Meeting Mind- An Hour Long Meeting Analyzed in Under 30 Seconds (Powered by Langflow)

MeetingMind is an AI-powered meeting assistant that helps you capture, analyze, and act on your meeting insights effortlessly. This project is built with Langflow, Next.js and Groq-based fast transcription service to analyze your meetings and generate insights.

Demo

Check out this demo video to see MeetingMind in action:

https://github.com/user-attachments/assets/50a9de7a-b24f-4167-9526-4e112b1d24f8

Features

Audio recording and file upload
AI-powered transcription
Automatic extraction of key information:
- Tasks
- Decisions
- Questions
- Insights
- Deadlines
- Attendees
- Follow-ups
- Risks
- Agenda

Getting Started

Prerequisites

Node.js 14.x or later
npm or yarn
A LangFlow server running locally
Git (for cloning the repository)

Caution

⚠️ Important: Groq Whisper used for transcription and analysis, currently supports files up to 25 MB only. There is a compression step in the process to reduce the file size to a manageable level. If your audio file is still larger than 25 MB, you will need to compress it before uploading. This limitation may affect the processing of longer meetings or high-quality audio recordings.

To compress your audio files further, you can use tools like:

Online audio compressors
FFmpeg (command-line tool for audio/video processing)

Ensure your compressed audio maintains sufficient quality for accurate transcription while staying under the 25 MB limit.

Installation

Clone the repository:

git clone https://github.com/yourusername/meetingmind.git
cd meetingmind

Install dependencies:
```
npm install
# or
yarn install
```
Set up LangFlow:
- Install and run the LangFlow backend server
- Upload the flow provided in the repo at utils/langflow_flow/Meeting Mind.json
- Note the URL of your LangFlow server
Create a .env.local file in the root directory and add the LangFlow URL:
```
LANGFLOW_FLOW_URL="http://127.0.0.1:7860/api/v1/run/5781a690-e689-4b26-b636-45da76a91915"
```
Replace the URL with your actual LangFlow server URL if different.

In the file app/api/transcribe/route.ts, locate the payload object and update the Groq component name to match your LangFlow component name. For example:
```
const payload = {
 output_type: 'text',
 input_type: 'text',
 tweaks: {
   'YourGroqComponentName': {
     audio_file: filePath
   },         
 }
}
```
Replace 'YourGroqComponentName' with the actual name of your Groq component in LangFlow.
Set up the database:

This project uses Prisma as an ORM. By default, it's configured to use SQLite as the database.

a. To use the local SQLite database:
- Ensure your .env file contains:
```
DATABASE_URL="file:./dev.db"
```
- Run the following commands to set up your database:
```
npx prisma generate
npx prisma migrate dev --name init
```
b. To use a different database (e.g., PostgreSQL with Neon):
- Update your .env file with the appropriate connection string:
```
DATABASE_URL="postgresql://username:password@host:port/database?schema=public"
```
- Update the provider in prisma/schema.prisma:
```
datasource db {
  provider = "postgresql"
  url      = env("DATABASE_URL")
}
```
- Run the Prisma commands as mentioned above to generate the client and run migrations.
Run the development server:
```
npm run dev
# or
yarn dev
```
Open http://localhost:3000 with your browser to see the result.

Usage

Navigate to the dashboard page.
Upload an audio file.
Wait for the AI to process and analyze the meeting.
Review the extracted information in the Dashboard.

Project Structure

app/: Contains the main application code
- components/: Reusable React components
- api/: API routes for server-side functionality
- dashboard/: Dashboard page component
- page.tsx: Home page component
public/: Static assets
prisma/: Database schema and migrations
utils/: Utility functions and configurations
lib/: Shared libraries and modules

Technologies Used

Langflow: For AI workflow management
Next.js: React framework for building the web application
React: JavaScript library for building user interfaces
Tailwind CSS: Utility-first CSS framework
Framer Motion: Animation library for React
Axios: Promise-based HTTP client
Prisma: ORM for database management
SQLite: Default database (can be changed to PostgreSQL or others)
Groq: AI model provider for transcription and analysis

Configuration

The project uses environment variables for configuration. Ensure all necessary variables are set in your .env.local file.
Tailwind CSS configuration can be found in tailwind.config.ts.
TypeScript configuration is in tsconfig.json.

API Routes

/api/meetings: Handles CRUD operations for meetings
/api/transcribe: Handles audio file transcription and analysis

Debugging

Use the browser's developer tools to debug client-side issues.
For server-side debugging, use console.log statements or attach a debugger to your Node.js process.

Performance Considerations

Large audio files may take longer to process. Consider implementing a progress indicator for better user experience.
Optimize database queries and indexes for improved performance as the number of meetings grows.

Screenshots

Landing Page

Dashboard

Meeting Summary

Meeting Details

These screenshots provide a visual representation of the application's main interfaces. The landing page showcases the initial user experience, while the dashboard displays the core functionality where users can upload audio files and view the AI-processed meeting information.

Contributing

Contributions are welcome! Please feel free to submit a Pull Request. Here are some ways you can contribute:

Report bugs and issues
Suggest new features
Improve documentation
Submit pull requests with bug fixes or new features

Please read our contributing guidelines before submitting a pull request.

License

This project is licensed under the MIT License. See the LICENSE file for details.

Support

If you encounter any problems or have questions, please open an issue on the GitHub repository.

Acknowledgements

Thanks to the Langflow team for providing the AI workflow management tool.
Special thanks to all contributors who have helped shape this project.

misbahsy / meetingmind

readme