How to use the epitran.xsampa.XSampa function in epitran

language/script pair that should be loaded
            preproc (bool): apply preprocessors
            postproc (bool): apply prostprocessors
            ligatures (bool): use precomposed ligatures instead of standard IPA
            cedict_filename (str): path to file containing the CC-CEDict
                                   dictionary; relevant only for Chinese
            rev (boolean): if True, load reverse transliteration
            rev_preproc (bool): if True, apply preprocessor when reverse transliterating
            rev_postproc (bool): if True, apply postprocessor when reverse transliterating
        """
        if code in self.special:
            self.epi = self.special[code](ligatures=ligatures, cedict_file=cedict_file)
        else:
            self.epi = SimpleEpitran(code, preproc, postproc, ligatures, rev, rev_preproc, rev_postproc)
        self.ft = panphon.featuretable.FeatureTable()
        self.xsampa = XSampa()
        self.puncnorm = PuncNorm()

def __init__(self, lang_script_codes, cedict_file=None):
        """Construct a Backoff object.

        Args:
            lang_script_codes (list): codes for languages to try, starting
            with the highest priority languages
            cedict_file (str): path to the CC-CEdict dictionary file
            (necessary only when cmn-Hans or cmn-Hant are used)
        """
        self.langs = [_epitran.Epitran(c, cedict_file=cedict_file)
                      for c in lang_script_codes]
        self.num_re = re.compile(r'\p{Number}+')
        self.ft = panphon.featuretable.FeatureTable()
        self.xsampa = XSampa()
        self.puncnorm = PuncNorm()
        self.dias = [StripDiacritics(c) for c in lang_script_codes]