app.py

import os
import time
import requests
from huggingface_hub import login
import torch
import torchaudio
from einops import rearrange
import gradio as gr
from stable_audio_tools import get_pretrained_model
from stable_audio_tools.inference.generation import generate_diffusion_cond

# Authenticate Hugging Face Hub
token = os.getenv("HUGGINGFACE_TOKEN")
if not token:
    raise RuntimeError("HUGGINGFACE_TOKEN not set")
login(token=token, add_to_git_credential=False)

# Load audio model
device = "cuda" if torch.cuda.is_available() else "cpu"
model, config = get_pretrained_model("stabilityai/stable-audio-open-small")
model = model.to(device)
sample_rate = config["sample_rate"]
sample_size = config["sample_size"]

# Audio generation function
def generate_audio(prompt):
    conditioning = [{"prompt": prompt, "seconds_total": 11}]
    with torch.no_grad():
        output = generate_diffusion_cond(
            model,
            steps=8,
            conditioning=conditioning,
            sample_size=sample_size,
            device=device
        )
    output = rearrange(output, "b d n -> d (b n)")
    output = output.to(torch.float32).div(torch.max(torch.abs(output))).clamp(-1, 1).mul(32767).to(torch.int16).cpu()
    path = "output.wav"
    torchaudio.save(path, output, sample_rate)
    return path

# Image generation function using Replicate
def generate_image(prompt):
    replicate_token = os.getenv("REPLICATE_API_TOKEN")
    if not replicate_token:
        raise RuntimeError("REPLICATE_API_TOKEN not set")

    url = "https://api.replicate.com/v1/predictions"
    headers = {
        "Authorization": f"Token {replicate_token}",
        "Content-Type": "application/json"
    }
    data = {
        "version": "5ee6b41748a4e3e3d3a212ed4a29379d6a13b9265fd00fe59e28c2767a5e82eb",
        "input": {
            "prompt": prompt,
            "style": "surreal"
        }
    }
    response = requests.post(url, headers=headers, json=data)
    response.raise_for_status()
    prediction = response.json()

    status = prediction["status"]
    get_url = prediction["urls"]["get"]

    while status not in ["succeeded", "failed"]:
        time.sleep(1.5)
        resp = requests.get(get_url, headers=headers)
        prediction = resp.json()
        status = prediction["status"]
    
    if status != "succeeded":
        raise RuntimeError(f"Image generation failed: {prediction}")
    
    image_url = prediction["output"]
    image_path = "output.png"
    image_data = requests.get(image_url).content
    with open(image_path, "wb") as f:
        f.write(image_data)
    
    return image_path

# Combined generation function
def generate_assets(prompt):
    audio_path = generate_audio(prompt)
    image_path = generate_image(prompt)
    return audio_path, image_path

# Gradio UI
gr.Interface(
    fn=generate_assets,
    inputs=gr.Textbox(
        label="🎤 Prompt your sonic + visual art",
        placeholder="e.g. 'drunk driving with mario and yung lean'"
    ),
    outputs=[
        gr.Audio(type="filepath", label="🧠 Generated Audio"),
        gr.Image(type="filepath", label="🎨 Generated Image")
    ],
    title='🌐 Hot Prompts in Your Area: "My Husband Is Dead"',
    description="Enter a fun sound idea — generate audio *and* visual from one prompt.",
    examples=[
        "ghosts peeing",
        "Tech startup boss villain entrance music",
        "Dolphin hootin'"
    ]
).launch()