From 810046e3df4910c8f5c1a90e4b53aca45b3397e8 Mon Sep 17 00:00:00 2001
From: wuhongsheng <664116298@qq.com>
Date: 星期一, 01 七月 2024 10:42:58 +0800
Subject: [PATCH] 优化merge segments 参数，解决新闻联播男女主持人“晚上好”合并一个speakid问题 (#1861)
---
funasr/tokenizer/korean_cleaner.py | 7 ++-----
1 files changed, 2 insertions(+), 5 deletions(-)
diff --git a/funasr/tokenizer/korean_cleaner.py b/funasr/tokenizer/korean_cleaner.py
index ee556d4..eceebe2 100644
--- a/funasr/tokenizer/korean_cleaner.py
+++ b/funasr/tokenizer/korean_cleaner.py
@@ -19,8 +19,7 @@
"9": "甑�",
}
new_text = "".join(
- number_to_kor[char] if char in number_to_kor.keys() else char
- for char in text
+ number_to_kor[char] if char in number_to_kor.keys() else char for char in text
)
return new_text
@@ -56,9 +55,7 @@
}
new_text = re.sub("[a-z]+", lambda x: str.upper(x.group()), text)
new_text = "".join(
- upper_alphabet_to_kor[char]
- if char in upper_alphabet_to_kor.keys()
- else char
+ upper_alphabet_to_kor[char] if char in upper_alphabet_to_kor.keys() else char
for char in new_text
)
--
Gitblit v1.9.1