INNER CODE UNIT · Python

TextCleaner

devnen/Kitten-TTS-Server · engine.py:276

class TextCleaner:
    """Text cleaner for KittenTTS - converts text to token indices."""

    def __init__(self):
        _pad = "$"
        _punctuation = ';:,.!?¡¿—…"«»"" '
        _letters = "ABCDEFGHIJKLMNOPQRSTUVWXYZabcdefghijklmnopqrstuvwxyz"
        _letters_ipa = "ɑɐɒæɓʙβɔɕçɗɖðʤəɘɚɛɜɝɞɟʄɡɠɢʛɦɧħɥʜɨɪʝɭɬɫɮʟɱɯɰŋɳɲɴøɵɸθœɶʘɹɺɾɻʀʁɽʂʃʈʧʉʊʋⱱʌɣɤʍχʎʏʑʐʒʔʡʕʢǀǁǂǃˈˌːˑʼʴʰʱʲʷˠˤ˞↓↑→↗↘'̩'ᵻ"

        symbols = [_pad] + list(_punctuation) + list(_letters) + list(_letters_ipa)

        self.word_index_dictionary = {}
        for i in range(len(symbols)):
            self.word_index_dictionary[symbols[i]] = i

    def __call__(self, text: str):
        indexes = []
        for char in text:

View source record →

📰 Research Paper
Loading…
⏳ Fetching content…