Tune the server engine's Silero VAD so batched file transcription stops dropping quiet speech

Co-Authored-By: Claude Opus 5.5 <[email protected]>

server-py/sidecar.py

@@ -27,11 +27,12 @@ interrupt the segment loop of the request in flight.
 
 Decoding strategy by mode (Decisions.md D-WS5a):
   "file"  faster-whisper's BatchedInferencePipeline with its Silero VAD
-          (vad_filter=True): silence is dropped, speech is cut into <= 30 s
-          chunks and BATCH_SIZE chunks decode per forward pass. Segments come
-          out in bursts (one burst per batch), so a cancel takes effect at the
-          next burst boundary; the Bun side's heartbeat (D-021) covers the
-          longer quiet stretch before the first burst.
+          (vad_filter=True, VAD_PARAMETERS): silence is dropped, speech is cut
+          into <= 30 s chunks and BATCH_SIZE chunks decode per forward pass.
+          Without word-level timestamps each chunk is one segment (<= 30 s).
+          Segments come out in bursts (one burst per batch), so a cancel
+          takes effect at the next burst boundary; the Bun side's heartbeat
+          (D-021) covers the longer quiet stretch before the first burst.
   "live"  the plain sequential WhisperModel.transcribe: server-live passes
           are short (<= 29 s, often a few seconds), so batching has nothing to
           batch and VAD only adds latency and can clip a short utterance.
@@ -60,6 +61,12 @@ import threading
 
 SAMPLE_RATE = 16000
 BATCH_SIZE = 8  # D-WS5a: VAD chunks decoded per batched forward pass
+# D-WS5a: Silero VAD tuned for file mode. faster-whisper's batched default
+# (threshold 0.5, 400 ms padding) dropped quiet and overlapped speech —
+# deletions nearly doubled on en_alien; 0.2 / 800 ms brought WER back to the
+# sequential decoder's level. min_silence_duration_ms stays at VadOptions'
+# 2000 ms default (a dict replaces the pipeline's 160 ms preset).
+VAD_PARAMETERS = {"threshold": 0.2, "speech_pad_ms": 800}
 
 
 def emit(message):
@@ -156,6 +163,7 @@ def main():
                     language=language,
                     task=task,
                     vad_filter=True,
+                    vad_parameters=dict(VAD_PARAMETERS),  # fresh: faster-whisper may mutate it
                     batch_size=BATCH_SIZE,
                 )
 

tests/transcribe.test.ts

@@ -519,6 +519,8 @@ describe("server-py/sidecar.py (stubbed faster-whisper, D-WS5a)", () => {
     expect(explicit.texts[0]).toContain("batched model=m1");
     expect(explicit.texts[0]).toContain("vad_filter=True");
     expect(explicit.texts[0]).toContain("batch_size=8");
+    // Tuned Silero VAD: the default (0.5 / 400 ms) dropped quiet speech (D-WS5a).
+    expect(explicit.texts[0]).toContain("vad_parameters={'threshold': 0.2, 'speech_pad_ms': 800}");
     expect(explicit.texts[0]).toContain("language=da");
     expect(explicit.texts[0]).toContain("task=transcribe");