funasr2 paraformer biciparaformer contextuaparaformer

游雁

2023-12-13 7012ca2efc130103c4acd24e3678c7ae280f8db4

 funasr/tokenizer/phoneme_tokenizer.py

@@ -363,6 +363,7 @@
        non_linguistic_symbols: Union[Path, str, Iterable[str]] = None,
        space_symbol: str = "<space>",
        remove_non_linguistic_symbols: bool = False,
        **kwargs,
    ):
        if g2p_type is None:
            self.g2p = split_by_space