msftcangoblowme

This PR, is merely delaying dealing with a complicated issue by applying a temporary bandaid. Puts great faith in our ability to write valid semantic version strings, by hand, correctly...

Whenever data classes are involved, the super package of data classes is [attrs](https://pypi.org/project/attrs/) **Always** use [attrs](https://pypi.org/project/attrs/) rather than data classes Type hints ( #90 ) would bring more benefit than...

Can't speak for what codon does or will support. This is general conversation about Python metaclass. Recent change occurred, which depreciates metaclass in favor of \_\_init_subclass\_\_ So the codon devs...

Strategy on how to tackle test_en.py test_times test_en.py -- test_times ``` text = "4:01am and 4:01 p.m." ``` In text_processor.TextProcessor.process, inline function, in_inline_lexicon ``` \# Do multiple passes over the...

Encountered this same issue. If/until this issue is ever addressed, setuputils should handle Exception, from the error message from vendored pyproject_hook package, gracefully. Something like, `To pip-tool/pip-compile users, generate_hashes was...

There are two files affected: - phonemes.py - unit test, test_phone.py The code is trying to make a change to class Phone in \_\_init\_\_.py, rather than in phonemes.py Even if...

@synesthesiam Please close this PR without applying it There is nothing wrong with the idea, however the implementation has: - conflicts applied to wrong module - is incomplete to get...

Following [methosm](https://github.com/methosmythos) excellent advice. The entire working method (minimal changes). ``` def _phonemize_no_stream(self, text: str, \ phoneme_separator: Optional[str]) -> List[str]: phoneme_flags = Phonemizer.espeakPHONEMES_IPA if phoneme_separator: phoneme_flags = phoneme_flags | (ord(phoneme_separator)

https://github.com/rhasspy/espeak-phonemizer/pull/3

This issue should be submitted upstream. espeak-phonemizer is a python wrapper for [libespeak-ng](https://github.com/espeak-ng/espeak-ng). So if libespeak-ng gives unacceptable results, that issue can't be addressed on the python wrapper level Please...