update Redme based on Huggingface TEmplate

README.md

@@ -6,18 +6,22 @@ tags:
 - music
 - art
 ---
+# Model Card for Model ID
+## Model Details
+### Model Description
+The model consists of a music encoder ```MERT-v1-300M```, a natural language decoder ```vicuna-7b-delta-v0```, and a linear projection laer between the two.
 
-# MusiLingo-short-v1
-This repo contains the code for the following paper. 
-__[MusiLingo: Bridging Music and Text with Pre-trained Language Models for Music Captioning and Query Response](https://arxiv.org/abs/2309.08730)__
+This checkpoint of MusiLingo is developed on the MusicInstruct (MI)-short and can answer short instructions with music raw audio, such as querying about the tempo, emotion, genre, tags information. You can use the [MI](https://huggingface.co/datasets/m-a-p/Music-Instruct) dataset for the following demo 
 
-You can refer to more information at the [GitHub repo](https://github.com/zihaod/MusiLingo)
 
-You can use the [MusicInstruct (MI)](https://huggingface.co/datasets/m-a-p/Music-Instruct) dataset for the following demo 
+### Model Sources [optional]
+- **Repository:** [GitHub repo](https://github.com/zihaod/MusiLingo)
+- **Paper [optional]:** __[MusiLingo: Bridging Music and Text with Pre-trained Language Models for Music Captioning and Query Response](https://arxiv.org/abs/2309.08730)__
+<!-- - **Demo [optional]:** [More Information Needed] -->
 
-This checkpoint is developped on the MI-short.
 
-# Inference Code
+
+## Getting Start
 ```
 from tqdm.auto import tqdm