Add model card and manifest for Synesthesia ONNX models

README.md

@@ -0,0 +1,92 @@
+---
+license: apache-2.0
+language:
+  - en
+tags:
+  - music-generation
+  - audio
+  - onnx
+  - directml
+  - synesthesia
+  - magenta
+  - performance-rnn
+  - musicvae
+  - ddsp
+library_name: onnxruntime
+pipeline_tag: audio-to-audio
+---
+
+# Synesthesia — AI Music Models
+
+ONNX model weights for [Synesthesia](https://github.com/kryptodogg/synesthesia), a cyber-physical synthesizer and 3D/4D signal workstation.
+
+## Models
+
+| Model | Source | Format | Size | Task |
+|-------|--------|--------|------|------|
+| Performance RNN | Magenta | ONNX | ~20MB | Note-level MIDI generation |
+| MusicVAE (Encoder) | Magenta | ONNX | ~80MB | Latent music encoding |
+| MusicVAE (Decoder) | Magenta | ONNX | ~80MB | Latent music decoding |
+| DDSP (Encoder) | Magenta | ONNX | ~30MB | Audio → harmonic params |
+| DDSP (Decoder) | Magenta | ONNX | ~30MB | Harmonic params → audio |
+| SpectroStream (Encoder) | Magenta RT | ONNX | TBD | Audio → spectral tokens |
+| SpectroStream (Decoder) | Magenta RT | ONNX | TBD | Spectral tokens → audio |
+| MusicCoCa (Text) | Google | ONNX | TBD | Text → music embedding |
+| MusicCoCa (Audio) | Google | ONNX | TBD | Audio → music embedding |
+| Gemma-3N | Google | ONNX | TBD | Vision → mood/energy JSON |
+
+## Runtime
+
+All models run locally via **ONNX Runtime with DirectML** (GPU acceleration on Windows).
+
+```toml
+# Cargo.toml
+[dependencies]
+ort = { version = "2", features = ["directml"] }
+```
+
+## Download
+
+```python
+from huggingface_hub import snapshot_download
+snapshot_download("Ashiedu/Synesthesia", local_dir="./models")
+```
+
+```rust
+// Rust (using hf-hub crate)
+use hf_hub::api::sync::Api;
+let api = Api::new().unwrap();
+let repo = api.model("Ashiedu/Synesthesia".to_string());
+let model_path = repo.get("perfrnn/model.onnx").unwrap();
+```
+
+## Structure
+
+```
+├── perfrnn/
+│   └── model.onnx
+├── musicvae/
+│   ├── encoder.onnx
+│   └── decoder.onnx
+├── ddsp/
+│   ├── encoder.onnx
+│   └── decoder.onnx
+├── spectrostream/
+│   ├── encoder.onnx
+│   └── decoder.onnx
+├── musiccoca/
+│   ├── text.onnx
+│   └── audio.onnx
+├── gemma3n/
+│   └── model.onnx
+└── manifest.json
+```
+
+## License
+
+Apache 2.0 — model weights may have additional upstream licenses (see individual model directories).
+
+## Links
+
+- **GitHub:** [kryptodogg/synesthesia](https://github.com/kryptodogg/synesthesia)
+- **Roadmap:** See GitHub Issues with `lane:ml` label

manifest.json

@@ -0,0 +1,76 @@
+{
+  "version": "0.1.0",
+  "runtime": "onnxruntime-directml",
+  "models": {
+    "perfrnn": {
+      "path": "perfrnn/model.onnx",
+      "task": "midi-generation",
+      "source": "magenta/performance-rnn",
+      "status": "pending",
+      "sha256": null
+    },
+    "musicvae_encoder": {
+      "path": "musicvae/encoder.onnx",
+      "task": "music-encoding",
+      "source": "magenta/music-vae",
+      "status": "pending",
+      "sha256": null
+    },
+    "musicvae_decoder": {
+      "path": "musicvae/decoder.onnx",
+      "task": "music-decoding",
+      "source": "magenta/music-vae",
+      "status": "pending",
+      "sha256": null
+    },
+    "ddsp_encoder": {
+      "path": "ddsp/encoder.onnx",
+      "task": "audio-analysis",
+      "source": "magenta/ddsp",
+      "status": "pending",
+      "sha256": null
+    },
+    "ddsp_decoder": {
+      "path": "ddsp/decoder.onnx",
+      "task": "audio-synthesis",
+      "source": "magenta/ddsp",
+      "status": "pending",
+      "sha256": null
+    },
+    "spectrostream_encoder": {
+      "path": "spectrostream/encoder.onnx",
+      "task": "audio-tokenization",
+      "source": "magenta-rt/spectrostream",
+      "status": "pending",
+      "sha256": null
+    },
+    "spectrostream_decoder": {
+      "path": "spectrostream/decoder.onnx",
+      "task": "audio-detokenization",
+      "source": "magenta-rt/spectrostream",
+      "status": "pending",
+      "sha256": null
+    },
+    "musiccoca_text": {
+      "path": "musiccoca/text.onnx",
+      "task": "text-embedding",
+      "source": "google/musiccoca",
+      "status": "pending",
+      "sha256": null
+    },
+    "musiccoca_audio": {
+      "path": "musiccoca/audio.onnx",
+      "task": "audio-embedding",
+      "source": "google/musiccoca",
+      "status": "pending",
+      "sha256": null
+    },
+    "gemma3n": {
+      "path": "gemma3n/model.onnx",
+      "task": "vision-mood",
+      "source": "google/gemma-3n",
+      "status": "pending",
+      "sha256": null
+    }
+  }
+}