Update app.py

app.py

@@ -6,7 +6,10 @@ from diffusers.utils import export_to_video
 from PIL import Image, ImageOps
 from gtts import gTTS
 from pydub import AudioSegment
-import whisper
+try:
+    import whisper
+except ImportError:
+    whisper = None
 import ffmpeg
 import requests
 from io import BytesIO
@@ -100,19 +103,33 @@ def generate_video(prompt, image_url):
     AudioSegment.from_mp3("voice.mp3").export("voice.wav", format="wav")
     
     # Step 6: Subtitles
-    model = whisper.load_model("base", device="cpu")
-    result = model.transcribe("voice.wav", task="transcribe", language="en")
-    
-    # Generate SRT subtitles manually since result["srt"] might not be available
-    srt_content = ""
-    for i, segment in enumerate(result["segments"]):
-        start_time = format_time(segment["start"])
-        end_time = format_time(segment["end"])
-        text = segment["text"].strip()
-        srt_content += f"{i + 1}\n{start_time} --> {end_time}\n{text}\n\n"
-    
-    with open("subtitles.srt", "w", encoding="utf-8") as f:
-        f.write(srt_content)
+    if whisper is not None:
+        try:
+            model = whisper.load_model("base", device="cpu")
+            result = model.transcribe("voice.wav", task="transcribe", language="en")
+            
+            # Generate SRT subtitles manually since result["srt"] might not be available
+            srt_content = ""
+            for i, segment in enumerate(result["segments"]):
+                start_time = format_time(segment["start"])
+                end_time = format_time(segment["end"])
+                text = segment["text"].strip()
+                srt_content += f"{i + 1}\n{start_time} --> {end_time}\n{text}\n\n"
+            
+            with open("subtitles.srt", "w", encoding="utf-8") as f:
+                f.write(srt_content)
+        except Exception as e:
+            print(f"Whisper transcription failed: {e}")
+            # Create a simple subtitle with the original prompt
+            srt_content = f"1\n00:00:00,000 --> 00:00:05,000\n{prompt}\n\n"
+            with open("subtitles.srt", "w", encoding="utf-8") as f:
+                f.write(srt_content)
+    else:
+        print("Whisper not available, using prompt as subtitle")
+        # Create a simple subtitle with the original prompt
+        srt_content = f"1\n00:00:00,000 --> 00:00:05,000\n{prompt}\n\n"
+        with open("subtitles.srt", "w", encoding="utf-8") as f:
+            f.write(srt_content)
     
     # Step 7: Merge video + audio + subtitles
     final_output = "final_with_audio.mp4"