paddlespeech.t2s.exps.ernie_sat.utils 模块

paddlespeech.t2s.exps.ernie_sat.utils.eval_durs(phns, target_lang: str = 'zh', fs: int = 24000, n_shift: int = 300)[来源]

paddlespeech.t2s.exps.ernie_sat.utils.get_dict(dictfile: str)[来源]

paddlespeech.t2s.exps.ernie_sat.utils.get_dur_adj_factor(orig_dur: List[int], pred_dur: List[int], phns: List[str])[来源]

paddlespeech.t2s.exps.ernie_sat.utils.get_span_bdy(mfa_start: List[float], mfa_end: List[float], span_to_repl: List[List[int]])[来源]

paddlespeech.t2s.exps.ernie_sat.utils.get_tmp_name(text: str)[来源]

paddlespeech.t2s.exps.ernie_sat.utils.get_voc_out(mel)[来源]

paddlespeech.t2s.exps.ernie_sat.utils.is_chinese(ch)[来源]

paddlespeech.t2s.exps.ernie_sat.utils.load_num_sequence_text(path: Union[Path, str], loader_type: str = 'csv_int') → Dict[str, List[Union[float, int]]][来源]

读取一个文本文件，指示数字序列

Examples:

key1 1 2 3 key2 34 5 6

>>> d = load_num_sequence_text('text')
>>> np.testing.assert_array_equal(d["key1"], np.array([1, 2, 3]))

paddlespeech.t2s.exps.ernie_sat.utils.read_2col_text(path: Union[Path, str]) → Dict[str, str][来源]

读取一个具有 2 列作为字典对象的文本文件。

Examples:

wav.scp:: key1 /some/path/a.wav key2 /some/path/b.wav

>>> read_2col_text('wav.scp')
{'key1': '/some/path/a.wav', 'key2': '/some/path/b.wav'}

paddlespeech.t2s.exps.ernie_sat.utils.str2md5(string)[来源]