[本地唤醒模型]：完成KWS模型下载和检查，包含manifest、配置和model-check

scripts/download_speech_models.py

@@ -16,6 +16,9 @@ if str(SRC_DIR) not in sys.path:
     sys.path.insert(0, str(SRC_DIR))
 
 from owner_voice_pet.speech_models import (
+    DEFAULT_KWS_DIR,
+    DEFAULT_KWS_KEYWORDS,
+    DEFAULT_KWS_URL,
     DEFAULT_STT_DIR,
     DEFAULT_STT_URL,
     DEFAULT_VAD_URL,
@@ -32,6 +35,7 @@ def main(argv: list[str] | None = None) -> int:
 
     target = Path(args.dir)
     target.mkdir(parents=True, exist_ok=True)
+    download_kws(target, force=args.force)
     download_vad(target, force=args.force)
     download_stt(target, force=args.force)
     manifest = write_default_manifest(target)
@@ -42,6 +46,38 @@ def main(argv: list[str] | None = None) -> int:
     return 0 if not status.missing_files else 1
 
 
+def download_kws(models_dir: Path, *, force: bool = False) -> Path:
+    output_dir = models_dir / "wake" / DEFAULT_KWS_DIR
+    if output_dir.exists() and any(output_dir.iterdir()) and not force:
+        print(f"skip existing {output_dir}")
+    else:
+        output_dir.parent.mkdir(parents=True, exist_ok=True)
+        last_error: Exception | None = None
+        for attempt in range(1, 4):
+            with tempfile.TemporaryDirectory() as tmp:
+                archive = Path(tmp) / f"{DEFAULT_KWS_DIR}.tar.bz2"
+                try:
+                    download_file(DEFAULT_KWS_URL, archive)
+                    extract_tar_bz2(archive, output_dir.parent)
+                    last_error = None
+                    break
+                except Exception as exc:
+                    last_error = exc
+                    print(f"download/extract KWS attempt {attempt} failed: {exc}")
+                    if output_dir.exists():
+                        shutil.rmtree(output_dir)
+        if last_error is not None:
+            raise last_error
+    keywords = models_dir / "wake" / "keywords.txt"
+    if force or not keywords.exists():
+        keywords.parent.mkdir(parents=True, exist_ok=True)
+        keywords.write_text(DEFAULT_KWS_KEYWORDS, encoding="utf-8")
+        print(f"wrote {keywords}")
+    else:
+        print(f"skip existing {keywords}")
+    return output_dir
+
+
 def download_vad(models_dir: Path, *, force: bool = False) -> Path:
     output = models_dir / "vad" / "silero_vad.onnx"
     if output.exists() and not force: