Update app.py

app.py

@@ -1,51 +1,40 @@
 import gradio as gr
 from transformers import pipeline
 from gtts import gTTS
+from moviepy.editor import ImageSequenceClip, CompositeVideoClip, ImageClip, AudioFileClip
 from PIL import Image, ImageDraw, ImageFont
-from moviepy.editor import CompositeVideoClip, concatenate_videoclips, ImageClip
 import numpy as np
 from scipy.io.wavfile import write
-from pydub import AudioSegment
-
 
 script_generator = pipeline("text-generation", model="gpt2", truncation=True, max_length=100)
 
+def generate_comedy_script(prompt):
+    script = script_generator(prompt)[0]['generated_text']
+    return script
 
 def text_to_speech(script):
-    if isinstance(script, list):
-        texts = [item['generated_text'] for item in script if isinstance(item, dict) and 'generated_text' in item]
-        script = " ".join(texts)
-        print("Final script for TTS:", script)  # Debugging line
-
-    if not script.strip():
-        raise ValueError("No text to convert to speech.")
-        
     tts = gTTS(text=script, lang='en')
     audio_file = 'output.mp3'
     tts.save(audio_file)
     return audio_file
 
-
 def generate_animation(script):
-    if isinstance(script, list):
-        texts = [item['generated_text'] for item in script if isinstance(item, dict) and 'generated_text' in item]
-        script = " ".join(texts)
-
     lines = script.split('. ')
-    clips = []
-    for line in lines:
+    frames = []
+    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))
-        img.save('/tmp/temp.png')
-        clips.append(ImageClip('/tmp/temp.png').set_duration(3))
-    final_video = concatenate_videoclips(clips, method="compose")
-    final_video.write_videofile("/tmp/final_video.mp4", fps=24)
+        frame_path = f'/tmp/frame_{i}.png'
+        img.save(frame_path)
+        frames.append(frame_path)
+    
+    clip = ImageSequenceClip(frames, fps=1)
+    clip.write_videofile("/tmp/final_video.mp4", fps=24)
     return "/tmp/final_video.mp4"
 
 
-
 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)
@@ -64,17 +53,30 @@ 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 = [TextClip(f"Kids Music: {theme}", fontsize=70, color='yellow', size=(800, 400), bg_color='blue', method='caption').set_duration(5)]
+    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)
     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 = script_generator(prompt)
-    print("Generated script:", script)  # Debugging line
+    script = generate_comedy_script(prompt)
     audio_file = text_to_speech(script)
     video_file = generate_animation(script)
-    return script, audio_file, video_file
+    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)
@@ -113,3 +115,4 @@ app.launch()
 
 
 
+