notesGPT: AI-Powered Voice Note-Taking
notesGPT is a cutting-edge platform designed to streamline the process of generating action items from voice notes, powered by advanced technologies such as Convex, Together.ai, and Whisper. With the aim of transforming note-taking, notesGPT helps users efficiently turn recorded speech into actionable tasks in just a few seconds.
Tech Stack
The technological backbone of notesGPT is composed of a variety of powerful tools and services:
- Convex: Utilized for its database and cloud functions, ensuring reliable and scalable data management.
- Next.js App Router: Provides a robust framework for the application's architecture.
- Together Inference and Embeddings: These AI technologies facilitate deep learning models for language processing (LLM Mixtral) and create embeddings for efficient search functions.
- Convex File Storage and Vector Search: Essential for storing voice notes and performing fast, vector-based searches.
- Replicate for Whisper Transcriptions: Enables high-quality speech-to-text conversion, allowing voice notes to be transcribed quickly and accurately.
- Clerk: A secure system for user authentication.
- Tailwind CSS: Offers a sleek design with customizable styling options.
Deploy Your Own
Deploying a notesGPT template involves several steps to configure and run the system using necessary environment variables:
- Dependency Installation: Begin by running
npm install
to download project dependencies. - Development Server: Use
npm run dev
to start the development server, which will guide you through logging into Convex and setting up a project. - Clerk Setup: Create a Clerk account, then configure it by copying necessary API keys into the
.env.local
file and setting theCLERK_ISSUER_URL
in Convex environment variables. - User Authentication: Once configured, the frontend and backend should operate, allowing login capabilities.
- API Key Registration: Get API keys by setting up accounts with Together.ai and Replicate, then save them in the Convex environment as specified keys.
Future Tasks
notesGPT has future enhancements lined up to improve user experience and functionality:
- Enhanced Recording Features: Plans to retain and display recordings for future use.
- Dynamic Interface Elements: Aiming to sync animated elements with real-time voice input.
- Action Items Management: Improving the way completed items are stored and edited.
- Responsive UI Design: Optimizing text and image sizes for various screen dimensions.
- Improved Search and PWA Accessibility: Smoother search tools and potential development into a Progressive Web App.
- Notion Integration: Seamlessly transferring transcripts, summaries, and action items to Notion.
- User Interface Refresh: Applying visual updates based on new design concepts.
- Whisper Migration: Transitioning to a faster Whisper version for even quicker transcriptions.
notesGPT positions itself as a revolutionary tool for anyone looking to enhance their productivity by simplifying the process of turning voice inputs into structured, actionable outputs.