Update app.py

app.py

@@ -5,14 +5,19 @@ from langchain.chains import LLMChain
 from openai import OpenAI
 import os
 import re
+from pathlib import Path
+from moviepy.editor import *
+import requests
+import tempfile
 
+ 
 # Initialize OpenAI client
 client = OpenAI()
 
 # Create a prompt template
 template = """
-You are a creative story writer. Given a topic, write a short story of about 150 words. 
-Divide the story into 3 paragraphs. Each paragraph should be a distinct part of the story.
+You are a creative teacher exclusively for kids. Given a topic, write a tutorial-based story of about 150 words teaching the child about the topic. 
+Divide the tutorial into 3 paragraphs. Each paragraph should be a distinct part of the tutorial.
 
 Topic: {topic}
 
@@ -30,7 +35,7 @@ def generate_image(prompt):
         response = client.images.generate(
             model="dall-e-3",
             prompt=prompt,
-            size="1024x1024",
+            size="256x256",
             quality="standard",
             n=1,
         )
@@ -39,7 +44,43 @@ def generate_image(prompt):
         print(f"Error generating image: {e}")
         return None
 
-def generate_story_with_images(topic):
+def generate_speech(text, filename):
+    try:
+        response = client.audio.speech.create(
+            model="tts-1",
+            voice="alloy",
+            input=text
+        )
+        response.stream_to_file(filename)
+        return filename
+    except Exception as e:
+        print(f"Error generating speech: {e}")
+        return None
+
+def download_image(url, filename):
+    response = requests.get(url)
+    with open(filename, 'wb') as f:
+        f.write(response.content)
+
+def create_video(paragraphs, image_files, audio_files, output_file):
+    clips = []
+    for paragraph, image_file, audio_file in zip(paragraphs, image_files, audio_files):
+        audio_clip = AudioFileClip(audio_file)
+        duration = audio_clip.duration
+        
+        image_clip = ImageClip(image_file).set_duration(duration)
+        
+        text_clip = TextClip(paragraph, fontsize=30, color='white', bg_color='black',
+                             size=(1024, 200), method='caption').set_position(('center', 'bottom')).set_duration(duration)
+        
+        composite_clip = CompositeVideoClip([image_clip, text_clip])
+        composite_clip = composite_clip.set_audio(audio_clip)
+        clips.append(composite_clip)
+    
+    final_clip = concatenate_videoclips(clips)
+    final_clip.write_videofile(output_file, fps=24)
+
+def generate_story_with_video(topic):
     # Generate the story
     story = story_chain.run(topic)
     
@@ -51,40 +92,50 @@ def generate_story_with_images(topic):
     while len(paragraphs) < 3:
         paragraphs.append("...")
 
-    # Generate images for each paragraph
-    image_urls = []
-    for paragraph in paragraphs:
-        image_url = generate_image(paragraph)
-        image_urls.append(image_url)
+    with tempfile.TemporaryDirectory() as temp_dir:
+        image_files = []
+        audio_files = []
+        
+        for i, paragraph in enumerate(paragraphs):
+            # Generate and download image
+            image_url = generate_image(paragraph)
+            image_file = os.path.join(temp_dir, f"image_{i}.png")
+            download_image(image_url, image_file)
+            image_files.append(image_file)
+            
+            # Generate audio
+            audio_file = os.path.join(temp_dir, f"audio_{i}.mp3")
+            generate_speech(paragraph, audio_file)
+            audio_files.append(audio_file)
+        
+        # Create video
+        output_file = os.path.join(temp_dir, "story_video.mp4")
+        create_video(paragraphs, image_files, audio_files, output_file)
+        
+        # Read the video file
+        with open(output_file, "rb") as f:
+            video_data = f.read()
     
-    return paragraphs, image_urls
+    return story, video_data
 
 # Create the Gradio interface
 def gradio_interface(topic):
     try:
-        paragraphs, image_urls = generate_story_with_images(topic)
-        return [
-            paragraphs[0], image_urls[0],
-            paragraphs[1], image_urls[1],
-            paragraphs[2], image_urls[2]
-        ]
+        story, video_data = generate_story_with_video(topic)
+        return story, video_data
     except Exception as e:
         print(f"Error in gradio_interface: {e}")
-        return ["An error occurred while generating the story and images."] + [""] * 5
+        return "An error occurred while generating the story and video.", None
 
 iface = gr.Interface(
     fn=gradio_interface,
     inputs=gr.Textbox(lines=2, placeholder="Enter a topic for the story..."),
     outputs=[
-        gr.Textbox(label="Paragraph 1"),
-        gr.Image(label="Image 1", type="filepath"),
-        gr.Textbox(label="Paragraph 2"),
-        gr.Image(label="Image 2", type="filepath"),
-        gr.Textbox(label="Paragraph 3"),
-        gr.Image(label="Image 3", type="filepath"),
+        gr.Textbox(label="Generated Story"),
+        gr.Video(label="Story Video")
     ],
-    title="Story Generator with Images",
-    description="Enter a topic, and the AI will generate a short story with an image for each paragraph."
+    title="Story Generator with Video",
+    description="Enter a topic, and the AI will generate a short story with a video combining images, captions, and narration."
 )
 
 # Launch the Gradio app