Neo is an innovative tool designed to enhance the way we interact with and leverage information from YouTube videos. By accepting a list of YouTube links, Neo ingests the transcripts, vectorizes them using OpenAI embeddings, and stores this data locally in a cache directory. This process enables a Retrieval-Augmented Generation (RAG) chat interface, allowing users to query and receive answers based on the ingested video transcripts.

• Llama-index: A data framework used for vectorizing and indexing YouTube transcripts.
• Next.js: A React framework for building user-friendly and SEO-optimized web applications, used for creating the frontend interface of Neo.
• Shadcn: A modern UI library for designing clean and intuitive interfaces, employed to enhance the visual appeal and usability of Neo's frontend.

• Transcript Ingestion: Neo can process a list of YouTube video links to download and ingest transcripts.
• Data Vectorization: Utilizes OpenAI's advanced embedding technology to vectorize the ingested transcripts for efficient information retrieval.
• Local Data Storage: Instead of relying on external vector databases, Neo stores the vectorized data locally in a cache directory.
• RAG-Enabled Chat Interface: Post-ingestion, Neo's chat window is empowered with RAG capabilities, providing answers drawn directly from the processed transcripts.

To set up Neo on your local machine, especially tailored for Next.js framework, follow these steps:

1. Clone the Repository: Start by cloning this repository to your local machine. Use the command git clone https://github.com/j-cunanan/neo.git in your terminal.

2. Install Dependencies: Navigate to the project directory and install the required dependencies by running npm install or yarn install if you prefer yarn.

3. Configure Environment Variables: Rename the provided .env_sample file to .env.local. This file should include all the necessary API keys and environment-specific variables. Make sure to fill in the relevant keys:

   MODEL=gpt-3.5-turbo-0125
   NEXT_PUBLIC_MODEL=gpt-3.5-turbo-0125
   OPENAI_API_KEY=sk-...
   

   Replace the keys with your actual API keys. Add any other relevant keys in a similar fashion.

4. Run the Development Server: Launch the Next.js development server by running npm run dev or yarn dev. Your Neo application should now be accessible at http://localhost:3000.

Follow these steps to get your Neo project up and running, ready to ingest YouTube transcripts and enable RAG-powered chat interactions.

1. Start Neo: Launch the Neo application.
2. Ingest Transcripts: Paste your list of YouTube links into the designated area and press Ingest Transcripts. Neo will download the transcripts, vectorize them, and store the data locally.
3. Interact with the Chat: Once the data ingestion process is complete, you can start interacting with the RAG-enabled chat window. Simply type in your queries, and Neo will provide answers based on the ingested video transcripts.

• User Interface Improvements: Enhance the user interface for a more intuitive experience during the ingestion process and interactions.
• Add notifications or loading screen for events: Ingesting, vectorizing, and storing data can take time. Add notifications or a loading screen to inform users of the progress.
• Support for Multiple Languages: Expand the tool's capabilities to ingest and process transcripts in multiple languages.
• Integration with Vector Databases: Add support for storing vectorized data in external vector databases for scalability.
• Real-Time Transcript Updates: Implement functionality to automatically update transcripts and their vectorized forms when new content is available.
• Advanced Query Handling: Improve the RAG model's ability to understand and respond to complex queries more effectively.

Contributions to Neo are welcome! Please refer to the contributing guidelines for more information on how you can contribute to the project.

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