| from nltk.tokenize import word_tokenize | |
| special_combs = { | |
| "Dzs" : "Ј", "dzs" : "ј", | |
| "Dz" : "Ъ", "dz" : "ъ", #Actually the sound of ds in kids | |
| "Cs" : "Ч", "cs" : "ч", | |
| "Zs" : "Ж", "zs" : "ж", | |
| "Sz" : "С", "sz" : "с", | |
| "Ly" : "y", "ly" : "y" | |
| } | |
| hungarian_dict = { | |
| "á" : "a", "Á" : "A", | |
| "é" : "e", "É" : "E", | |
| "í" : "i", "Í" : "I", | |
| "ó" : "o", "Ó" : "O", | |
| "ö" : "a", "Ö" : "A", | |
| "ő" : "a", "Ő" : "A", | |
| "ú" : "u", "Ú" : "U", | |
| "ü" : "ю", "Ü" : "Ю", | |
| "ű" : "ю", "Ű" : "Ю", | |
| "j" : "y", "J" : "Y", | |
| "s" : "sh", "S": "Sh" | |
| } | |
| cyrillic_equiv_dict = { | |
| "ъ" : "ds", "ь" : "Ds", | |
| "ч" : "ch", "Ч" : "Ch", | |
| "ж" : "zh", "Ж" : "Zh", | |
| "ш" : "sh", "Ш" : "Sh", | |
| "ј" : "j" , "Ј" : "J", | |
| "ю" : "yu", "Ю" : "Yu", | |
| "с" : "s" , "С" : "S" | |
| } | |
| def check_special_comb(word): | |
| for comb in special_combs: | |
| if comb in word: | |
| word = word.replace(comb,special_combs[comb]) | |
| return word | |
| def hungarian_letter_to_eng(letter): | |
| if letter in hungarian_dict: | |
| return hungarian_dict[letter] | |
| else: | |
| return letter | |
| def cyrillic_to_eng(word): | |
| for cyrillic in cyrillic_equiv_dict: | |
| if cyrillic in word: | |
| word = word.replace(cyrillic,cyrillic_equiv_dict[cyrillic]) | |
| return word | |
| def hungarian_sentence_to_latin(word): | |
| assert type(word)==str, "Input must be a string" | |
| # print("Original word: ", word) | |
| word = check_special_comb(word) | |
| # print("Just after special combination replacement: -",word) | |
| word = ''.join([hungarian_letter_to_eng(letter) for letter in word]) | |
| # print("After regular word replacement: -",word) | |
| word = cyrillic_to_eng(word) | |
| # print("Simplified pronunciation: -",word) | |
| return word |