Update README.md

README.md

@@ -6,4 +6,34 @@ base_model: pyannote/segmentation-3.0
 
 https://huggingface.co/pyannote/segmentation-3.0 with ONNX weights to be compatible with Transformers.js.
 
+
+## Torch → ONNX conversion code:
+```py
+# pip install torch onnx https://github.com/pyannote/pyannote-audio/archive/refs/heads/develop.zip
+import torch
+from pyannote.audio import Model
+
+model = Model.from_pretrained(
+  "pyannote/segmentation-3.0", 
+  use_auth_token="hf_...", # <-- Set your HF token here
+).eval()
+
+dummy_input = torch.zeros(2, 1, 160000)
+torch.onnx.export(
+    model,
+    dummy_input,
+    'model.onnx',
+    do_constant_folding=True,
+    input_names=["input_features"],
+    output_names=["logits"],
+    dynamic_axes={
+        "input_features": {0: "batch_size", 1: "num_channels", 2: "num_samples"},
+        "logits": {0: "batch_size", 1: "num_frames"},
+    },
+)
+```
+
+---
+
+
 Note: Having a separate repo for ONNX weights is intended to be a temporary solution until WebML gains more traction. If you would like to make your models web-ready, we recommend converting to ONNX using [🤗 Optimum](https://huggingface.co/docs/optimum/index) and structuring your repo like this one (with ONNX weights located in a subfolder named `onnx`).