I can use text to get audio response and barge in

requirements.txt

@@ -0,0 +1,37 @@
+# Web Framework
+fastapi>=0.109.0
+uvicorn[standard]>=0.27.0
+websockets>=12.0
+python-multipart>=0.0.6
+
+# WebRTC (optional - for WebRTC transport)
+aiortc>=1.6.0
+
+# Audio Processing
+av>=12.1.0
+numpy>=1.26.3
+onnxruntime>=1.16.3
+
+# Configuration
+pydantic>=2.5.3
+pydantic-settings>=2.1.0
+python-dotenv>=1.0.0
+toml>=0.10.2
+
+# Logging
+loguru>=0.7.2
+
+# HTTP Client
+aiohttp>=3.9.1
+
+# AI Services - LLM
+openai>=1.0.0
+
+# AI Services - TTS
+edge-tts>=6.1.0
+pydub>=0.25.0  # For audio format conversion
+
+# Microphone client dependencies
+sounddevice>=0.4.6
+soundfile>=0.12.1
+pyaudio>=0.2.13  # More reliable audio on Windows