Update app.py

app.py

@@ -112,30 +112,27 @@ def merge_audio_video(video_filename, audio_filename, output_filename, overlay_a
         print(f"Error loading video or main audio clip: {e}")
         return None
 
-    # Create a list starting with the main audio clip
+    # Start with the main audio clip
     audio_clips = [main_audio_clip]
 
-    # If there's an overlay audio file, adjust its volume and add to the composite
     if overlay_audio_path and os.path.exists(overlay_audio_path):
         try:
+            # Load the overlay audio clip
             overlay_audio_clip = AudioFileClip(overlay_audio_path)
             # Adjust the overlay audio clip's volume to 20%
             overlay_audio_clip = overlay_audio_clip.volumex(0.2)
-            audio_clips.append(overlay_audio_clip.set_duration(main_audio_clip.duration))
+            # Ensure the overlay audio clip matches the main audio clip's duration
+            if overlay_audio_clip.duration > main_audio_clip.duration:
+                overlay_audio_clip = overlay_audio_clip.subclip(0, main_audio_clip.duration)
+            audio_clips.append(overlay_audio_clip)
         except Exception as e:
             print(f"Error processing overlay audio clip: {e}")
-            # Depending on your needs, you may choose to return None here or continue without the overlay
+            # Optionally handle the error or continue without the overlay
 
-    # Combine the audio clips
+    # Combine the audio clips into a composite
     composite_audio_clip = CompositeAudioClip(audio_clips)
 
-    # Ensure the composite audio clip does not exceed the video's duration
-    if composite_audio_clip.duration > video_clip.duration:
-        # Extend the video with the last frame if the audio is longer
-        last_frame = video_clip.to_ImageClip(t=video_clip.duration - 1).set_duration(composite_audio_clip.duration - video_clip.duration)
-        video_clip = concatenate_videoclips([video_clip, last_frame])
-
-    # Set the composite audio as the video's audio and write the output file
+    # Set the video's audio to the composite audio clip and write the output file
     final_clip = video_clip.set_audio(composite_audio_clip)
     final_clip.write_videofile(output_filename, codec='libx264', audio_codec="aac")
 
@@ -145,7 +142,49 @@ def merge_audio_video(video_filename, audio_filename, output_filename, overlay_a
     if 'overlay_audio_clip' in locals():
         overlay_audio_clip.close()
 
-    return output_filename
+    return output_filename    
+# def merge_audio_video(video_filename, audio_filename, output_filename, overlay_audio_path=None):
+#     try:
+#         video_clip = VideoFileClip(video_filename)
+#         main_audio_clip = AudioFileClip(audio_filename)
+#     except Exception as e:
+#         print(f"Error loading video or main audio clip: {e}")
+#         return None
+
+#     # Create a list starting with the main audio clip
+#     audio_clips = [main_audio_clip]
+
+#     # If there's an overlay audio file, adjust its volume and add to the composite
+#     if overlay_audio_path and os.path.exists(overlay_audio_path):
+#         try:
+#             overlay_audio_clip = AudioFileClip(overlay_audio_path)
+#             # Adjust the overlay audio clip's volume to 20%
+#             overlay_audio_clip = overlay_audio_clip.volumex(0.2)
+#             audio_clips.append(overlay_audio_clip.set_duration(main_audio_clip.duration))
+#         except Exception as e:
+#             print(f"Error processing overlay audio clip: {e}")
+#             # Depending on your needs, you may choose to return None here or continue without the overlay
+
+#     # Combine the audio clips
+#     composite_audio_clip = CompositeAudioClip(audio_clips)
+
+#     # Ensure the composite audio clip does not exceed the video's duration
+#     if composite_audio_clip.duration > video_clip.duration:
+#         # Extend the video with the last frame if the audio is longer
+#         last_frame = video_clip.to_ImageClip(t=video_clip.duration - 1).set_duration(composite_audio_clip.duration - video_clip.duration)
+#         video_clip = concatenate_videoclips([video_clip, last_frame])
+
+#     # Set the composite audio as the video's audio and write the output file
+#     final_clip = video_clip.set_audio(composite_audio_clip)
+#     final_clip.write_videofile(output_filename, codec='libx264', audio_codec="aac")
+
+#     # Cleanup
+#     video_clip.close()
+#     main_audio_clip.close()
+#     if 'overlay_audio_clip' in locals():
+#         overlay_audio_clip.close()
+
+#     return output_filename
     
 # def merge_audio_video(video_filename, audio_filename, output_filename, overlay_audio_file=None):
 #     try:
@@ -210,8 +249,18 @@ def main():
         st.error("OpenAI API key is not set in .env.local")
         return
 
-    uploaded_file = st.file_uploader("Select a video file", type=["mp4", "avi"])
-    overlay_audio_file = st.file_uploader("Upload overlay audio (optional)", type=["mp3", "wav"])
+    uploaded_video_file = st.file_uploader("Select a video file", type=["mp4", "avi"])
+    if uploaded_video_file is not None:
+        # Display a preview of the uploaded video file
+        st.video(uploaded_video_file)
+
+    uploaded_audio_file = st.file_uploader("Upload overlay audio (optional)", type=["mp3", "wav"])
+    if uploaded_audio_file is not None:
+        # Convert the uploaded audio file to bytes for st.audio to display
+        # Streamlit's st.audio requires the data to be in bytes
+        audio_bytes = uploaded_audio_file.read()
+        # Display a preview of the uploaded audio file
+        st.audio(audio_bytes, format='audio/wav')
 
     voice_options = {'Echo (Male)': 'echo', 'Fable (Male)': 'fable', 'Onyx (Male)': 'onyx', 'Nova (Female)': 'nova', 'Shimmer (Female)': 'shimmer', 'Alloy (Female)': 'alloy'}
     voice = st.selectbox('Choose the voice you want', list(voice_options.keys()))
@@ -270,6 +319,10 @@ def main():
             output_video_filename = os.path.splitext(video_filename)[0] + "_output.mp4"
             final_video_filename = merge_audio_video(video_filename, audio_filename, output_video_filename, overlay_audio_path)
 
+            # Display the generated script to the user
+            st.subheader("Generated Script")
+            st.write(text)  # Use st.text(text) if you prefer plain text without markdown
+            
             if final_video_filename:
                 st.video(final_video_filename)