Add bounded CPU VAD diagnostic with real model contract tests

tools/speech-vad/test_native.py

@@ -0,0 +1,53 @@
+"""Actual CPU model contract checks; explicit asset paths required."""
+import hashlib
+import json
+import os
+from pathlib import Path
+import struct
+import subprocess
+import tempfile
+import unittest
+
+
+class NativeContract(unittest.TestCase):
+    @classmethod
+    def setUpClass(cls):
+        cls.worker = Path(os.environ["SEARCHI_VAD_DIAGNOSTIC"])
+        cls.model = Path(os.environ["SEARCHI_VAD_MODEL"])
+        if hashlib.sha256(cls.model.read_bytes()).hexdigest() != "2aa269b785eeb53a82983a20501ddf7c1d9c48e33ab63a41391ac6c9f7fb6987":
+            raise ValueError("Expected frozen Silero v6.2.0 export")
+
+    def run_pcm(self, data):
+        with tempfile.TemporaryDirectory(prefix="searchi-vad-contract-") as scratch:
+            pcm = Path(scratch) / "samples.f32"
+            pcm.write_bytes(data)
+            return subprocess.run([str(self.worker), str(self.model), str(pcm)],
+                                  capture_output=True, timeout=30, check=False)
+
+    def test_real_model_silence_tail(self):
+        result = self.run_pcm(b"\0" * (16001 * 4))
+        self.assertEqual(result.returncode, 0, result.stderr.decode("utf-8", errors="replace"))
+        report = json.loads(result.stdout)
+        self.assertEqual(report["samples"], 16001)
+        self.assertEqual(len(report["probabilities"]), 32)
+        self.assertEqual(report["intervals_cs"], [])
+        self.assertEqual(report["threads"], 1)
+        self.assertIs(report["gpu"], False)
+        for key in ("load_ms", "detect_ms", "segment_ms"):
+            self.assertGreaterEqual(report[key], 0)
+
+    def test_invalid_pcm(self):
+        for data in (b"", b"x", struct.pack("<f", float("nan")), struct.pack("<f", 2)):
+            with self.subTest(data=data):
+                result = self.run_pcm(data)
+                self.assertEqual(result.returncode, 2)
+                self.assertEqual(result.stdout, b"")
+
+    def test_unknown_arguments_fail(self):
+        result = subprocess.run([str(self.worker), "--unknown"], capture_output=True, timeout=5)
+        self.assertEqual(result.returncode, 2)
+        self.assertEqual(result.stdout, b"")
+
+
+if __name__ == "__main__":
+    unittest.main()