Compress logging audio format

src/core/usage_logger.py

@@ -299,16 +299,29 @@ def _compute_audio_id(audio: np.ndarray, ts: datetime) -> str:
     return f"{audio_hash}:{ts.strftime('%Y%m%dT%H%M%S')}"
 
 
-def _encode_audio_flac(audio: np.ndarray, sample_rate: int, audio_id: str) -> str:
-    """Encode audio to a temp FLAC file; returns the file path."""
+def _encode_audio_ogg(audio: np.ndarray, sample_rate: int, audio_id: str) -> str:
+    """Encode audio to a temp OGG Opus file (32kbps); returns the file path."""
     import soundfile as sf
+    import subprocess
 
     tmp_dir = LOG_DIR / "tmp_audio"
     tmp_dir.mkdir(parents=True, exist_ok=True)
     safe_id = audio_id.replace(":", "-")
-    filepath = tmp_dir / f"{safe_id}.flac"
-    sf.write(str(filepath), audio, sample_rate, format="FLAC")
-    return str(filepath)
+
+    wav_path = tmp_dir / f"{safe_id}.wav"
+    ogg_path = tmp_dir / f"{safe_id}.ogg"
+    sf.write(str(wav_path), audio, sample_rate, format="WAV")
+    try:
+        subprocess.run(
+            ["ffmpeg", "-y", "-i", str(wav_path),
+             "-c:a", "libopus", "-b:a", "32k",
+             "-ar", "16000", "-ac", "1",
+             str(ogg_path)],
+            capture_output=True, check=True,
+        )
+    finally:
+        wav_path.unlink(missing_ok=True)
+    return str(ogg_path)
 
 
 def _sync_row_to_scheduler(row: Dict[str, Any]) -> None:
@@ -461,7 +474,7 @@ def log_alignment(
         def _encode_and_append():
             """Encode FLAC and register row with scheduler/fallback."""
             try:
-                row["audio"] = _encode_audio_flac(audio, sample_rate, audio_id)
+                row["audio"] = _encode_audio_ogg(audio, sample_rate, audio_id)
             except Exception as e:
                 print(f"[USAGE_LOG] FLAC encoding failed: {e}")
             if _aligner_scheduler is None: