Augmentative and alternative communication (AAC) software assists impaired speakers to communicate. Enabling context awareness, authenticity and ease of use goes a long way in empowering communication. Our work presents an easy to use AAC tool that allows for message composition with text or emoji input by typing as well as speaking with contextually relevant word-level suggestions. Any transcription errors are corrected, and contextually relevant phrases with the appropriately chosen emotional tone are suggested by large-language models. The user is then able to use modern text-to-speech synthesis to be able to synthesize the composed message in the user’s own voice. Our system additionally maintains features from AAC software such as word-cards, pre-composed and frequently used messages.