LLM is large language model, the brain of your voice AI, STT is speech to text, the transcriber for turning what the human says into text, and TTS is text to speech, turning whatever output your LLM gives back into spoken words. Here's an article with more info: https://docs.vapi.ai/quickstart