Phase 4 STT pipeline implemented — Silero VAD + faster-whisper — still not working well at all

stt/requirements.txt

@@ -0,0 +1,25 @@
+# STT Container Requirements
+
+# Core dependencies
+fastapi==0.115.6
+uvicorn[standard]==0.32.1
+websockets==14.1
+aiohttp==3.11.11
+
+# Audio processing
+numpy==2.2.2
+soundfile==0.12.1
+librosa==0.10.2.post1
+
+# VAD (CPU)
+torch==2.9.1  # Latest PyTorch
+torchaudio==2.9.1
+silero-vad==5.1.2
+
+# STT (GPU)
+faster-whisper==1.2.1  # Latest version (Oct 31, 2025)
+ctranslate2==4.5.0  # Required by faster-whisper
+
+# Utilities
+python-multipart==0.0.20
+pydantic==2.10.4