Initial commit: audio-chat with fixes

- Created AGENTS.md with architecture documentation
- Fixed race conditions and async patterns
- Added conversation history to LLM prompts
- Fixed TTS audio shape handling
- Added buffer limits and graceful shutdown
- Fixed client.py with file sending support
- Removed duplicate requirements
- Added .gitignore

config.py

@@ -0,0 +1,29 @@
+import os
+from pathlib import Path
+from dotenv import load_dotenv
+
+env_path = Path(__file__).parent / ".env"
+load_dotenv(env_path)
+
+
+class Config:
+    # Models
+    STT_MODEL = os.getenv("STT_MODEL", "Systran/faster-whisper-large-v3")
+    LLM_MODEL = os.getenv("LLM_MODEL", "Qwen/Qwen2.5-7B-Instruct")
+    TTS_MODEL = os.getenv("TTS_MODEL", "facebook/mms-tts-rus")
+
+    # Audio settings
+    SAMPLE_RATE = int(os.getenv("SAMPLE_RATE", "16000"))
+    AUDIO_BUFFER_SECONDS = float(os.getenv("AUDIO_BUFFER_SECONDS", "2"))
+    CHUNK_SIZE = int(os.getenv("CHUNK_SIZE", "1024"))
+
+    # Server
+    HOST = os.getenv("HOST", "0.0.0.0")
+    PORT = int(os.getenv("PORT", "8000"))
+
+    # LLM settings
+    LLM_MAX_TOKENS = int(os.getenv("LLM_MAX_TOKENS", "512"))
+    LLM_TEMPERATURE = float(os.getenv("LLM_TEMPERATURE", "0.7"))
+
+    # GPU
+    DEVICE = os.getenv("DEVICE", "auto")