Standalone Speech-to-Intent System with High Accuracy Voice Recognition

I'm looking for a standalone speech-to-intent system that can interact with a Linux system running Ubuntu 16.04. This system should not be dependent on the internet, and I have a c++ application already running within the Linux system. This application would need to be made to run within the arduino or other. Would like to eventually move to a chipset if able. Key Requirements: - Voice Recognition: The system should have high accuracy voice recognition capabilities. - Intent Extraction: It should be able to extract specific intents from the recognized speech. - Standalone Operation: The system should function independently without requiring an internet connection. Specific Intents: - The system should be able to recognize and act upon commands related to controlling mouse functions, such as cursor coordinates and clicks. Ideal Candidate: - Experience in developing speech-to-intent systems. - Proficiency in programming microcontrollers or Arduino for standalone operation. - Strong background in voice recognition technology, particularly achieving high accuracy levels. - Familiarity with Linux systems, particularly Ubuntu 16.04. - Previous work involving interaction between microcontrollers and Linux systems would be advantageous. See additional information here. [login to view URL]