Upload 3 files

README.md

@@ -1,15 +1,39 @@
 ---
-title: Stable Audio 3.0 API
-emoji: ⚡
-colorFrom: gray
-colorTo: gray
+title: Stable Audio 3 Small
+emoji: 🎵
+colorFrom: purple
+colorTo: blue
 sdk: gradio
-sdk_version: 6.22.0
-python_version: '3.12'
+sdk_version: 6.3.0
 app_file: app.py
 pinned: false
-license: mit
-short_description: api
 ---
 
-Check out the configuration reference at https://huggingface.co/docs/hub/spaces-config-reference
+# Stable Audio 3 Small - Music Generation
+
+A CPU-based Hugging Face Space for generating music using [Stability AI's Stable Audio 3](https://github.com/Stability-AI/stable-audio-3) Small model.
+
+## Features
+
+- **Model**: `small-music` (433M parameters)
+- **Duration**: 1–120 seconds
+- **Sample rate**: 44.1 kHz stereo
+- **Runs on CPU** (no GPU required)
+- Gradio 6.3.0 interface
+
+## Usage
+
+1. Enter a text prompt describing the music you want
+2. Adjust duration, steps, CFG scale, and seed
+3. Click **Generate** and wait for the audio
+
+## Setup
+
+```bash
+pip install -r requirements.txt
+python app.py
+```
+
+## License
+
+This project uses the Stable Audio 3 model. Please refer to Stability AI's [license](https://huggingface.co/stabilityai/stable-audio-3-small) for usage terms.

app.py

@@ -0,0 +1,77 @@
+import os
+import torch
+import torchaudio
+import gradio as gr
+from einops import rearrange
+from stable_audio_3 import StableAudioModel
+
+# Load model once at startup
+print("Loading Stable Audio 3 Small model...")
+model = StableAudioModel.from_pretrained("small-music", device="cpu")
+print("Model loaded successfully!")
+
+
+def generate_audio(prompt, duration, steps, cfg_scale, seed):
+    print(f"Generating: prompt='{prompt}', duration={duration}s, steps={steps}, cfg={cfg_scale}, seed={seed}")
+
+    audio = model.generate(
+        prompt=prompt,
+        duration=duration,
+        steps=steps,
+        cfg_scale=cfg_scale,
+        seed=seed,
+        batch_size=1
+    )
+
+    # Post-process: (batch, channels, samples) -> stereo waveform
+    audio = rearrange(audio, "b d n -> d (b n)")
+    audio = audio.to(torch.float32).clamp(-1, 1).mul(32767).to(torch.int16).cpu()
+
+    output_path = "output.wav"
+    torchaudio.save(output_path, audio, 44100)
+    print("Generation complete!")
+    return output_path
+
+
+with gr.Blocks(title="Stable Audio 3 Small") as demo:
+    gr.Markdown("# 🎵 Stable Audio 3 Small - Music Generation")
+    gr.Markdown("Generate music using Stability AI's Stable Audio 3 Small model. Runs on CPU.")
+
+    with gr.Row():
+        with gr.Column():
+            prompt = gr.Textbox(
+                label="Prompt",
+                placeholder="Describe the music you want to generate...",
+                lines=2
+            )
+            duration = gr.Slider(
+                minimum=1, maximum=120, value=30, step=1,
+                label="Duration (seconds)"
+            )
+            steps = gr.Slider(
+                minimum=1, maximum=50, value=8, step=1,
+                label="Steps"
+            )
+            cfg_scale = gr.Slider(
+                minimum=0.0, maximum=10.0, value=1.0, step=0.1,
+                label="CFG Scale"
+            )
+            seed = gr.Number(
+                value=-1, label="Seed (-1 for random)"
+            )
+            btn = gr.Button("Generate", variant="primary")
+
+        with gr.Column():
+            audio_output = gr.Audio(
+                label="Generated Audio",
+                type="filepath"
+            )
+
+    btn.click(
+        fn=generate_audio,
+        inputs=[prompt, duration, steps, cfg_scale, seed],
+        outputs=audio_output
+    )
+
+
+demo.queue(max_size=4, default_concurrency_limit=1).launch()

requirements.txt

@@ -0,0 +1,7 @@
+stable-audio-3[ui]
+gradio==6.3.0
+soundfile>=0.13.1
+einops
+torch>=2.7.1
+torchaudio>=2.7.1
+huggingface-hub>=1.7.1