Fix

Dockerfile

@@ -23,8 +23,8 @@ RUN python -c "from sentence_transformers import SentenceTransformer; SentenceTr
 # Skip Whisper model preloading - it will be downloaded on first use
 # We don't need to preload it during build
 
-# Run NLTK downloads
-RUN python -m nltk.downloader punkt wordnet
+# Run NLTK downloads - add all needed NLTK data
+RUN python -c "import nltk; nltk.download('punkt'); nltk.download('wordnet'); nltk.download('omw-1.4')"
 
 # Cache the torch hub directory to avoid unnecessary downloads
 RUN mkdir -p /root/.cache/torch/hub

topic_segmenter.py

@@ -7,6 +7,16 @@ from keybert import KeyBERT
 import re
 from collections import defaultdict
 from sklearn.cluster import AgglomerativeClustering
+import logging
+
+# Ensure NLTK resources are available
+try:
+    import nltk
+    nltk.download('punkt', quiet=True)
+    nltk.download('wordnet', quiet=True)
+    nltk.download('omw-1.4', quiet=True)
+except Exception as e:
+    logging.warning(f"NLTK resource download issue: {str(e)}")
 
 class TopicSegmenter:
     def __init__(self, window_size=3, similarity_threshold=0.2, context_size=2, min_segment_size=2, topic_similarity_threshold=0.3,