip_pauses=False))
assert_equal(len(utterances), 2)
assert_equal(len(utterances[0]), 7)
assert_equal(len(utterances[1]), 1)
for i, word in enumerate(utterances[0]):
assert_equal(words[i].beg, word.beg)
assert_equal(words[i].end, word.end)
for i, word in enumerate(utterances[1], 7):
assert_equal(words[i].beg, word.beg)
assert_equal(words[i].end, word.end)
def test_strip_end_multiple(self):
pause = Pause(beg=1.25, end=1.95)
zero = Word('', 1.95, 1.95)
utt_strip = Utterance()
utt_strip._words = self.words + [pause, zero]
utt_strip.strip()
for entry in {pause, zero}:
assert_not_in(entry, utt_strip)
assert_equal(utt_strip.words(), self.words)
def test_final_pause_no_strip(self):
final_pause = Pause('<SIL>', 1.19, 1.25)
words = self.words + [final_pause]
utterances = list(words_to_utterances(words, strip_pauses=False))
assert_equal(len(utterances), 1)
assert_equal(len(utterances[0]), 7)
for i, word in enumerate(utterances[0]):
assert_equal(words[i].beg, word.beg)
assert_equal(words[i].end, word.end)
def test_strip_beg_multiple(self):
pause = Pause(beg=0, end=0.39)
zero = Word('', 0.39, 0.39)
utt_strip = Utterance()
utt_strip._words = [pause, zero] + self.words[2:]
utt_strip.strip()
for entry in {pause, zero}:
assert_not_in(entry, utt_strip)
assert_equal(utt_strip.words(), self.words[2:])
def test_speech_rate_pause_at_end(self):
utt = Utterance(self.words)
utt.append(Pause(beg=1.25, end=1.50))
# raises
assert_raises(ValueError, utt.speech_rate)
assert_raises(ValueError, utt.speech_rate, False, 'raises')
# zero
assert_equal(utt.speech_rate(no_syllables='zero'), 10 / 3)
# squeeze
assert_equal(utt.speech_rate(no_syllables='squeeze'), 4.0)
def test_short_medial_pause(self):
words = self.words[:]
words[5:] = [
Pause('<SIL>', 0.91, 1.0),
Word('mat', 1.0, 1.28, ['m', 'ae', 't'], ['m', 'ae', 't'])
]
utterances = list(words_to_utterances(words))
assert_equal(len(utterances), 1)
assert_equal(words[5].entry, utterances[0][5].entry)
for i, word in enumerate(utterances[0]):
assert_equal(words[i].beg, word.beg)
assert_equal(words[i].end, word.end)
def test_medial_pause_no_strip(self):
# insert a pause into a copy of the word list
words = self.words[:]
words[5:] = [
Pause('<SIL>', 0.91, 1.42),
Word('mat', 1.42, 1.7, ['m', 'ae', 't'], ['m', 'ae', 't'])
]
utterances = list(words_to_utterances(words, strip_pauses=False))
assert_equal(len(utterances), 2)
assert_equal(len(utterances[0]), 6)
assert_equal(len(utterances[1]), 1)
for i, word in enumerate(utterances[0]):
assert_equal(words[i].beg, word.beg)
assert_equal(words[i].end, word.end)
def test_medial_pause(self):
# insert a pause into a copy of the word list
words = self.words[:]
words[5:] = [
Pause('<SIL>', 0.91, 1.42),
Word('mat', 1.42, 1.7, ['m', 'ae', 't'], ['m', 'ae', 't'])
]
utterances = list(words_to_utterances(words))
assert_equal(len(utterances), 2)
assert_equal(len(utterances[0]), 5)
assert_equal(len(utterances[1]), 1)
for i, word in enumerate(utterances[0]):
assert_equal(words[i].beg, word.beg)
assert_equal(words[i].end, word.end)
assert_equal(words[6].beg, utterances[1][0].beg)
assert_equal(words[6].end, utterances[1][0].end)
assert_equal(words[6].orthography, utterances[1][0].orthography)
assert_equal(words[6].phonemic, utterances[1][0].phonemic)
assert_equal(words[6].phonetic, utterances[1][0].phonetic)
def test_speech_rate_pause_at_end_raise(self):
utt = Utterance(self.words)
utt.append(Pause(beg=1.25, end=1.50))
assert_raises(ValueError, utt.speech_rate)
def test_speech_rate_pause_at_end_ignore(self):
utt = Utterance(self.words)
utt.append(Pause(beg=1.25, end=1.50))
assert_equal(utt.speech_rate(ignore_missing_syllables=True), 10 / 3)
def test_final_pause(self):
final_pause = Pause('<SIL>', 1.19, 1.25)
utterances = list(words_to_utterances(self.words + [final_pause]))
yield self.check_expected, utterances
def test_initial_pause(self):
initial_pause = Pause('<SIL>', 0.0, 0.05)
utterances = list(words_to_utterances([initial_pause] + self.words))
yield self.check_expected, utterances
def setup(self):
self.pause = Pause('<SIL>', 0.0, 0.05)
def test_speech_rate_pause_at_beg_and_middle_ignore(self):
words = [Pause(beg=0, end=0.39)] + self.words[2:]
words[2] = Pause(beg=0.55, end=0.73)
utt = Utterance(words)
assert_equal(utt.speech_rate(ignore_missing_syllables=True), 2.4)
def test_speech_rate_pause_at_end_and_middle_ignore(self):
words = self.words + [Pause(beg=1.25, end=1.50)]
words[-3] = Pause(beg=0.73, end=0.80)
utt = Utterance(words)
assert_equal(utt.speech_rate(ignore_missing_syllables=True), 8 / 3)
def test_str_pause(self):
utt = Utterance(self.words)
utt.append(Pause('VOCNOISE', 1.25, 1.50))
assert_equal(str(utt), '<Utterance "the cat is on the mat VOCNOISE">')
def test_empty_pause(self):
empty_pause = Pause()
assert_is_none(empty_pause.entry)
assert_is_none(empty_pause.beg)
assert_is_none(empty_pause.end)
def test_speech_rate_pause_at_end_and_middle_raise(self):
words = self.words + [Pause(beg=1.25, end=1.50)]
words[-3] = Pause(beg=0.73, end=0.80)
utt = Utterance(words)
assert_raises(ValueError, utt.speech_rate)
def test_misaligned_zero(self):
zero_pause = Pause('', 0.40, 0.40)
assert_false(zero_pause.misaligned)
def test_speech_rate_pause_at_beg_and_middle_raise(self):
words = [Pause(beg=0, end=0.39)] + self.words[2:]
words[2] = Pause(beg=0.55, end=0.73)
utt = Utterance(words)
assert_raises(ValueError, utt.speech_rate)