Spaces:
Building
Building
File size: 3,103 Bytes
c9f9492 836b0d2 c9f9492 836b0d2 c9f9492 abc6394 c9f9492 836b0d2 c9f9492 f5e28c7 c9f9492 836b0d2 c9f9492 f5e28c7 c9f9492 dd969eb |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 |
# Define a list of question adverbs
opened_question_adverbs = ["how", "when", "where", "why", "how much", "how many", "how often", "how long", "what", "which", "who", "whose", "whom"]
# time adverbs to be moved at the beginning of ASL Gloss sentences
time_words = ["yesterday", "today", "tomorrow", "now", "before", "after", "morning", "afternoon", "evening", "night"]
# ASL glossing rules implemented in functions
def gloss_word(word):
return word.upper()
def handle_fingerspelling(word):
return '-'.join(list(word.upper()))
def handle_lexicalized_fingerspelling(word):
return f"#{word.upper()}"
def handle_repetition(word, count):
return f"{word.upper()}{'+' * (count - 1)}" if count > 1 else word.upper()
def handle_role_shift(sentence):
return f"rs {sentence}"
def handle_indexing(token, index):
return f"ix_{index} {token.upper()}"
def gloss_sentence(doc):
glossed_sentence = []
for token in doc:
glossed_word = gloss_word(token.text)
glossed_sentence.append(glossed_word)
return " ".join(glossed_sentence)
def add_time_indicator(gloss_sentence_):
for word in gloss_sentence_:
if word.text.lower() in time_words:
return f"{word.text.upper()} {gloss_sentence_.replace(word.text.upper(), '').strip()}"
return gloss_sentence_
# skip stop_words
def skip_stop_words(word):
if word.lower() == 'the' or word.lower() == 'a':
return ''
else:
return word
def question_type(doc):
try:
if doc[-1].text == '?':
if doc[0].text.lower() in opened_question_adverbs:
return "wh-question"
else:
return "yes-no-question"
return None
except IndexError:
return None
def process_sentence(doc):
nms = {
"wh-question": "wh-q",
"yes-no-question": "y/n-q"
}
classifiers = {
"car": "CL:3",
"person": "CL:1"
}
basic_verbs = {
"is": "BE",
"am": "BE",
"are": "BE",
"was": "BE",
"were": "BE",
"be": "BE"
}
glossed_sentence = []
for token in doc:
word = token.lemma_.lower()
if token.pos_ == "PROPN":
glossed_word = token.text.upper()
elif word in ["i", "me"]:
glossed_word = handle_indexing("I", 1)
elif word in ["you"]:
glossed_word = handle_indexing("YOU", 2)
elif word in classifiers:
glossed_word = classifiers[word]
elif word in basic_verbs:
glossed_word = basic_verbs[word]
else:
glossed_word = gloss_word(word)
glossed_word = skip_stop_words(glossed_word)
glossed_sentence.append(glossed_word)
for gloss in glossed_sentence:
if gloss.lower() in time_words:
glossed_sentence.insert(0, glossed_sentence.pop(glossed_sentence.index(gloss)))
break
type_doc = question_type(doc)
if type_doc != None:
glossed_sentence.insert(0, nms[type_doc])
return " ".join(glossed_sentence) |