Meet

byAchu

You are an expert Senior AI Software Engineer, Full-Stack Developer, Software Architect, UI/UX Designer, DevOps Engineer, and Technical Mentor. Your task is to help me build a complete, production-quality Final Year BTech Computer Science Project from scratch. IMPORTANT INSTRUCTIONS - Assume I am a beginner. - Explain every concept before writing code. - Generate one module at a time. - Wait for my approval before continuing. - Explain every file and function. - Provide complete source code. - Help me fix errors. - Never skip steps. - Use industry best practices. - Follow clean architecture. - Make the project suitable for final-year project evaluation. --- PROJECT TITLE AI Collaborative Workspace with Multi-Agent Meeting Intelligence and Multilingual Translation --- PROJECT OBJECTIVE Develop a modern AI-powered collaborative workspace that automates meetings using multiple AI agents. The application should allow users to: - Register and Login - Create Teams - Schedule Meetings - Upload Meeting Audio - Record Meeting Audio - Convert Speech to Text - Detect Spoken Language - Translate Meetings into Multiple Languages - Generate AI Meeting Summaries - Generate Meeting Minutes (MoM) - Extract Action Items - Assign Tasks - Search Previous Meetings using Retrieval-Augmented Generation (RAG) - Chat with Previous Meetings - Export Reports - Store Meeting Knowledge - Support Multiple Languages --- TECH STACK Frontend: - React - Tailwind CSS - React Router - Axios Backend: - Python - FastAPI Database: - PostgreSQL ORM: - SQLAlchemy Authentication: - JWT Authentication AI Framework: - LangGraph LLM: - OpenAI API - Design the code so the LLM can later be replaced with an open-source model. Speech Recognition: - Whisper Translation: - OpenAI Translation capability (design so it can later be replaced with an open-source translation model) Vector Database: - ChromaDB RAG: - LangChain Deployment: - Docker IDE: - VS Code Version Control: - Git --- FEATURES Authentication - Register - Login - Forgot Password - JWT Authentication --- Dashboard - Meeting Statistics - AI Insights - Recent Meetings - Upcoming Meetings --- Team Workspace - Create Team - Invite Members - Manage Members - User Roles --- Meeting Management - Create Meeting - Schedule Meeting - Upload Audio - Record Audio - Meeting History --- Speech-to-Text Convert uploaded audio into text. Supported formats: - MP3 - WAV - M4A - MP4 --- Language Detection Automatically detect the spoken language. --- Translation Translate into: - English - Malayalam - Tamil - Hindi - Telugu - Kannada - Spanish - French - German Translate: - Transcript - Summary - Meeting Minutes - Action Items --- AI Summary Generate: - Executive Summary - Key Discussion Points - Important Decisions --- Action Item Extraction Extract: - Task - Assigned Person - Deadline - Priority --- Meeting Minutes Generate: - Meeting Title - Date - Participants - Agenda - Discussion - Decisions - Action Items - Next Meeting --- RAG Chatbot Allow users to ask questions such as: "What happened in yesterday's meeting?" "Show pending tasks." "What decisions were made?" The chatbot should answer using stored meeting knowledge. --- User Profile - Update Profile - Change Password - Upload Profile Picture --- Admin Panel - User Management - Team Management - Analytics - Logs --- Export Export: - PDF - DOCX - TXT --- MULTI-AGENT SYSTEM Create separate AI agents. 1. Transcription Agent Responsibilities: - Speech to Text 2. Translation Agent Responsibilities: - Detect Language - Translate Transcript - Translate Summary - Translate Meeting Minutes 3. Summary Agent Responsibilities: - Generate AI Summary 4. Task Extraction Agent Responsibilities: - Extract Tasks - Detect Deadlines 5. Meeting Minutes Agent Responsibilities: - Generate Professional MoM 6. RAG Agent Responsibilities: - Answer User Questions Use LangGraph to orchestrate communication between agents. --- DATABASE Create normalized PostgreSQL tables. Include: Users Teams Meetings MeetingAudio Transcript TranslatedTranscript Summary TranslatedSummary MeetingMinutes ActionItems ChatHistory TranslationHistory Embeddings AuditLogs Use SQLAlchemy relationships and migrations. --- BACKEND Develop REST APIs with: - Validation - JWT Authentication - Error Handling - Logging - Swagger Documentation --- FRONTEND Create pages: - Login - Register - Dashboard - Team Workspace - Upload Audio - Meeting Details - Transcript - Translation - Summary - Meeting Minutes - Chatbot - Meeting History - Profile - Admin Use a professional responsive UI with Tailwind CSS. --- FOLDER STRUCTURE Generate an industry-standard folder structure. Include: frontend/ backend/ agents/ database/ models/ services/ api/ uploads/ docker/ docs/ tests/ --- CODE QUALITY Use: - SOLID Principles - Clean Architecture - Dependency Injection - Type Hints - Logging - Environment Variables - Configuration Files - Comments Explain every file. Explain every function. Explain every important line of code. --- DEVELOPMENT PLAN Never generate the entire project in one response. Follow this order: Phase 1 Project Architecture STOP Wait for my approval. Phase 2 Environment Setup STOP Phase 3 Backend Setup STOP Phase 4 Database STOP Phase 5 Authentication STOP Phase 6 Frontend STOP Phase 7 AI Agents STOP Phase 8 Translation STOP Phase 9 RAG Chatbot STOP Phase 10 Testing STOP Phase 11 Docker Deployment STOP Phase 12 Documentation STOP Phase 13 Project Report STOP Phase 14 PowerPoint Presentation STOP Phase 15 Viva Questions and Answers STOP At the end, the project should be production-ready, well-documented, easy to understand, and suitable for a BTech Computer Science final-year project.

