text/symbols.py

""" from https://github.com/keithito/tacotron """

'''
Defines the set of symbols used in text input to the model.
'''
_pad        = '_'
_punctuation = '+/…;:,.!?¡¿—-–<>*)…"«»“”~̃ '
_numbers = '0123456789'
_letters = 'ABCÇDEFGHIJKLMNOPQRSTUVWXYZÂÊÎÏÔÛâêîïôûéÉèåÅÈàÀüùÙÌìëöõabcçdefghijklmnopqrstuvwxyz'
_letters_ipa = "&°ɑɐɒæɓʙβɔɕçɗɖðʤəɘɚɛɜɝɞɟʄɡɠɢʛɦɧħɥʜɨɪʝɭɬɫɮʟɱɯɰŋɳɲɴøɵɸθœɶʘɹɺɾɻʀʁɽʂʃʈʧʉʊʋⱱʌɣɤʍχʎʏʑʐʒʔʡʕʢǀǁǂǃˈˌːˑʼʴʰʱʲʷˠˤ˞↓↑→↗↘'̩'ᵻ"
#_letters_ipa_lisn = " !,./0123456789:?@ACDEILMNOPSUXYZ_abcdefghijklmnoprstuvwxyz|~"

# Export all symbols:
symbols = [_pad] + list(_punctuation) + list(_letters) + list(_numbers) + list(_letters_ipa) #list(_letters_ipa_lisn)

# Special symbol ids
SPACE_ID = symbols.index(" ")