stefantaubert / cmudict-parser

Python parser for CMUDict files. It returns ARBAbet and IPA transciption of dictionary words.

Hi 👋

• 🎓 I’m currently working on my PhD on speech synthesis
• 📫 How to reach me: [email protected]
• 🐍 You can have a look at my PyPI projects and Zenodo publications
• 🖖 My y.at is: 🐍💻🔬🤓
• 💬 Web apps: en-tts, zho-tts, pinyin-to-ipa

cmudict-parser's Issues

Assert Exception in sentence_to_ipa

The method sentence_to_ipa throws an exception for the input: "'Whose name is'--.

Conversion failed

Conversion failed for Suckin'--I mean helpin' people an' fightin' an' all that.

Traceback:

Traceback (most recent call last):
  File "/home/mi/.local/share/virtualenvs/tacotron2-vfc1XCWN/src/text-utils/text_utils/text.py", line 86, in en_to_ipa_cmu_epitran
    result = CMU_CACHE.sentence_to_ipa(
  File "/home/mi/.local/share/virtualenvs/tacotron2-vfc1XCWN/src/cmudict-parser/cmudict_parser/CMUDict.py", line 51, in sentence_to_ipa
    return get_ipa_of_sentence(self._entries_first_ipa, sentence, replace_unknown_with, use_caching)
  File "/home/mi/.local/share/virtualenvs/tacotron2-vfc1XCWN/src/cmudict-parser/cmudict_parser/SentenceToIPA.py", line 17, in sentence_to_ipa
    ipa_words = [get_ipa_of_word_in_sentence_cache(
  File "/home/mi/.local/share/virtualenvs/tacotron2-vfc1XCWN/src/cmudict-parser/cmudict_parser/SentenceToIPA.py", line 17, in <listcomp>
    ipa_words = [get_ipa_of_word_in_sentence_cache(
  File "/home/mi/.local/share/virtualenvs/tacotron2-vfc1XCWN/src/cmudict-parser/cmudict_parser/SentenceToIPA.py", line 30, in get_ipa_of_word_in_sentence_cache
    ipa = get_ipa_of_word_in_sentence(dict, word, replace_unknown_with)
  File "/home/mi/.local/share/virtualenvs/tacotron2-vfc1XCWN/src/cmudict-parser/cmudict_parser/SentenceToIPA.py", line 38, in get_ipa_of_word_in_sentence
    ipa = get_ipa_of_word_with_punctuation(dict, word, replace_unknown_with)
  File "/home/mi/.local/share/virtualenvs/tacotron2-vfc1XCWN/src/cmudict-parser/cmudict_parser/SentenceToIPA.py", line 50, in get_ipa_of_word_with_punctuation
    return ipa_of_punctuation_and_words_combined(dict, punctuation_before_word, word_without_punctuation, punctuation_after_word, replace_unknown_with)
  File "/home/mi/.local/share/virtualenvs/tacotron2-vfc1XCWN/src/cmudict-parser/cmudict_parser/SentenceToIPA.py", line 85, in ipa_of_punctuation_and_words_combined
    ipa_of_word_without_punct = f"{get_ipa_of_words_with_hyphen(dict, word_without_punctuation, replace_unknown_with)}{char_at_end}"
  File "/home/mi/.local/share/virtualenvs/tacotron2-vfc1XCWN/src/cmudict-parser/cmudict_parser/SentenceToIPA.py", line 107, in get_ipa_of_words_with_hyphen
    ipa = find_combination_of_certain_length_in_dict(
  File "/home/mi/.local/share/virtualenvs/tacotron2-vfc1XCWN/src/cmudict-parser/cmudict_parser/SentenceToIPA.py", line 119, in find_combination_of_certain_length_in_dict
    word, apos_before, apos_after = strip_apos_at_beginning_and_end_if_they_do_not_belong_to_word(
  File "/home/mi/.local/share/virtualenvs/tacotron2-vfc1XCWN/src/cmudict-parser/cmudict_parser/SentenceToIPA.py", line 130, in strip_apos_at_beginning_and_end_if_they_do_not_belong_to_word
    word, apos_before = strip_apos(word, 0)
  File "/home/mi/.local/share/virtualenvs/tacotron2-vfc1XCWN/src/cmudict-parser/cmudict_parser/SentenceToIPA.py", line 144, in strip_apos
    while word[pos] == "'":
IndexError: string index out of range

Exception on sentence to ipa for "A1"

see testcase in branch bugfix_numbers

"hier" was pushed

The folder "hier" was pushed. Please remove it from the repo.

