large punc model modelscope pipeline

mengzhe.cmz

2023-07-25 edcd1a7292ce5f7c4a05d4144cfb15ab102f93db

 funasr/datasets/preprocessor.py

@@ -11,7 +11,7 @@
import numpy as np
import scipy.signal
import soundfile

import jieba

from funasr.text.build_tokenizer import build_tokenizer
from funasr.text.cleaner import TextCleaner
@@ -659,7 +659,6 @@
        self.split_text_name = split_text_name
        self.seg_jieba = seg_jieba
        if self.seg_jieba:
            import jieba
            jieba.load_userdict(seg_dict_file)

    @classmethod