Update app.py

app.py

@@ -1,12 +1,11 @@
 import gradio as gr
 from transformers import pipeline
 from gtts import gTTS
-from moviepy.editor import ImageSequenceClip, CompositeVideoClip, ImageClip, AudioFileClip, VideoFileClip
+from moviepy.editor import ImageSequenceClip, CompositeVideoClip, ImageClip, AudioFileClip, VideoFileClip, TextClip
 from PIL import Image, ImageDraw, ImageFont
 import numpy as np
 from scipy.io.wavfile import write
 
-# Load and Initialize Models
 script_generator = pipeline("text-generation", model="gpt2", truncation=True, max_length=100)
 
 def generate_comedy_script(prompt):
@@ -19,22 +18,49 @@ def text_to_speech(script):
     tts.save(audio_file)
     return audio_file
 
-def generate_animation(script):
+def create_images_from_script(script):
     lines = script.split('. ')
-    frames = []
+    image_paths = []
     for i, line in enumerate(lines):
         img = Image.new('RGB', (800, 400), color=(0, 0, 0))
         d = ImageDraw.Draw(img)
         fnt = ImageFont.load_default()
         d.text((10, 180), line, font=fnt, fill=(255, 255, 255))
-        frame_path = f'/tmp/frame_{i}.png'
-        img.save(frame_path)
-        frames.append(frame_path)
+        image_path = f'/tmp/image_{i}.png'
+        img.save(image_path)
+        image_paths.append(image_path)
+    return image_paths
+
+def generate_animation_with_images(script):
+    image_paths = create_images_from_script(script)
+    clips = []
+    for i, img_path in enumerate(image_paths):
+        text_clip = TextClip(script.split('. ')[i], fontsize=40, color='white', size=(800, 400), method='caption')
+        text_clip = text_clip.set_duration(3).set_position(('center', 'bottom'))
+        
+        image_clip = ImageClip(img_path).set_duration(3).set_position(('center', 'center'))
+        
+        composite_clip = CompositeVideoClip([image_clip, text_clip])
+        clips.append(composite_clip)
     
-    clip = ImageSequenceClip(frames, fps=1)
-    clip.write_videofile("/tmp/final_video.mp4", fps=24)
+    final_video = concatenate_videoclips(clips, method="compose")
+    final_video.write_videofile("/tmp/final_video.mp4", fps=24)
     return "/tmp/final_video.mp4"
 
+def combine_audio_video(video_path, audio_path):
+    video = VideoFileClip(video_path)
+    audio = AudioFileClip(audio_path)
+    final_video = video.set_audio(audio)
+    final_video.write_videofile("/tmp/final_comedy_video.mp4", fps=24)
+    return "/tmp/final_comedy_video.mp4"
+
+def generate_comedy_and_animation(prompt):
+    script = generate_comedy_script(prompt)
+    audio_file = text_to_speech(script)
+    video_file = generate_animation_with_images(script)
+    final_video = combine_audio_video(video_file, audio_file)
+    return script, audio_file, final_video
+
 def generate_sine_wave(frequency, duration, sample_rate=44100, amplitude=0.5):
     t = np.linspace(0, duration, int(sample_rate * duration), False)
     wave = amplitude * np.sin(2 * np.pi * frequency * t)
@@ -54,30 +80,20 @@ def generate_kids_music(theme, output_music_file="kids_music.wav"):
 def generate_kids_animation_with_music(theme, output_video_file="kids_animation.mp4"):
     music_file = generate_kids_music(theme)
     clips = []
-    img = Image.new('RGB', (800, 400), color=(0, 0, 255))
-    d = ImageDraw.Draw(img)
-    fnt = ImageFont.load_default()
-    d.text((10, 180), f"Kids Music: {theme}", font=fnt, fill=(255, 255, 0))
-    img.save('/tmp/kids_temp.png')
-    clips.append(ImageClip('/tmp/kids_temp.png').set_duration(5))
-    video = CompositeVideoClip(clips)
-    video.write_videofile(output_video_file, fps=24)
+    for i in range(5):
+        img = Image.new('RGB', (800, 400), color=(0, 0, 255))
+        d = ImageDraw.Draw(img)
+        fnt = ImageFont.load_default()
+        d.text((10, 180), f"Kids Music: {theme}", font=fnt, fill=(255, 255, 0))
+        frame_path = f'/tmp/kids_temp_{i}.png'
+        img.save(frame_path)
+        clips.append(ImageClip(frame_path).set_duration(1).set_position(('center', 'center')))
+    
+    final_video = CompositeVideoClip(clips, size=(800, 400))
+    final_video = final_video.set_audio(AudioFileClip(music_file))
+    final_video.write_videofile(output_video_file, fps=24)
     return output_video_file, music_file
 
-def combine_audio_video(video_path, audio_path):
-    video = VideoFileClip(video_path)
-    audio = AudioFileClip(audio_path)
-    final_video = video.set_audio(audio)
-    final_video.write_videofile("/tmp/final_comedy_video.mp4", fps=24)
-    return "/tmp/final_comedy_video.mp4"
-
-def generate_comedy_and_animation(prompt):
-    script = generate_comedy_script(prompt)
-    audio_file = text_to_speech(script)
-    video_file = generate_animation(script)
-    final_video = combine_audio_video(video_file, audio_file)
-    return script, audio_file, final_video
-
 def generate_kids_content(theme):
     video_file, music_file = generate_kids_animation_with_music(theme)
     return music_file, video_file
@@ -110,10 +126,5 @@ with gr.Blocks() as app:
             outputs=[kids_music_audio, kids_music_video]
         )
 
-app.launch()
-
-
-
-
-
 
+app.launch()