Add pipeline tag, library name, paper link, and usage section

README.md

@@ -1,11 +1,13 @@
 ---
+datasets:
+- as2m
+license: cc-by-4.0
+pipeline_tag: audio-classification
+library_name: espnet
 tags:
 - espnet
 - audio
 - classification
-datasets:
-- as2m
-license: cc-by-4.0
 ---
 
 ## ESPnet2 CLS model
@@ -14,6 +16,36 @@ license: cc-by-4.0
 
 This model was trained by Shikhar Bharadwaj using as2m recipe in [espnet](https://github.com/espnet/espnet/).
 
+It is presented in the paper [OpenBEATs: A Fully Open-Source General-Purpose Audio Encoder](https://huggingface.co/papers/2507.14129).
+
+* **GitHub Repository:** [Shikhar-S/OpenBEATs](https://github.com/Shikhar-S/OpenBEATs)
+
+### Usage
+
+You can run this model using the `openbeats` library:
+
+```bash
+pip install openbeats
+```
+
+```python
+from openbeats.model import OpenBeats
+from openbeats.utils import load_audio
+
+# load model
+model = OpenBeats.from_pretrained("espnet/OpenBEATS-Base-i3-as2m", device="cuda")
+
+# from a file with any sample rate
+out = model.encode_file("audio.wav")              # pass chunk_seconds=10 for long audio
+
+# or load the waveform in 16khz monoaural array with values in [-1,1]
+wav, sr = load_audio("audio.wav")
+# and pass it
+out = model.encode(wav, sr)
+
+print(out["patch_embeddings"].shape)
+```
+
 ## CLS config
 
 <details><summary>expand</summary>
@@ -780,4 +812,4 @@ task: cls
 
 
 
-```
+```