clamsproject · keighrim · Nov 11, 2025 · Nov 6, 2025
diff --git a/newshour-transcript-sync/golds/cpb-aacip-507-154dn40c26.tsv b/newshour-transcript-sync/golds/cpb-aacip-507-154dn40c26.tsv
diff --git a/newshour-transcript-sync/golds/cpb-aacip-507-1v5bc3tf81.tsv b/newshour-transcript-sync/golds/cpb-aacip-507-1v5bc3tf81.tsv
diff --git a/newshour-transcript-sync/golds/cpb-aacip-507-4746q1t25k.tsv b/newshour-transcript-sync/golds/cpb-aacip-507-4746q1t25k.tsv
diff --git a/newshour-transcript-sync/golds/cpb-aacip-507-4t6f18t178.tsv b/newshour-transcript-sync/golds/cpb-aacip-507-4t6f18t178.tsv
diff --git a/newshour-transcript-sync/golds/cpb-aacip-507-6h4cn6zk04.tsv b/newshour-transcript-sync/golds/cpb-aacip-507-6h4cn6zk04.tsv
diff --git a/newshour-transcript-sync/golds/cpb-aacip-507-6w96689725.tsv b/newshour-transcript-sync/golds/cpb-aacip-507-6w96689725.tsv
diff --git a/newshour-transcript-sync/golds/cpb-aacip-507-7659c6sk7z.tsv b/newshour-transcript-sync/golds/cpb-aacip-507-7659c6sk7z.tsv
diff --git a/newshour-transcript-sync/golds/cpb-aacip-507-9882j68s35.tsv b/newshour-transcript-sync/golds/cpb-aacip-507-9882j68s35.tsv
diff --git a/newshour-transcript-sync/golds/cpb-aacip-507-cf9j38m509.tsv b/newshour-transcript-sync/golds/cpb-aacip-507-cf9j38m509.tsv
diff --git a/newshour-transcript-sync/golds/cpb-aacip-507-n29p26qt59.tsv b/newshour-transcript-sync/golds/cpb-aacip-507-n29p26qt59.tsv
diff --git a/newshour-transcript-sync/golds/cpb-aacip-507-nk3610wp6s.tsv b/newshour-transcript-sync/golds/cpb-aacip-507-nk3610wp6s.tsv
diff --git a/newshour-transcript-sync/golds/cpb-aacip-507-pc2t43js98.tsv b/newshour-transcript-sync/golds/cpb-aacip-507-pc2t43js98.tsv
diff --git a/newshour-transcript-sync/golds/cpb-aacip-507-pr7mp4wf25.tsv b/newshour-transcript-sync/golds/cpb-aacip-507-pr7mp4wf25.tsv
diff --git a/newshour-transcript-sync/golds/cpb-aacip-507-r785h7cp0z.tsv b/newshour-transcript-sync/golds/cpb-aacip-507-r785h7cp0z.tsv
diff --git a/newshour-transcript-sync/golds/cpb-aacip-507-v11vd6pz5w.tsv b/newshour-transcript-sync/golds/cpb-aacip-507-v11vd6pz5w.tsv
diff --git a/newshour-transcript-sync/golds/cpb-aacip-507-v40js9j432.tsv b/newshour-transcript-sync/golds/cpb-aacip-507-v40js9j432.tsv
diff --git a/newshour-transcript-sync/golds/cpb-aacip-507-vd6nz81n6r.tsv b/newshour-transcript-sync/golds/cpb-aacip-507-vd6nz81n6r.tsv
diff --git a/newshour-transcript-sync/golds/cpb-aacip-507-vm42r3pt6h.tsv b/newshour-transcript-sync/golds/cpb-aacip-507-vm42r3pt6h.tsv
diff --git a/newshour-transcript-sync/golds/cpb-aacip-507-zk55d8pd1h.tsv b/newshour-transcript-sync/golds/cpb-aacip-507-zk55d8pd1h.tsv
diff --git a/newshour-transcript-sync/golds/cpb-aacip-507-zw18k75z4h.tsv b/newshour-transcript-sync/golds/cpb-aacip-507-zw18k75z4h.tsv
diff --git a/newshour-transcript-sync/process.py b/newshour-transcript-sync/process.py
@@ -194,7 +194,7 @@ def srt_to_tsv(srt_filename, gold_transcript_filename, tsv_filename, generate_ht
 
     with open(gold_transcript_filename, 'r', encoding='utf-8') as f:
         gold_content = f.read()
-    reference_text = gold_content.strip()
+    reference_text = gold_content
     reference_tokens = tokenize(reference_text) # These have real spans
 
     metrics, alignment = analyze_diff(hypothesis_tokens, reference_tokens)