Vibe coded digital AI assistant with voice creation and text to speech response via Qwen3 and locally installed large language models. Multiple personas with persistent contextual memory and individually configurable personalities. Text to speech can be disabled for fluid response and conversation.