Suggest words

haiku.py

from nltk import word_tokenize, ngrams, FreqDist
from nltk.corpus import cmudict
import string

f = open('poem.txt', 'r')
doc = f.readlines()
texts = ' '.join(doc)
texts = texts.replace('\n', ' ')
tokenized = word_tokenize(texts)

phoneme_dict = dict(cmudict.entries())

bigram_freq = FreqDist()
for bigram in ngrams(tokenized, 2):
	bigram_freq[bigram] += 1


bigram_dict = {}
bigram_list = list(bigram_freq)
for item in bigram_list:
	if item[0] in bigram_dict:
		bigram_dict[item[0]].append(item[1])
	else:
		bigram_dict[item[0]] = [item[1]]



def filter_words_by_syllable_count(arr, max=1):
	filtered = filter(lambda s: syllables_in_word(s) <= max, arr)
	return list(set(filtered))


def suggest_words_from_string(str, **kwargs):
	ret = filter(lambda s: s.lower().startswith(str), tokenized)
	syllables = kwargs.get('syllables')
	if syllables:
		ret = filter_words_by_syllable_count(ret, max=syllables)
	return list(set(ret))


def syllables_in_word(word):
    word = word.lower()
    count = 0
    vowels = "aeiouy"
    if word[0] in vowels:
        count += 1
    for index in range(1, len(word)):
        if word[index] in vowels and word[index - 1] not in vowels:
            count += 1
    if word.endswith("e") and not word.endswith('le'):
        count -= 1
    if count == 0:
        count += 1
    return count


def suggest_from_previous_word(prev_word, **kwargs):
	ret = bigram_dict[prev_word]
	syllables = kwargs.get('syllables')
	if syllables:
		ret = filter_words_by_syllable_count(ret, max=syllables)
	try:
		return ret
	except KeyError:
		return []