From f201ef509d6b6236f26dd54bae950343e5aa7d12 Mon Sep 17 00:00:00 2001
From: "Gan, Jimmy" <jimmy.gan@adidas.com>
Date: Wed, 3 Jun 2026 02:56:59 +0800
Subject: [PATCH] Enforce local audio extraction rules in transcribe.py

Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
---
 transcribe.py | 16 ++++++++++++++--
 1 file changed, 14 insertions(+), 2 deletions(-)

diff --git a/transcribe.py b/transcribe.py
index 0e6eedb..f5b13f9 100644
--- a/transcribe.py
+++ b/transcribe.py
@@ -1,13 +1,25 @@
 #!/usr/bin/env python3
 import sys
 import whisper
+import os
 
 if len(sys.argv) < 2:
-    print("Usage: python transcribe.py <audio_or_video_file> [language]")
+    print("Usage: python transcribe.py <audio_file> [language]")
     print("Example: python transcribe.py audio.mp3 es")
     sys.exit(1)
 
+filename = sys.argv[1]
+ext = os.path.splitext(filename)[1].lower()
+video_exts = {'.mp4', '.avi', '.mov', '.mkv', '.flv', '.wmv', '.webm'}
+
+if ext in video_exts:
+    print(f"Error: Video files ({ext}) are not allowed per DEVELOPMENT_RULES.md.")
+    print("Please extract audio first (e.g., using ffmpeg -i input.mp4 -vn -acodec libmp3lame output.mp3)")
+    sys.exit(1)
+
+print(f"Transcribing {filename}...")
 model = whisper.load_model("base")
 language = sys.argv[2] if len(sys.argv) > 2 else None
-result = model.transcribe(sys.argv[1], language=language)
+result = model.transcribe(filename, language=language)
+print("-" * 20)
 print(result["text"])