Add audio training roadmap to model card

README.md

@@ -289,9 +289,26 @@ fused_embedding = model.encode_multimodal(
 
 - Optimized for English; multilingual performance may vary
 - Image resolution fixed at 512x512
-- Audio modality available but not trained in this release
+- Audio encoder included but not yet trained (see Roadmap)
 - Best for semantic similarity, not generative tasks
 
+## Roadmap
+
+### Audio Modality Training (Planned)
+
+The model architecture includes a Whisper audio encoder, but this release only trained on image-text data. Future releases will add audio-text alignment using:
+
+| Dataset | Size | Description |
+|---------|------|-------------|
+| [WavCaps](https://huggingface.co/datasets/cvssp/WavCaps) | 400K | Largest audio-caption dataset |
+| [AudioCaps](https://github.com/cdjkim/audiocaps) | 46K | YouTube audio with human captions |
+| [Clotho](https://zenodo.org/record/3490684) | 6K | High-quality multi-annotator captions |
+
+This will enable:
+- Audio-to-text retrieval
+- Text-to-audio retrieval  
+- Audio-image-text multimodal fusion
+
 ## Citation
 
 ```bibtex