speech_recognition/FunASR
1
0
Fork 0
mirror of https://github.com/modelscope/FunASR synced 2025-09-15 14:48:36 +08:00
Code Issues Actions Packages Projects Releases Wiki Activity

Merge pull request #167 from alibaba-damo-academy/dev_lhn

Browse Source 
fix text postprocess bug
This commit is contained in:
zhifu gao 2023-03-01 11:17:18 +08:00 committed by GitHub
commit 5d4b0c3994
No known key found for this signature in database
GPG Key ID: 4AEE18F83AFDEB23
1 changed files with 6 additions and 2 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

8
funasr/utils/postprocess_utils.py
View File

@ -6,7 +6,7 @@ from typing import Any, List, Union
def isChinese(ch: str): def isChinese(ch: str):
if '\u4e00' <= ch <= '\u9fff' or '\u0030' <= ch <= '\u0039': if '\u4e00' <= ch <= '\u9fff' or '\u0030' <= ch <= '\u0039' or ch == '@':
return True return True
return False return False
@ -17,6 +17,8 @@ def isAllChinese(word: Union[List[Any], str]):
cur = i.replace(' ', '') cur = i.replace(' ', '')
cur = cur.replace('</s>', '') cur = cur.replace('</s>', '')
cur = cur.replace('<s>', '') cur = cur.replace('<s>', '')
cur = cur.replace('<unk>', '')
cur = cur.replace('<OOV>', '')
word_lists.append(cur) word_lists.append(cur)
if len(word_lists) == 0: if len(word_lists) == 0:
@ -34,6 +36,8 @@ def isAllAlpha(word: Union[List[Any], str]):
cur = i.replace(' ', '') cur = i.replace(' ', '')
cur = cur.replace('</s>', '') cur = cur.replace('</s>', '')
cur = cur.replace('<s>', '') cur = cur.replace('<s>', '')
cur = cur.replace('<unk>', '')
cur = cur.replace('<OOV>', '')
word_lists.append(cur) word_lists.append(cur)
if len(word_lists) == 0: if len(word_lists) == 0:
@ -144,7 +148,7 @@ def sentence_postprocess(words: List[Any], time_stamp: List[List] = None):
else: else:
word = i.decode('utf-8') word = i.decode('utf-8')
if word in ['<s>', '</s>', '<unk>']: if word in ['<s>', '</s>', '<unk>', '<OOV>']:
continue continue
else: else:
middle_lists.append(word) middle_lists.append(word)