Monika is an advanced AI assistant designed to facilitate natural and seamless interaction through a combination of cutting-edge technologies. By integrating speech-to-text (STT), natural language processing (NLP), and text-to-speech (TTS), Monika offers a comprehensive conversational experience that caters to user needs.

Key Features

• Speech-to-Text (STT): Effortlessly converts spoken audio into text using OpenAI's Whisper, enabling accurate transcription for real-time interaction.
• Natural Language Processing (NLP): Leverages Google Gemini to refine user input processing, ensuring that responses are relevant and contextually appropriate.
• Text-to-Speech (TTS): Delivers natural-sounding speech synthesized by RealtimeTTS, which enhances the conversational experience by providing clear and articulate responses.
• Emotional Expression: Utilizes Orpheus technology to convey emotions during interactions, adding a human touch to conversations.
• Voice Activity Detection (VAD): Automatically detects when users are speaking, facilitating smoother dialogues without unnecessary delays.
• Interactive Web Interface: A user-friendly interface allows for intuitive interaction with Monika, making it accessible to everyone.

Video Demonstration

Experience Monika in action by watching the demo available on YouTube: Watch Monika in Action

How It Works

Monika operates by starting a Flask server that hosts the web interface. Users can interact by speaking into their microphone, prompting Monika to transcribe their speech, process it, and respond accordingly. The system’s endpoints provide organized functionalities, managing various tasks such as audio transcription and voice response streaming:

• Main Web Interface: Accessed via the root endpoint (/).
• Transcription Endpoint: Handles audio input for transcription purposes (/transcribe).
• Processing Endpoint: Utilizes Gemini to process text (/gemini_process).
• Speech Synthesis Endpoint: Streams synthesized speech through the TTS system (/tts).

Future Enhancements

Planned improvements aim to enhance user experience, including reducing TTS latency, enabling interruption handling while Monika speaks, expanding language support, providing custom voice options, and developing offline functionality for basic operations without internet connectivity.

Monika stands as a powerful tool for anyone seeking a reliable and expressive AI assistant, suitable for various applications ranging from personal use to business solutions.