AssertionError in SentenceToIPA

Please fix the error occuring on together, higgledy-piggledy, the.

Callback:

Exception has occurred: AssertionError
  File "/home/mi/code/cmudict-parser/cmudict_parser/SentenceToIPA.py", line 115, in get_ipa_of_words_with_hyphen
    assert ipa is not None
  File "/home/mi/code/cmudict-parser/cmudict_parser/SentenceToIPA.py", line 89, in ipa_of_punctuation_and_words_combined
    ipa_of_word_without_punct = f"{get_ipa_of_words_with_hyphen(dict, word_without_punctuation, replace_unknown_with)}{char_at_end}"
  File "/home/mi/code/cmudict-parser/cmudict_parser/SentenceToIPA.py", line 54, in get_ipa_of_word_with_punctuation
    return ipa_of_punctuation_and_words_combined(dict, punctuation_before_word, word_without_punctuation, punctuation_after_word, replace_unknown_with)
  File "/home/mi/code/cmudict-parser/cmudict_parser/SentenceToIPA.py", line 42, in get_ipa_of_word_in_sentence
    ipa = get_ipa_of_word_with_punctuation(dict, word, replace_unknown_with)
  File "/home/mi/code/cmudict-parser/cmudict_parser/SentenceToIPA.py", line 34, in get_ipa_of_word_in_sentence_cache
    ipa = get_ipa_of_word_in_sentence(dict, word, replace_unknown_with)
  File "/home/mi/code/cmudict-parser/cmudict_parser/SentenceToIPA.py", line 21, in <listcomp>
    ipa_words = [get_ipa_of_word_in_sentence_cache(
  File "/home/mi/code/cmudict-parser/cmudict_parser/SentenceToIPA.py", line 21, in sentence_to_ipa
    ipa_words = [get_ipa_of_word_in_sentence_cache(
  File "/home/mi/code/cmudict-parser/cmudict_parser/CMUDict.py", line 51, in sentence_to_ipa
    return get_ipa_of_sentence(self._entries_first_ipa, sentence, replace_unknown_with, use_caching)
  File "/home/mi/code/cmudict-parser/cmudict_parser/playground.py", line 9, in <module>
    res = cmu.sentence_to_ipa(x, replace_unknown_with="_")

Implement caching

Cache the IPA of already converted words.

IndexError on sentence with comma

The following text raised an error: it is not a real gain, for the modern printer throws the gain away by putting inordinately wide spaces between his lines, which, probably,

Traceback

Traceback (most recent call last):
  File "/home/mi/.local/share/virtualenvs/tacotron2-vfc1XCWN/src/text-utils/text_utils/text.py", line 86, in en_to_ipa_cmu_epitran
    result = CMU_CACHE.sentence_to_ipa(
  File "/home/mi/.local/share/virtualenvs/tacotron2-vfc1XCWN/src/cmudict-parser/cmudict_parser/CMUDict.py", line 63, in sentence_to_ipa
    ipa = self.get_ipa_of_word_in_sentence(word, replace_unknown_with)
  File "/home/mi/.local/share/virtualenvs/tacotron2-vfc1XCWN/src/cmudict-parser/cmudict_parser/CMUDict.py", line 70, in get_ipa_of_word_in_sentence
    ipa = self.get_ipa_of_words_with_punctuation(word, replace_unknown_with)
  File "/home/mi/.local/share/virtualenvs/tacotron2-vfc1XCWN/src/cmudict-parser/cmudict_parser/CMUDict.py", line 95, in get_ipa_of_words_with_punctuation
    if self.contains(word_with_apo_at_beginning) and punctuations_before_word[-1] == "'":
IndexError: string index out of range

Fix Indentation

In method: get_ipa_of_word_in_sentence