app.py

import gradio as gr
import torch
import torchaudio
from einops import rearrange
from stable_audio_tools import get_pretrained_model
from stable_audio_tools.inference.generation import generate_diffusion_cond

device = "cuda" if torch.cuda.is_available() else "cpu"

# Download model
model, model_config = get_pretrained_model("stabilityai/stable-audio-open-1.0")
sample_rate = model_config["sample_rate"]
sample_size = model_config["sample_size"]

model = model.to(device)

def generate_audio(prompt, bpm, duration):
    # Set up text and timing conditioning
    conditioning = [{
        "prompt": f"{bpm} BPM {prompt}",
        "seconds_start": 0, 
        "seconds_total": duration
    }]

    # Generate stereo audio
    output = generate_diffusion_cond(
        model,
        steps=100,
        cfg_scale=7,
        conditioning=conditioning,
        sample_size=sample_size,
        sigma_min=0.3,
        sigma_max=500,
        sampler_type="dpmpp-3m-sde",
        device=device
    )

    # Rearrange audio batch to a single sequence
    output = rearrange(output, "b d n -> d (b n)")

    # Peak normalize, clip, convert to int16, and save to file
    output = output.to(torch.float32).div(torch.max(torch.abs(output))).clamp(-1, 1).mul(32767).to(torch.int16).cpu()
    
    return sample_rate, output

inputs = [
    gr.inputs.Textbox(label="Prompt"),
    gr.inputs.Number(label="BPM", default=128),
    gr.inputs.Number(label="Duration (seconds)", default=30)
]

output = gr.outputs.Audio(type="numpy", label="Generated Audio")

gr.Interface(
    fn=generate_audio, 
    inputs=inputs,
    outputs=output,
    title="Stable Audio Generation",
    description="Generate audio using Stable Audio Open 1.0"
).launch()