def process_utterance(self, utt):
#print('target nodes: %s' % utt.xpath(self.target_nodes))
# import pdb; pdb.set_trace()
for node in utt.xpath(self.target_nodes):
assert node.has_attribute(self.split_attribute)
to_split = node.get(self.split_attribute)
child_chunks = self.splitting_function(to_split)
for chunk in child_chunks:
#print '=='
#print(chunk)
child = Element(self.child_node_type)
child.set(self.split_attribute, chunk)
if self.add_token_classes:
token_class = self.classify_token(chunk)
#print(token_class)
child.set(self.class_attribute, token_class)
if self.add_safetext:
token_safetext = self.safetext_token(chunk)
child.set(self.safetext_attribute, token_safetext)
node.add_child(child)
def process_utterance(self, utt):
# print 'target nodes: %s'%(utt.xpath(self.target_nodes))
for node in utt.xpath(self.target_nodes):
assert node.has_attribute(self.split_attribute)
to_split = node.get(self.split_attribute)
#normalised = self.normalise(to_split)
normalised = self.normalise_ice(to_split)
child_chunks = self.splitting_function(normalised)
for chunk in child_chunks:
# print '=='
# print chunk
child = Element(self.child_node_type)
child.set(self.split_attribute, chunk)
if self.add_token_classes:
token_class = self.classify_token(chunk)
# print token_class
child.set(self.class_attribute, token_class)
if self.add_safetext:
token_safetext = self.safetext_token(chunk)
child.set(self.safetext_attribute, token_safetext)
node.add_child(child)
def process_utterance(self, utt):
for node in utt.xpath(self.target_nodes):
assert node.has_attribute(self.class_attribute)
assert node.has_attribute(self.target_attribute)
current_class = node.attrib[self.class_attribute]
if current_class in self.word_classes:
word = node.attrib[self.target_attribute]
pos = node.attrib.get(self.part_of_speech_attribute, None) # default to None
(pronunciation, method) = self.get_phonetic_segments(word, part_of_speech=pos)
node.set('phones_from', method)
NodeProcessors.add_syllable_structure(node, pronunciation, syllable_delimiter='|', syllable_tag='syllable', \
phone_tag='segment', pronunciation_attribute='pronunciation', stress_attribute='stress')
elif current_class in self.probable_pause_classes:
pronunciation = c.PROB_PAUSE # [c.PROB_PAUSE]
child = Element('segment')
child.set('pronunciation', pronunciation)
node.add_child(child)
elif current_class in self.possible_pause_classes:
pronunciation = c.POSS_PAUSE # [c.POSS_PAUSE]
child = Element('segment')
child.set('pronunciation', pronunciation)
node.add_child(child)
else:
sys.exit('Class "%s" not in any of word_classes, probable_pause_classes, possible_pause_classes')
def process_utterance(self, utt):
print('target nodes: %s' % (utt.xpath(self.target_nodes)))
node_count = 0
for node in utt.xpath(self.target_nodes):
node_count += 1
assert node.has_attribute(self.split_attribute)
text_to_normalise = node.get(self.split_attribute)
print(text_to_normalise)
normalised = self.normalizer.normalize(text_to_normalise)
# get tokenized arr from normalizer, don't split here again
child_chunks = self.splitting_function(normalised)
for chunk in child_chunks:
# print '=='
# print chunk
child = Element(self.child_node_type)
child.set(self.split_attribute, chunk)
if self.add_token_classes:
token_class = self.classify_token(chunk)
# print token_class
child.set(self.class_attribute, token_class)
if self.add_safetext:
token_safetext = self.safetext_token(chunk)
child.set(self.safetext_attribute, token_safetext)
node.add_child(child)
def process_utterance(self, utt):
for node in utt.xpath(self.target_nodes):
assert node.has_attribute(self.class_attribute)
assert node.has_attribute(self.target_attribute)
current_class = node.attrib[self.class_attribute]
if current_class in self.word_classes:
word = node.attrib[self.target_attribute]
children = self.get_phonetic_segments(word)
elif current_class in self.probable_pause_classes:
children = [c.PROB_PAUSE]
elif current_class in self.possible_pause_classes:
children = [c.POSS_PAUSE]
else:
sys.exit('Class "%s" not in any of word_classes, probable_pause_classes, possible_pause_classes')
for chunk in children:
child = Element(self.child_node_type)
child.set(self.output_attribute, chunk)
node.add_child(child)
def process_utterance(self, utt):
#print 'target nodes: %s'%(utt.xpath(self.target_nodes))
for node in utt.xpath(self.target_nodes):
assert node.has_attribute(self.split_attribute)
to_split = node.get(self.split_attribute)
child_chunks = self.splitting_function(to_split)
for chunk in child_chunks:
if chunk == "_END_" :
continue
child = Element(self.child_node_type)
child.set(self.split_attribute, chunk)
if self.add_token_classes:
token_class = self.classify_token(chunk)
child.set(self.class_attribute, token_class)
if self.add_safetext:
token_safetext = self.safetext_token(chunk)
child.set(self.safetext_attribute, token_safetext)
node.add_child(child)