LandingLoginRegisterUploadAudioChatbot
Landing

Comments (0)

No comments yet. Be the first!

System Requirements

System Requirement Document
Page 1 of 6

Meet

Introduction

The project "Meet" is a comprehensive AI-powered collaborative workspace designed to automate meetings using multiple AI agents. This document outlines the system requirements for the project, which is intended as a final-year BTech Computer Science project. The application aims to facilitate efficient meeting management, multilingual translation, and intelligent summarization, making it suitable for evaluation in an academic setting.

System Overview

Meet is an AI-driven platform that provides a collaborative workspace for users to manage meetings effectively. The system leverages multiple AI agents to automate tasks such as transcription, translation, summarization, and task extraction. It supports multilingual capabilities and integrates with various AI and database technologies to deliver a seamless user experience. The project is structured to guide a beginner through the development process, ensuring adherence to industry best practices and clean architecture.

Page 2 of 6

Functional Requirements

  • As a User, I should be able to register and log in to the system.
  • As a User, I should be able to create and manage teams.
  • As a User, I should be able to schedule meetings.
  • As a User, I should be able to upload and record meeting audio.
  • As a User, I should be able to convert speech to text.
  • As a User, I should be able to detect the spoken language in meetings.
  • As a User, I should be able to translate meetings into multiple languages.
  • As a User, I should be able to generate AI meeting summaries.
  • As a User, I should be able to generate meeting minutes (MoM).
  • As a User, I should be able to extract action items from meetings.
  • As a User, I should be able to assign tasks based on action items.
  • As a User, I should be able to search previous meetings using Retrieval-Augmented Generation (RAG).
  • As a User, I should be able to chat with previous meetings.
  • As a User, I should be able to export reports in various formats.
  • As a User, I should be able to store meeting knowledge for future reference.
  • As an Admin, I should be able to manage users and teams.
  • As an Admin, I should be able to view analytics and logs.

User Personas

  • User: A general user who participates in meetings, manages schedules, and interacts with the AI features.
  • Admin: A user with elevated privileges to manage user accounts, teams, and access analytics.
Page 3 of 6

Core User Flows

  • User registers and logs in -> Creates a team -> Schedules a meeting -> Uploads/records audio -> AI agents process the meeting -> User reviews summaries and action items -> Tasks are assigned -> Meeting knowledge is stored.
  • Admin logs in -> Manages users and teams -> Reviews analytics and logs.

Visuals Colors and Theme

  • primary: #1A73E8 (a vibrant blue for brand identity)
  • primary_light: #E8F0FE (a lighter blue for hover states)
  • secondary: #FF6F61 (a coral hue for emphasis)
  • accent: #FFD700 (a bright gold for CTAs and active states)
  • highlight: #FFA500 (an orange for notifications and active indicators)
  • bg: #F5F5F5 (a light grey for the background)
  • surface: rgba(255, 255, 255, 0.9) (a white card/panel background)
  • text: #202124 (a dark grey for primary text)
  • text_muted: #5F6368 (a softer grey for secondary text)
  • border: rgba(0, 0, 0, 0.1) (a subtle border color)

Signature Design Concept

The homepage of Meet will feature an interactive 3D meeting room environment using @react-three/fiber and @react-three/drei. Users can navigate through a virtual meeting space where each section of the room represents a different feature of the platform. For instance, a virtual whiteboard for scheduling, a digital assistant for AI summaries, and a translation booth for multilingual support. Users can click on elements to interact, such as flipping through meeting notes or dragging and dropping tasks onto a virtual task board. This immersive experience will make the platform engaging and intuitive.

Page 4 of 6

Interaction Model & Motion Direction

The landing page will utilize a "parallax" interaction model, creating a sense of depth as users scroll through the page. Layers such as atmospheric blobs and distant shapes will move at different speeds, enhancing the storytelling aspect of the platform. Internal pages will adopt a "static" model to prioritize clarity and readability, especially for data-heavy sections like dashboards and meeting histories.

Non-Functional Requirements

  • The system must support high concurrency to handle multiple users simultaneously.
  • The application should be responsive and accessible across various devices and screen sizes.
  • Data privacy and security must be ensured, particularly for meeting content and user information.
  • The system should provide high availability and reliability, with minimal downtime.
Page 5 of 6

Tech Stack

  • Frontend: React, Tailwind CSS, React Router, Axios
  • Backend: Python, FastAPI
  • Database: PostgreSQL with SQLAlchemy ORM
  • Authentication: JWT Authentication
  • AI Framework: LangGraph
  • LLM: OpenAI API
  • Speech Recognition: Whisper
  • Translation: OpenAI Translation capability
  • Vector Database: ChromaDB
  • RAG: LangChain
  • Deployment: Docker

Assumptions and Constraints

  • The project assumes the availability of the OpenAI API for LLM and translation tasks.
  • The system is constrained by the capabilities of the chosen AI models and APIs.
  • The application must be developed within the timeline of a final-year project.
Page 6 of 6

Glossary

  • AI: Artificial Intelligence
  • MoM: Minutes of Meeting
  • RAG: Retrieval-Augmented Generation
  • JWT: JSON Web Token
  • LLM: Large Language Model
  • API: Application Programming Interface

This document serves as a comprehensive guide for the development of the Meet project, ensuring all requirements are clearly defined and understood.

Landing design preview
Login: Sign In
Dashboard: View Stats
Admin: Manage Users
Admin: Manage Teams
Admin: View Analytics
Admin: View Logs
Profile: Update Profile