Update app.py

app.py

@@ -2,7 +2,7 @@ import os
 import google.generativeai as genai
 import gradio as gr
 import requests
-from moviepy.editor import ImageClip, AudioFileClip
+from moviepy.editor import AudioFileClip, ImageClip, CompositeVideoClip
 
 # Configure Google Gemini API
 genai.configure(api_key=os.getenv("GEMINI_API_KEY"))
@@ -72,18 +72,27 @@ def text_to_speech(text):
     except Exception as e:
         return f"Error generating audio: {e}"
 
-# Function to create video from image and audio
-def generate_video(image_path, audio_path):
+# Function to create video from image, audio, and add logo overlay
+def create_video(image, audio):
     try:
-        if audio_path is None or "Error" in audio_path:
-            return "Error generating video: No valid audio file."
+        # Load the audio file
+        audio_clip = AudioFileClip(audio)
         
-        #image_clip = ImageClip(image_path).set_duration(AudioFileClip(audio_path).duration)
-        #audio_clip = AudioFileClip(audio_path)
-        #video_clip = image_clip.set_audio(audio_clip)
-        video_output_path = gr.make_waveform((16000, audio_path), bg_image=image_path)
-        #video_clip.write_videofile(video_output_path, codec="libx264", audio_codec="aac")
-        return video_output_path
+        # Load the main image and set its duration to match the audio
+        image_clip = ImageClip(image).set_duration(audio_clip.duration)
+        
+        # Load the logo image, resize it, and position it in the top-right corner
+        logo = ImageClip("Logo.png").resize(height=50)  # Adjust the height as needed
+        logo = logo.set_position(("right", "top")).set_duration(audio_clip.duration)
+        
+        # Create a composite video with the main image and the logo overlay
+        video_clip = CompositeVideoClip([image_clip, logo]).set_audio(audio_clip)
+        
+        # Save the video to a temporary file
+        output_path = "/tmp/output_video_with_logo.mp4"
+        video_clip.write_videofile(output_path, fps=30)
+        
+        return output_path
     except Exception as e:
         return f"Error generating video: {e}"
 
@@ -91,13 +100,13 @@ def generate_video(image_path, audio_path):
 def process_roast(image_path):
     roast_text = generate_roast(image_path)
     audio_path = text_to_speech(roast_text)
-    video_path = generate_video(image_path, audio_path)
+    video_path = create_video(image_path, audio_path)
     return roast_text, audio_path, video_path
 
 # Gradio Interface
 with gr.Blocks(theme=theme) as demo:
     gr.Markdown("# Image Roasting App with TTS and Video")
-    gr.Markdown("Upload an image, click 'Roast Image', and the AI will roast it, convert the roast to audio, and generate a video.")
+    gr.Markdown("Upload an image, click 'Roast Image', and the AI will roast it, convert the roast to audio, and generate a video with a logo overlay.")
 
     with gr.Row():
         image_input = gr.Image(type="filepath", label="Upload Image")