Document direct Hugging Face loading

README.md

@@ -35,6 +35,22 @@ print(model.labels[index])
 This call downloads `perch_v2_torch.pt` from this Hugging Face repository and
 caches it locally.
 
+### With `huggingface_hub`
+
+From the [GitHub repository](https://github.com/janclemenslab/perch2_torch)
+checkout:
+
+```python
+from huggingface_hub import hf_hub_download
+from perch import load_model
+
+checkpoint = hf_hub_download(
+    "DAS-DeepAudioSegmenter/perch2-torch",
+    "perch_v2_torch.pt",
+)
+model = load_model(checkpoint).eval()
+```
+
 Inputs are mono, 32 kHz waveforms with shape `(time,)` or `(batch, time)`.
 Audio longer than five seconds is processed in overlapping windows and pooled.