Complete DDS training workflow and delivery package

backend/scripts/local_stt.py

@@ -0,0 +1,84 @@
+"""Локальный whisper.cpp server для Whisper small (macOS / Windows / Linux).
+
+Сначала скачайте GGML-вес через download_local_models.py. Исполняемый файл
+задаётся WHISPER_SERVER_BIN или ищется в PATH / локальной сборке. Сеть не нужна.
+"""
+
+import os
+import shutil
+import subprocess
+import sys
+import time
+import urllib.error
+import urllib.request
+from pathlib import Path
+
+ROOT = Path(__file__).resolve().parents[1]
+MODEL = ROOT / "models" / "whisper-small" / "ggml-small-q5_1.bin"
+PORT = 18082
+
+
+def binary_path() -> str:
+    built = ROOT / "models" / "bin" / "whisper.cpp-1.9.4" / "build" / "bin" / "whisper-server"
+    binary = (os.environ.get("WHISPER_SERVER_BIN") or shutil.which("whisper-server")
+              or shutil.which("whisper-server.exe")
+              # The in-tree build is macOS arm64, not a portable executable.
+              or (str(built) if sys.platform == "darwin" and built.is_file() else None))
+    if not binary or not Path(binary).is_file():
+        raise RuntimeError("нет whisper-server; установите бинарник whisper.cpp и задайте WHISPER_SERVER_BIN")
+    return binary
+
+
+def command(binary: str, threads: int = 4) -> list[str]:
+    if not MODEL.is_file():
+        raise RuntimeError(f"нет модели: {MODEL} — запустите make local-models")
+    return [binary, "-m", str(MODEL), "--host", "127.0.0.1", "--port", str(PORT),
+            "--language", "ru", "--threads", str(threads), "--no-fallback"]
+
+
+def ready() -> bool:
+    try:
+        with urllib.request.urlopen(f"http://127.0.0.1:{PORT}/", timeout=1) as response:
+            return response.status == 200
+    except (OSError, urllib.error.HTTPError):
+        return False
+
+
+def main() -> int:
+    try:
+        if ready():
+            raise RuntimeError(f"порт {PORT} уже занят сервером")
+        argv = command(binary_path())
+    except RuntimeError as exc:
+        print(exc, file=sys.stderr)
+        return 2
+    flags = subprocess.CREATE_NEW_PROCESS_GROUP if os.name == "nt" else 0
+    process = subprocess.Popen(argv, cwd=ROOT, creationflags=flags, start_new_session=os.name != "nt")
+    try:
+        deadline = time.monotonic() + 120
+        while time.monotonic() < deadline:
+            if process.poll() is not None:
+                raise RuntimeError("whisper-server завершился до готовности")
+            if ready():
+                print(f"Whisper small готов на 127.0.0.1:{PORT}; Ctrl+C остановит сервер", flush=True)
+                while process.poll() is None:
+                    time.sleep(0.5)
+                raise RuntimeError("whisper-server неожиданно завершился")
+            time.sleep(0.5)
+        raise RuntimeError("Whisper small не стал готов за 120 секунд")
+    except KeyboardInterrupt:
+        return 0
+    except RuntimeError as exc:
+        print(exc, file=sys.stderr)
+        return 1
+    finally:
+        if process.poll() is None:
+            process.terminate()
+        try:
+            process.wait(timeout=5)
+        except subprocess.TimeoutExpired:
+            process.kill()
+
+
+if __name__ == "__main__":
+    raise SystemExit(main())