Update app.py

app.py

@@ -1,17 +1,13 @@
-!pip install transformers diffusers gtts moviepy safetensors
-
+import gradio as gr
 import torch
-
-# Continue with your code...
-
+from transformers import pipeline
 from diffusers import AnimateDiffPipeline, MotionAdapter, EulerDiscreteScheduler
 from diffusers.utils import export_to_gif
 from huggingface_hub import hf_hub_download
 from safetensors.torch import load_file
-from transformers import pipeline
 from gtts import gTTS
-from moviepy.editor import *
-from IPython.display import Video
+from moviepy.editor import VideoFileClip, AudioFileClip, concatenate_videoclips
+import os
 
 # Load the text generation model
 generator = pipeline('text-generation', model='distilgpt2')
@@ -20,55 +16,54 @@ def generate_text(prompt):
     response = generator(prompt, max_length=150, num_return_sequences=1)
     return response[0]['generated_text']
 
-# Text-to-speech conversion
 def text_to_speech(text, filename='output_audio.mp3'):
     tts = gTTS(text)
     tts.save(filename)
     return filename
 
-# Generate animation using AnimateDiffPipeline
 def create_animation(prompt, output_file='animation.gif'):
-    device = "cuda"
-    dtype = torch.float16
+    device = "cuda" if torch.cuda.is_available() else "cpu"
+    dtype = torch.float16 if device == "cuda" else torch.float32
     step = 4
     repo = "ByteDance/AnimateDiff-Lightning"
     ckpt = f"animatediff_lightning_{step}step_diffusers.safetensors"
     base = "emilianJR/epiCRealism"
 
-    # Load adapter and pipeline
     adapter = MotionAdapter().to(device, dtype)
     adapter.load_state_dict(load_file(hf_hub_download(repo, ckpt), device=device))
     pipe = AnimateDiffPipeline.from_pretrained(base, motion_adapter=adapter, torch_dtype=dtype).to(device)
     pipe.scheduler = EulerDiscreteScheduler.from_config(pipe.scheduler.config, timestep_spacing="trailing", beta_schedule="linear")
 
-    # Generate animation based on prompt
     output = pipe(prompt=prompt, guidance_scale=1.0, num_inference_steps=step)
     export_to_gif(output.frames[0], output_file)
 
     return output_file
 
-# Combine animation and audio into a video
 def create_video(animation_file, audio_file, output_file='output_video.mp4'):
     clip = VideoFileClip(animation_file)
     audio = AudioFileClip(audio_file)
     clip = clip.set_audio(audio)
     clip.write_videofile(output_file, fps=24)
+    return output_file
 
 def generate_educational_video(prompt):
-    # Step 1: Generate text from prompt
     generated_text = generate_text(prompt)
-    
-    # Step 2: Convert text to speech
     audio_file = text_to_speech(generated_text)
-    
-    # Step 3: Create animation based on prompt
     animation_file = create_animation(prompt)
-    
-    # Step 4: Assemble the video
-    create_video(animation_file, audio_file)
-    
-    # Step 5: Display the video
-    return Video("output_video.mp4", embed=True)
+    video_file = create_video(animation_file, audio_file)
+    return video_file
+
+# Define Gradio Interface
+def gradio_interface(prompt):
+    video_path = generate_educational_video(prompt)
+    return video_path
+
+interface = gr.Interface(
+    fn=gradio_interface,
+    inputs="text",
+    outputs="video",
+    title="Educational Video Generator",
+    description="Enter a prompt to generate a video."
+)
 
-# Example usage
-generate_educational_video("give me a jock?")
+interface.launch()