A voice-enabled AI assistant powered by Google's Gemma 3n-e4b-it large language model. This project demonstrates how to create an interactive, on-device AI agent capable of understanding natural language, generating intelligent productivity plans, and interacting with users via both text and speech.
The primary purpose is to provide a personal productivity helper that offers smart task breakdowns, planning strategies, and helpful advice, showcasing the power of local LLM inference combined with multimodal input/output.
🌟 Features
Intelligent Task Planning: Utilizes the Gemma 3n-e4b-it model to intelligently break down complex tasks, suggest actionable steps, and provide structured plans.
Productivity Strategy Advice: Offers personalized suggestions using well-known productivity techniques like the Pomodoro Technique or time blocking.
Voice Input (Speech-to-Text): Seamlessly take commands and task descriptions through your microphone using Google Speech Recognition.
Spoken Responses (Text-to-Speech): The assistant speaks its responses aloud, providing a natural and engaging conversational experience.
Text Interaction: Supports traditional text input and displays responses directly in the console.
Minimalist Design: Focuses on core AI agent functionality without unnecessary complexities, ideal for demonstrating on-device LLM capabilities.
Basic Reminder Guidance: While it doesn't set system reminders, it intelligently acknowledges time-based requests and advises users on how to use external tools (like phone alarms or calendar apps) for actual reminders.
A voice-enabled AI assistant powered by Google's Gemma 3n-e4b-it large language model. This project demonstrates how to create an interactive, on-device AI agent capable of understanding natural language, generating intelligent productivity plans, and interacting with users via both text and speech.
The primary purpose is to provide a personal productivity helper that offers smart task breakdowns, planning strategies, and helpful advice, showcasing the power of local LLM inference combined with multimodal input/output.
🌟 Features
Intelligent Task Planning: Utilizes the Gemma 3n-e4b-it model to intelligently break down complex tasks, suggest actionable steps, and provide structured plans.
Productivity Strategy Advice: Offers personalized suggestions using well-known productivity techniques like the Pomodoro Technique or time blocking.
Voice Input (Speech-to-Text): Seamlessly take commands and task descriptions through your microphone using Google Speech Recognition.
Spoken Responses (Text-to-Speech): The assistant speaks its responses aloud, providing a natural and engaging conversational experience.
Text Interaction: Supports traditional text input and displays responses directly in the console.
Minimalist Design: Focuses on core AI agent functionality without unnecessary complexities, ideal for demonstrating on-device LLM capabilities.
Basic Reminder Guidance: While it doesn't set system reminders, it intelligently acknowledges time-based requests and advises users on how to use external tools (like phone alarms or calendar apps) for actual reminders.