def prosodic(self):
if not hasattr(self,'_prosodic'):
import prosodic as p
p.config['print_to_screen']=0
self._prosodic=pd={}
numlines=len(self.lined)
for _i,(i,line) in enumerate(sorted(self.lined.items())):
line=line.replace('-',' ').replace("'","").strip()
pd[i]=p.Text(line,meter=self.meter)
return self._prosodic
def getSymEnglish():
PHO = []
with open("english.txt", encoding="utf-8") as f:
content = f.read().splitlines()
for line in content:
x = p.Text(line)
x.parse()
PAR = x.bestParses()[0]
PHO.extend(x.phonemes())
print(PHO)
PHO = set(PHO)
return PHO
def meters(line):
p.config['print_to_screen'] = False
final_jsons = []
mline = p.Text(line)
mline.parse()
i = mline.bestParses()
final_json = {}
if i != [] and i[0] is not None:
meter = i[0].str_meter()
final_json['meter'] = meter
# final_json.clear()
else:
final_json['meter'] = '???'
return final_json
def parse_files(meter, path='*.txt'):
files = glob.glob(path)
for f in files:
terminal_out = sys.stdout
sys.stdout = open('parses/parse_{}_{}'.format(meter, f), 'w')
printable = set(string.printable)
file_text = '\n'.join(
[line.rstrip().split('\t')[0] for line in open(f, 'r')])
file_text = filter(lambda x: x in printable, file_text)
t = p.Text(file_text, meter=meter)
t.parse()
t.scansion()
sys.stdout.close()
sys.stdout = terminal_out
def compute(year):
f_names = os.path.join('..', 'data', 'social_security_info',
'yob%s.txt' % year)
f_save = os.path.join(save_dest, 'yob%s.csv' % year)
if os.path.exists(f_save):
print("Already completed", year)
return False
df = pd.read_csv(f_names,
header=None,
names=["name", "gender", "counts"],
encoding='utf-8')
data = []
#df = df[:5]
for _, row in tqdm(df.iterrows()):
name = row["name"]
sx = p.Text(name).syllables()
item = {}
item["name"] = name
item["gender"] = row["gender"]
item["counts"] = row["counts"]
item["n_syllables"] = len(sx)
item['IPA'] = ' '.join(map(str, sx))
if not item["IPA"]:
continue
item['IPA_stress'] = ''.join(
['1' if x[0] in "'`" else '0' for x in item["IPA"].split()])
#print(item)
data.append(item)
df = pd.DataFrame(data).set_index(["name", "gender"])
df = df.sort_values("counts", ascending=False)
df.to_csv(f_save, encoding='utf-8')
print(df)
print(f_save)
def go_ans(event):
output_text.delete(1.0, 'end')
text=input_text.get(1.0, 'end-1c')
t = p.Text(text,meter='default_english')
t.parse()
txt=[]
for parse in t.bestParses():
# print(parse)
txt.append(str(parse))
syllablegram=fsyllablegram(txt)
# print(syllablegram)
to_print=dinoit(syllablegram)
# print(to_print)
for line in to_print:
for word in line:
output_text.insert('end',word+' ')
output_text.insert('end','\n')
def normEng(eng, delemit):
x = p.Text(eng)
x.parse()
PAR = str(x.bestParses()[0]).split("|")
SYL = x.syllables()
if len(PAR) != len(SYL):
print("check dif len: ", eng)
result = ""
for i, syl in enumerate(SYL):
syllable = str(syl).replace("'", "").replace("ː", "").replace("ɑ", "a")
if PAR[i].lower().upper() == PAR[i]:
result += syllable + "'5" + " "
else:
result += syllable + "'1" + " "
result = result.rstrip(" ")
if delemit != "":
takemore = ""
for r in result:
if r in symbols:
takemore += delemit + r
result = takemore
return result
def mark_line(content):
t = p.Text(content)
words = t.words()
results = []
for i, word in enumerate(words):
if word.isMonoSyllab():
result = mark_syllable(syl_text(word.syllables()[0]), 'P')
# print (i, word.token.decode('utf-8'), word.syllables()[0], result)
else:
vowel_cluster_positions = find_vowel_clusters(word)
syl_count = len(word.syllables())
token = word.token.decode('utf-8')
ok = len(vowel_cluster_positions) == syl_count
if not ok:
revise_cluster_positions(vowel_cluster_positions, syl_count,
token)
ok = len(vowel_cluster_positions) == syl_count
if ok:
result = mark_lexical_stress_from_vowel_clusters(
word, vowel_cluster_positions)
else:
result = mark_lexical_stress_by_prosodic(
word, vowel_cluster_positions)
if result == token:
# As a last resort, just mark the first vowel in the word.
ok = False
result = mark_syllable(token, 'P')
if ok:
pass
# print (i, vowel_clusters_ct, syl_count, vowel_cluster_positions, token, result)
else:
print('##############', i, syl_count, vowel_cluster_positions,
token, result)
results.append(result)
return u' '.join(results)
#encoding=utf-8 import prosodic as p p.config['print_to_screen'] = True input_text = u"""Let me not to the marriage of true minds Admit impediments. Love is not love Which alters when it alteration finds, Or bends with the remover to remove. O no, it is an ever-fixèd mark That looks on tempests and is never shaken; It is the star to every wand'ring bark, Whose worth's unknown, although his height be taken. Love's not Time's fool, though rosy lips and cheeks Within his bending sickle's compass come; Love alters not with his brief hours and weeks, But bears it out even to the edge of doom. If this be error and upon me proved, I never writ, nor no man ever loved.""" # input_text is some string text = p.Text(input_text) text.parse()
import prosodic as p
t = p.Text('sonnet1.txt')
t.parse(meter='prose_rhythm_iambic_violable')
for a in t.allParses():
print(a)
for a in t.stats():
print(a)
import prosodic as p
#t = p.Text('have you reckoned a thousand acres much?')
t = p.Text('corpora/corppoetry_fi/fi.koskenniemi.txt')
t.more()
print
print ">> printing all heavy syllables..."
print t.feature('+prom.weight', True)
def is_ntV(word):
phonemes = word.phonemes()
if phonemes[-3:-1] != [p.Phoneme("n"), p.Phoneme("t")]:
return False
return phonemes[-1].feature("syll")
print
print ">> printing all -ntV words..."
nta = [word for word in t.words() if is_ntV(word)]
print nta