Upload 5 files

README.md

@@ -1,12 +1,17 @@
----
-title: Scriptyfa
-emoji: 🏃
-colorFrom: blue
-colorTo: red
-sdk: gradio
-sdk_version: 5.36.2
-app_file: app.py
-pinned: false
----
-
-Check out the configuration reference at https://huggingface.co/docs/hub/spaces-config-reference
+# Scriptify AI - Prompt to Image Generator (Free Hugging Face Space)
+
+A free, beautiful, and powerful AI app that turns text into high-quality images using Stable Diffusion XL.
+
+## Features:
+- Powered by SDXL 1.0
+- Custom UI with Gradio
+- Responsive layout
+- Works on mobile
+- Free & Unlimited (via Hugging Face)
+
+## Instructions:
+1. Upload to Hugging Face Space (select `Gradio` + `Public`)
+2. Make sure to install all requirements
+3. Start creating!
+
+> Made with 💙 by Ridam & ChatGPT (o4)

app.py

@@ -0,0 +1,47 @@
+import gradio as gr
+from utils import load_model, generate_image, prompt_presets
+
+pipe = load_model()
+
+with gr.Blocks(css="style.css") as demo:
+    gr.Markdown("# 🚀 Scriptify AI - Ultra Image Generator with ControlNet + Voice")
+    gr.Markdown("Create beautiful, controlled images using AI. Supports voice prompts, styles, and ControlNet!")
+
+    with gr.Row():
+        prompt_input = gr.Textbox(placeholder="e.g., cinematic shot of a samurai in rain", label="📝 Text Prompt", lines=2)
+        voice_input = gr.Audio(source="microphone", type="filepath", label="🎤 Or Speak Prompt")
+
+    with gr.Accordion("🔧 Advanced Settings", open=False):
+        with gr.Row():
+            style_dropdown = gr.Dropdown(choices=list(prompt_presets.keys()), value="None", label="🎨 Style Presets")
+            guidance = gr.Slider(minimum=1, maximum=20, value=7.5, label="🎚️ Guidance Scale")
+            steps = gr.Slider(minimum=10, maximum=50, value=30, label="🧠 Inference Steps")
+
+        with gr.Row():
+            width = gr.Slider(minimum=512, maximum=1024, step=64, value=768, label="🖼️ Width")
+            height = gr.Slider(minimum=512, maximum=1024, step=64, value=768, label="🖼️ Height")
+
+    generate_button = gr.Button("✨ Generate Image")
+    output_image = gr.Image(type="pil", label="🖼️ Result Image")
+    status = gr.Textbox(visible=False)
+
+    def run(prompt, voice, style, guidance, steps, width, height):
+        import whisper
+        if voice:
+            model = whisper.load_model("base")
+            result = model.transcribe(voice)
+            prompt = result["text"]
+
+        if style != "None":
+            prompt = f"{prompt}, {prompt_presets[style]}"
+        try:
+            image = generate_image(pipe, prompt, guidance, steps, width, height)
+            return image, ""
+        except Exception as e:
+            return None, f"⚠️ {str(e)}"
+
+    generate_button.click(fn=run,
+                          inputs=[prompt_input, voice_input, style_dropdown, guidance, steps, width, height],
+                          outputs=[output_image, status])
+
+demo.launch()

requirements.txt

@@ -0,0 +1,8 @@
+gradio==4.24.0
+torch
+diffusers
+transformers
+accelerate
+safetensors
+whisper
+ffmpeg

style.css

@@ -0,0 +1,15 @@
+body {
+  background-color: #0f172a;
+  color: white;
+  font-family: 'Segoe UI', sans-serif;
+}
+
+h1 {
+  color: #38bdf8;
+}
+
+.gr-button {
+  background-color: #2563eb !important;
+  border-radius: 12px !important;
+  font-weight: bold !important;
+}

utils.py

@@ -0,0 +1,27 @@
+from diffusers import DiffusionPipeline
+import torch
+
+prompt_presets = {
+    "Cinematic": "cinematic lighting, epic composition, 8k",
+    "Realistic Portrait": "realistic face, shallow depth of field, photography",
+    "Anime Style": "anime, cel-shading, crisp lines, colorful",
+    "Fantasy": "mythical, magical light, detailed, fantasy world",
+    "None": ""
+}
+
+def load_model():
+    print("Loading SDXL model...")
+    pipe = DiffusionPipeline.from_pretrained(
+        "stabilityai/stable-diffusion-xl-base-1.0",
+        torch_dtype=torch.float16 if torch.cuda.is_available() else torch.float32,
+        variant="fp16" if torch.cuda.is_available() else None
+    )
+    pipe.to("cuda" if torch.cuda.is_available() else "cpu")
+    return pipe
+
+def generate_image(pipe, prompt: str, guidance: float, steps: int, width: int, height: int):
+    if not prompt or len(prompt.strip()) < 5:
+        raise ValueError("Prompt too short. Please describe your idea better.")
+
+    result = pipe(prompt, guidance_scale=guidance, num_inference_steps=steps, height=height, width=width)
+    return result.images[0]