From ed7e74583d4ca661e8828efd1641ebbc174b111f Mon Sep 17 00:00:00 2001
From: YILING0013 <xielingqing123@gmail.com>
Date: Sat, 1 Feb 2025 10:20:40 +0800
Subject: [PATCH] =?UTF-8?q?#13=20=E4=BF=AE=E6=AD=A3=E8=AF=A5=E9=94=99?=
 =?UTF-8?q?=E8=AF=AF?=
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

---
 README.md          | 4 ++--
 novel_generator.py | 2 +-
 2 files changed, 3 insertions(+), 3 deletions(-)

diff --git a/README.md b/README.md
index fc85b5f..41ec5bd 100644
--- a/README.md
+++ b/README.md
@@ -45,11 +45,11 @@
    ```bash
    pip install -r requirements.txt
    ```
-3. （可选）若需要手动安装 `nltk` 的 `punkt` 数据包：
+3. （可选）若需要手动安装 `nltk` 的 `punkt_tab` 数据包：
    ```bash
    python
    >>> import nltk
-   >>> nltk.download('punkt')
+   >>> nltk.download('punkt_tab')
    ```
 
 ---
diff --git a/novel_generator.py b/novel_generator.py
index dd02e31..779b509 100644
--- a/novel_generator.py
+++ b/novel_generator.py
@@ -583,7 +583,7 @@ def advanced_split_content(content: str,
     """
     将文本先按句子切分，然后根据语义相似度进行合并，最后根据max_length进行二次切分。
     """
-    nltk.download('punkt', quiet=True)  # 确保 punkt 数据可用
+    nltk.download('punkt_tab', quiet=True)
     sentences = nltk.sent_tokenize(content)
 
     if not sentences: