feat: move non-secret settings from .env to config.yaml

voice-to-mermaid/backend/config.yaml

@@ -1,11 +1,28 @@
 # voice-to-mermaid backend — non-sensitive app config
-# Volume-mounted in Docker; edit without rebuilding.
+# Volume-mounted in Docker; edit here without rebuilding.
 #
-# Secrets (OLLAMA_URL, API_KEY, etc.) stay in .env / Infisical.
+# Secrets (OLLAMA_URL, OPENAI_API_KEY, etc.) stay in .env / Infisical.
 
 ollama:
+  default_model: "qwen3:8b"
   # Models shown in the UI picker — exact names or prefixes (e.g. "qwen2.5:").
   # Leave empty to show all installed models.
   model_filter:
     - "qwen2.5:14b"
     - "qwen3.5:27b"
+    - "qwen3:8b"
+    - "qwen3:14b"
+    - "llama3.2:3b"
+
+openai:
+  model: "gpt-4o-mini"
+
+whisper:
+  enabled: false   # set true only when requirements.stt.txt deps are installed
+  model: "medium.en"
+  language: "en"
+  device: "auto"   # "auto", "cuda", "cpu" — maps to WhisperLiveKitConfig backend
+
+paths:
+  prompt: "prompts/mermaid.txt"
+  log_dir: "data/logs"

voice-to-mermaid/backend/docker-compose.yml

@@ -2,6 +2,7 @@ services:
   ollama:
     image: ollama/ollama:latest
     container_name: vtm-ollama
+    restart: unless-stopped
     ports:
       - "11434:11434"
     volumes:
@@ -18,6 +19,7 @@ services:
   api:
     build: .
     container_name: vtm-api
+    restart: unless-stopped
     ports:
       - "${PORT:-7625}:7625"
     env_file: .env

voice-to-mermaid/backend/main.py

@@ -47,39 +47,47 @@
 logging.basicConfig(level=logging.INFO, format="%(asctime)s %(levelname)s %(message)s")
 log = logging.getLogger(__name__)
 
-# ── Config from env ───────────────────────────────────────────────────────────
-
-OLLAMA_URL = os.environ.get("OLLAMA_URL", "http://localhost:11434").rstrip("/")
-OLLAMA_MODEL = os.environ.get("OLLAMA_MODEL", "qwen3:8b")
-OPENAI_URL = os.environ.get("OPENAI_BASE_URL", "").rstrip("/")
-OPENAI_KEY = os.environ.get("OPENAI_API_KEY", "")
-OPENAI_MODEL = os.environ.get("OPENAI_MODEL", "gpt-4o-mini")
-WHISPER_ENABLED = os.environ.get("WHISPER_ENABLED", "0") == "1"
-WHISPER_MODEL = os.environ.get("WHISPER_MODEL", "medium.en")
-WHISPER_LANG = os.environ.get("WHISPER_LANG", "en")
-PROMPT_PATH = Path(os.environ.get("MERMAID_PROMPT_PATH", "prompts/mermaid.txt"))
-LOG_DIR = Path(os.environ.get("LOG_DIR", "data/logs"))
-
-_ollama_models_cache: list[dict] | None = None
+# ── Config (config.yaml for settings, .env for secrets) ──────────────────────
 
 # config.yaml lives next to this file in source; /app/config.yaml in Docker
 _CONFIG_PATH = Path(__file__).parent / "config.yaml"
 
 
-def _load_model_filter() -> list[str]:
+def _load_config() -> dict:
     try:
         import yaml  # type: ignore[import-untyped]
-
-        raw = yaml.safe_load(_CONFIG_PATH.read_text()) or {}
-        return [str(p) for p in (raw.get("ollama", {}).get("model_filter") or [])]
+        raw = yaml.safe_load(_CONFIG_PATH.read_text())
+        if not isinstance(raw, dict):
+            log.warning("config.yaml is not a mapping — using defaults")
+            return {}
+        return raw
     except FileNotFoundError:
-        return []
+        log.warning("config.yaml not found — using defaults")
+        return {}
     except Exception as exc:
         log.warning("Failed to load config.yaml: %s", exc)
-        return []
+        return {}
 
 
-_MODEL_FILTER: list[str] = _load_model_filter()
+_cfg = _load_config()
+
+# Secrets — env only
+OLLAMA_URL = os.environ.get("OLLAMA_URL", "http://localhost:11434").rstrip("/")
+OPENAI_URL = os.environ.get("OPENAI_BASE_URL", "").rstrip("/")
+OPENAI_KEY = os.environ.get("OPENAI_API_KEY", "")
+
+# Settings — config.yaml only; edit config.yaml to change these
+OLLAMA_MODEL = _cfg.get("ollama", {}).get("default_model", "qwen3:8b")
+OPENAI_MODEL = _cfg.get("openai", {}).get("model", "gpt-4o-mini")
+WHISPER_ENABLED = _cfg.get("whisper", {}).get("enabled", False)
+WHISPER_MODEL = _cfg.get("whisper", {}).get("model", "medium.en")
+WHISPER_LANG = _cfg.get("whisper", {}).get("language", "en")
+WHISPER_DEVICE = _cfg.get("whisper", {}).get("device", "auto")
+PROMPT_PATH = Path(_cfg.get("paths", {}).get("prompt", "prompts/mermaid.txt"))
+LOG_DIR = Path(_cfg.get("paths", {}).get("log_dir", "data/logs"))
+_MODEL_FILTER: list[str] = [str(p) for p in (_cfg.get("ollama", {}).get("model_filter") or [])]
+
+_ollama_models_cache: list[dict] | None = None
 
 
 async def _fetch_ollama_models() -> list[dict]:
@@ -248,15 +256,20 @@ async def _lifespan(app: FastAPI):
     global _transcription_engine
     _init_log_file()
     if WHISPER_ENABLED:
-        from whisperlivekit import TranscriptionEngine
-        from whisperlivekit.config import WhisperLiveKitConfig
-
-        log.info("Loading WhisperLiveKit  model=%s  lang=%s", WHISPER_MODEL, WHISPER_LANG)
+        try:
+            from whisperlivekit import TranscriptionEngine
+            from whisperlivekit.config import WhisperLiveKitConfig
+        except ImportError:
+            log.warning("whisper.enabled=true but whisperlivekit not installed — STT disabled. Install requirements.stt.txt to enable.")
+            yield
+            return
+
+        log.info("Loading WhisperLiveKit  model=%s  lang=%s  device=%s", WHISPER_MODEL, WHISPER_LANG, WHISPER_DEVICE)
         _transcription_engine = TranscriptionEngine(
             config=WhisperLiveKitConfig(
                 model_size=WHISPER_MODEL,
                 lan=WHISPER_LANG,
-                backend="auto",
+                backend=WHISPER_DEVICE,
                 backend_policy="simulstreaming",
                 vad=True,
                 vac=False,
@@ -268,7 +281,7 @@ async def _lifespan(app: FastAPI):
         )
         log.info("WhisperLiveKit ready")
     else:
-        log.info("STT disabled (WHISPER_ENABLED=0) — audio transcription unavailable")
+        log.info("STT disabled (whisper.enabled=false) — audio transcription unavailable")
     yield
     log.info("Server shutting down")