docs: update readme

README.md

@@ -8,3 +8,30 @@ pinned: true
 ---
 
 Check out the configuration reference at https://huggingface.co/docs/hub/spaces-config-reference
+
+# Music Source Splitter 🎶
+<a href="https://huggingface.co/spaces/fabiogra/st-music-splitter"><img src="https://img.shields.io/badge/🤗%20Hugging%20Face-Spaces-blue" alt="Hugging Face Spaces"></a>
+
+
+This is a streamlit demo of the [Music Source Separation](https://huggingface.co/spaces/fabiogra/st-music-splitter).
+
+The model can separate the vocals, drums, bass, and other from a music track.
+
+## Usage
+
+You can use the demo [here](https://huggingface.co/spaces/fabiogra/st-music-splitter), or run it locally with:
+
+```bash
+streamlit run app.py
+```
+> **Note**: In order to run the demo locally, you need to install the dependencies with `pip install -r requirements.txt`.
+
+
+## How it works
+
+The app uses a pretrained model called Hybrid Spectrogram and Waveform Source Separation from <a href="https://github.com/facebookresearch/demucs">facebook/htdemucs</a>.
+
+
+## Acknowledgements
+ - HtDemucs model from  <a href="https://github.com/facebookresearch/demucs">facebook/htdemucs</a>
+ - Streamlit Audio Recorder from <a href="https://github.com/stefanrmmr/streamlit_audio_recorder">stefanrmmr/streamlit_audio_recorder</a>