cmutosyllables.py

# dictionary: http://svn.code.sf.net/p/cmusphinx/code/trunk/cmudict/cmudict-0.7b

import string
import json

dictionary = {}
with open('cmudict-0.7b', 'r') as dictionary_file:
    for line in dictionary_file:
        # skip comment lines
        if (line[0] == ';'):
            continue

        word, token_string = line.split('  ')

        # skip words that start with punctuation (punctuation characters, or just trash)
        # skip words that end with punctuation (additional pronounciations, or trash)
        if word[0] in string.punctuation or word[-1] in string.punctuation:
            continue

        tokens = token_string.split(' ')

        syllables = [token for token in tokens if token[-1] in string.digits]
        dictionary[word.lower()] = len(syllables)

with open('cmudict-0.7b.json', 'w+') as json_file:
    json.dump(dictionary, json_file)