Simplify client modes and add VAD retention policy

bridge/.env.example

@@ -26,7 +26,8 @@ WAV_KEEP_FILES=10
 
 VAD_ENABLED=true
 VAD_DIR=/data/vad
-VAD_KEEP_FILES=100
+VAD_KEEP_FILES=200
+VAD_MAX_AGE_DAYS=7
 VAD_PREROLL_MS=1000
 VAD_POSTROLL_MS=1000
 VAD_START_THRESHOLD=900

bridge/Dockerfile

@@ -30,7 +30,8 @@ ENV WS_HOST=0.0.0.0 \
     WAV_KEEP_FILES=10 \
     VAD_ENABLED=true \
     VAD_DIR=/data/vad \
-    VAD_KEEP_FILES=100 \
+    VAD_KEEP_FILES=200 \
+    VAD_MAX_AGE_DAYS=7 \
     VAD_PREROLL_MS=1000 \
     VAD_POSTROLL_MS=1000 \
     VAD_START_THRESHOLD=900 \

bridge/README.md

@@ -96,7 +96,8 @@ You can build automations on these events (for STT/TTS pipelines or Node-RED han
   - `VAD_POSTROLL_MS=1000` keeps 1s after speech end
   - `VAD_START_THRESHOLD` / `VAD_STOP_THRESHOLD` tune sensitivity
   - `VAD_DIR` stores per-utterance WAV files
-  - `VAD_KEEP_FILES` limits stored VAD WAV files
+  - `VAD_KEEP_FILES=200` limits number of stored VAD WAV files
+  - `VAD_MAX_AGE_DAYS=7` deletes VAD WAV files older than 7 days
 - MQTT is recommended for control/events, WebSocket for streaming audio
 
 ## 7) Build and push to Gitea registry
@@ -148,7 +149,8 @@ services:
       WAV_KEEP_FILES: "10"
       VAD_ENABLED: "true"
       VAD_DIR: "/data/vad"
-      VAD_KEEP_FILES: "100"
+      VAD_KEEP_FILES: "200"
+      VAD_MAX_AGE_DAYS: "7"
       VAD_PREROLL_MS: "1000"
       VAD_POSTROLL_MS: "1000"
       VAD_START_THRESHOLD: "900"

bridge/app.py

@@ -53,7 +53,8 @@ WAV_HEADER_BYTES = 44
 
 VAD_ENABLED = getenv_bool("VAD_ENABLED", True)
 VAD_DIR = Path(os.getenv("VAD_DIR", "/data/vad"))
-VAD_KEEP_FILES = int(os.getenv("VAD_KEEP_FILES", "100"))
+VAD_KEEP_FILES = int(os.getenv("VAD_KEEP_FILES", "200"))
+VAD_MAX_AGE_DAYS = int(os.getenv("VAD_MAX_AGE_DAYS", "7"))
 VAD_PREROLL_MS = int(os.getenv("VAD_PREROLL_MS", "1000"))
 VAD_POSTROLL_MS = int(os.getenv("VAD_POSTROLL_MS", "1000"))
 VAD_START_THRESHOLD = int(os.getenv("VAD_START_THRESHOLD", "900"))
@@ -136,17 +137,34 @@ async def call_ha_webhook(event: str, payload: dict) -> None:
         log.exception("ha webhook call failed")
 
 
-def enforce_wav_retention(directory: Path, keep_files: int) -> None:
-    if keep_files <= 0:
+def enforce_wav_retention(directory: Path, keep_files: int, max_age_days: int = 0) -> None:
+    if keep_files <= 0 and max_age_days <= 0:
         return
     try:
         directory.mkdir(parents=True, exist_ok=True)
-        wavs = sorted(
-            [p for p in directory.glob("*.wav") if p.is_file()],
-            key=lambda p: p.stat().st_mtime,
-        )
-        while len(wavs) > keep_files:
-            oldest = wavs.pop(0)
+        wavs = []
+        now = time.time()
+        max_age_seconds = max_age_days * 86400
+        for p in directory.glob("*.wav"):
+            if not p.is_file():
+                continue
+            try:
+                st = p.stat()
+            except Exception:
+                continue
+            if max_age_seconds > 0 and (now - st.st_mtime) > max_age_seconds:
+                try:
+                    p.unlink()
+                    log.info("deleted old wav by age: %s", p)
+                except Exception:
+                    log.exception("failed to delete old wav by age: %s", p)
+                continue
+            wavs.append((p, st.st_mtime))
+
+        wavs.sort(key=lambda x: x[1])
+        files = [p for (p, _) in wavs]
+        while keep_files > 0 and len(files) > keep_files:
+            oldest = files.pop(0)
             try:
                 oldest.unlink()
                 log.info("deleted old wav: %s", oldest)
@@ -198,7 +216,7 @@ def write_vad_wav_segment(session: DeviceSession, pcm: bytes) -> Optional[str]:
         wf.writeframes(pcm)
         wf.close()
         session.vad_segment_index += 1
-        enforce_wav_retention(VAD_DIR, VAD_KEEP_FILES)
+        enforce_wav_retention(VAD_DIR, VAD_KEEP_FILES, VAD_MAX_AGE_DAYS)
         return str(path)
     except Exception:
         log.exception("failed to write vad wav segment for %s", session.device_id)