That's one sentence; but the instruction is no more than 2 sentences, so 2 is fine. Let me make it two sentences for clarity.

English | 简体中文 | 日本語

🎧 Listen to samples · Quick start · Tools · FAQ

Audio Studio MCP gives Claude Code, Claude Desktop, Cursor and any other MCP client a sound studio that runs on your own Mac. Ask for background music or a sound effect in plain language; the model generates it locally with Stable Audio 3. Then it can loop it, normalize the loudness and export it for your game engine or video editor. There is no cloud service and no API key, and generating doesn't cost anything.

✨ Highlights

• Reproducible: every file gets a .json sidecar with its prompt, seed and settings.

• Prompting built in: the server teaches the client the official Stable Audio 3 prompt format, so a request like "make it sound like a creaky door" becomes a prompt the model understands.

• Same output as upstream: for the same seed, results are bit-for-bit identical to the official sa3 CLI.

Related MCP server: AbletonMCP

🎧 Hear it

All of these were generated by this server on an M1 Mac and were not edited by hand. Open the sample page →

MP3s are in docs/samples/, with prompts and seeds in samples.json.

🧭 How it works

flowchart LR
    U["You<br/>“make a cozy cafe BGM”"] --> C["Claude Code · Desktop<br/>Cursor · any MCP client"]
    C -- "MCP" --> S["audio-studio-mcp<br/>generate · edit · process"]
    S --> M["Stable Audio 3<br/>MLX on Apple Silicon"]
    M --> F["WAV / OGG / FLAC + .json<br/>ready for game & video"]

🚀 Quick start

Requirements:

• Apple Silicon Mac (M1 or later), 16 GB RAM recommended

• About 6 GB of disk for the model

• git and ffmpeg (brew install ffmpeg)

git clone https://github.com/kanaworksai-hue/audio-studio-mcp.git
cd audio-studio-mcp
./install.sh

The installer sets up uv if needed, clones the official Stable Audio 3 runtime into vendor/ at a tested commit, creates .venv/, and downloads the medium model (1.4B, ~5.9 GB) from Hugging Face. It finishes by printing the exact command to register the server:

claude mcp add --scope user audio-studio -- /path/to/audio-studio-mcp/.venv/bin/audio-studio-mcp

Add this to claude_desktop_config.json, .cursor/mcp.json or your client's MCP config:

{
  "mcpServers": {
    "audio-studio": {
      "command": "/path/to/audio-studio-mcp/.venv/bin/audio-studio-mcp"
    }
  }
}

Downloads are anonymous. Logging in with hf auth login raises Hugging Face's rate limit.

Then just ask:

• "Make a 60-second lo-fi BGM loop at 85 BPM for a cafe scene, give me 3 candidates."

• "I need a heavy wooden door slamming in a stone hallway. Export it as OGG for Godot."

• "Extend this track to 90 seconds and normalize it to −16 LUFS."

🧰 Tools

• audio_type: music / instrument / sfx / raw. The server adds the prefix the model was trained with (TrackType: Music, VocalType: Instrumental, …). Without the SFX prefix, sound effects often come out as music.

• variations: 1–4 candidates with seeds seed, seed+1, …. Use seed: -1 or leave it out for a random seed.

• sound_source / sound_action / sound_material / sound_space: describe a sound effect by what makes it, how it evolves, its texture and the space it is in.

• peak_ceiling_db (default −1): a safety ceiling. Louder peaks are turned down and quieter audio is never boosted. Use process_audio to normalize.

• process_audio normalization: peak is common for SFX. loudness normalizes to a target LUFS: −14 for streaming/video, −16 for game BGM, −23 for broadcast.

Generated files are 44.1 kHz 24-bit stereo WAV, written to outputs/ in this folder. Each one has a <name>.json next to it. Processed files keep the original generation info under source.

⚙️ Configuration

Pass them with claude mcp add ... -e NAME=value or in the env block of your client config.

Each client session normally starts its own server process. Models load on the first call and are freed after the idle timeout. To keep one warm process for every session, run it as a local HTTP server. It only listens on 127.0.0.1.

.venv/bin/audio-studio-mcp --http 8765 --preload
claude mcp add --scope user --transport http audio-studio http://127.0.0.1:8765/mcp

⚡ Performance

Medium model on an M1 Mac with 16 GB:

Loaded models use about 5–7 GB of memory. They are freed after the idle timeout or by unload_models.

📝 Prompting tips

1. English, 1–2 sentences, up to ~45 words. Avoid vocal words such as vocals, singer or lyrics.

2. Music: genre + adjectives → lead instruments → supporting layers → drums/rhythm → mood. Put the tempo in bpm.

   Lo-fi hip-hop chill track with mellow electric piano, soft vinyl crackle, low-pass filtered drums and soft plucked bass for a relaxed, dreamy vibe.

3. Sound effects: fill in source, action, material and space.

   source: glass bottle / action: shattering on impact, fragments scattering / material: thick green glass, sharp bright transient / space: concrete floor, close-up, dry

4. Match the length to the content: impacts 1–3 s, actions 3–6 s, ambience 6–15 s+, loops 6–20 s, full tracks 120–300 s (max 380 s).

5. Timing instructions like "climax at 15 s" are not followed reliably. Generate first, then use inpaint or continue.

6. The defaults steps=8 and cfg=1.0 are the official settings. For stricter prompt adherence, or to use negative_prompt, try cfg 3–7 (about 2× slower).

7. Results vary a lot by seed. Ask for 2–4 variations and pick the best.

The full guide is also sent to the client as the server's instructions.

❓ FAQ

Not yet. It needs an Apple Silicon Mac today. The model runtime sits behind a backend interface, and a Windows backend for NVIDIA GPUs is planned.

No. Stable Audio 3 generates instrumental music and sound effects. It cannot produce intelligible singing or lyrics.

The code here is MIT. The model weights are under the Stability AI Community License, which sets the terms for the model and what it generates. Read it before commercial use. At the time of writing it allows commercial use for individuals and organizations with less than US$1M in annual revenue.

It hasn't been tested. The medium model needs about 5–7 GB while loaded, so 16 GB is recommended. The small sm-music / sm-sfx models are lighter.

Use WAV or OGG for loops. MP3 adds padding at the start and end, so it can't loop seamlessly. Also try a longer loop_crossfade_ms, or ask the model to cut on a bar line.

🗺️ Roadmap

• Apple Silicon backend (Stable Audio 3, MLX)

• Windows backend for NVIDIA GPUs

• LoRA support (the upstream runtime already has it)

Issues and pull requests are welcome.

🛠️ Development

uv run pytest                              # offline tests, no model needed
.venv/bin/python scripts/smoke_test.py     # end-to-end with the real model (~30 s)

src/audio_studio_mcp/
  server.py            MCP tools (backend-independent)
  prompts.py           prompt building and the instructions sent to clients
  post.py              post-processing and export
  backends/base.py     backend interface
  backends/sa3_mlx.py  Stable Audio 3 on Apple Silicon

A new platform is added as another backend in backends/.

📄 License and credits

• This project: MIT.

• Stable Audio 3 runtime: downloaded into vendor/ at install time. It is © Stability AI and MIT-licensed.

• Model weights: licensed under the Stability AI Community License. They are downloaded from Hugging Face, not distributed here.

• Brand: the KanaWorks_AI name, logo and the Kana character are © KanaWorks_AI and are not covered by the MIT license. See docs/brand.

• Built on: Stable Audio 3, MLX and the Model Context Protocol.

This is an independent project. It is not affiliated with or endorsed by Stability AI, Apple or Anthropic.