Trivia Host: An AI-Powered Conversational Game

Step into a new era of trivia with Trivia Host, an interactive web application where you compete against a dynamic AI. Choose your host's personality, answer challenging questions using your voice, and experience a truly personalized trivia game, all powered by the Google Gemini API.

✨ Features

• Dynamic AI Personalities: Choose from a variety of unique hosts, such as a "Sarcastic Robot" or a "Mysterious Wizard," each with their own style of asking questions and giving feedback.
• Voice-Activated Answers: No typing required! Simply hold the microphone button and speak your answer.
• Real-time, Grounded Questions: The AI generates fresh trivia questions on random topics, using Google Search grounding to ensure accuracy and timeliness.
• Conversational Feedback: The AI host provides in-character feedback, tells you if you're correct, and shares interesting facts related to the questions.
• Text-to-Speech Output: Hear your host's personality come to life as it speaks the questions and its witty responses.
• Session History: Track your progress with a detailed history of past games, including your final score, all questions, and your answers. Data is saved locally in your browser.
• Sleek, Modern UI: A responsive and futuristic interface built with Angular and Tailwind CSS.

🤖 How It Works

The application creates a seamless, conversational trivia experience through a series of interactions with the Gemini API and browser features.

1. Personality Selection: The user chooses a personality for the AI host.
2. Question Generation: The frontend calls the Gemini API (gemini-2.5-flash) with a prompt that includes the chosen personality. It requests a structured JSON object for a trivia question and uses Google Search grounding to pull from up-to-date information.
3. Voice Input: The user holds the microphone button to record their answer using the browser's MediaRecorder API.
4. Answer Evaluation: The recorded audio is sent to the Gemini API as a base64 string. The API transcribes the speech, evaluates if the answer is correct against the provided context, and generates a conversational response in the host's personality.
5. Audio Feedback: The host's response text is spoken back to the user via the browser's Web Speech API (Text-to-Speech).
6. History Persistence: When a game is finished, the entire session (personality, score, and all rounds) is saved to the browser's localStorage for future review.

🛠️ Tech Stack

• Framework: Angular (v20+, Zoneless, Standalone Components)
• State Management: Angular Signals
• AI/LLM: Google Gemini API (@google/genai)
• Styling: Tailwind CSS
• Browser APIs:
  • Web Speech API (SpeechSynthesis for Text-to-Speech)
  • MediaStream API (for microphone input)

🚀 Running the Project

This application is designed to run in a modern web browser that supports the required Web APIs.

API Key Configuration

The application requires a Google Gemini API key to function. It is configured to read the key from an environment variable:

process.env.API_KEY

Ensure this variable is available in the execution environment when the application is run.