This article details how to set up a custom voice pipeline in Home Assistant using free self-hosted tools like Whisper and Piper, replacing cloud-based services for full control over speech-to-text and text-to-speech processing.
A tutorial on building a private, offline Retrieval Augmented Generation (RAG) system using Ollama for embeddings and language generation, and FAISS for vector storage, ensuring data privacy and control.
1. **Document Loader:** Extracts text from various file formats (PDF, Markdown, HTML) while preserving metadata like source and page numbers for accurate citations.
2. **Text Chunker:** Splits documents into smaller text segments (chunks) to manage token limits and improve retrieval accuracy. It uses overlapping and sentence boundary detection to maintain context.
3. **Embedder:** Converts text chunks into numerical vectors (embeddings) using the `nomic-embed-text` model via Ollama, which runs locally without internet access.
4. **Vector Database:** Stores the embeddings using FAISS (Facebook AI Similarity Search) for fast similarity search. It uses cosine similarity for accurate retrieval and saves the database to disk for quick loading in future sessions.
5. **Large Language Model (LLM):** Generates answers using the `llama3.2` model via Ollama, also running locally. It takes the retrieved context and the user's question to produce a response with citations.
6. **RAG System Orchestrator:** Coordinates the entire workflow, managing the ingestion of documents (loading, chunking, embedding, storing) and the querying process (retrieving relevant chunks, generating answers).
1. **Document Loader:** Extracts text from various file formats (PDF, Markdown, HTML) while preserving metadata like source and page numbers for accurate citations.
2. **Text Chunker:** Splits documents into smaller text segments (chunks) to manage token limits and improve retrieval accuracy. It uses overlapping and sentence boundary detection to maintain context.
3. **Embedder:** Converts text chunks into numerical vectors (embeddings) using the `nomic-embed-text` model via Ollama, which runs locally without internet access.
4. **Vector Database:** Stores the embeddings using FAISS (Facebook AI Similarity Search) for fast similarity search. It uses cosine similarity for accurate retrieval and saves the database to disk for quick loading in future sessions.
5. **Large Language Model (LLM):** Generates answers using the `llama3.2` model via Ollama, also running locally. It takes the retrieved context and the user's question to produce a response with citations.
6. **RAG System Orchestrator:** Coordinates the entire workflow, managing the ingestion of documents (loading, chunking, embedding, storing) and the querying process (retrieving relevant chunks, generating answers).
Mail-Archiver is an open-source, self-hosted email archiving system that allows you to take back ownership of your communication data. It's built with .NET and PostgreSQL, offering a responsive web interface for managing and backing up emails from multiple accounts.
Inference Snaps are generative AI models packaged for efficient performance on local hardware, automatically optimizing for CPU, GPU, or NPU.
SPINQ Triangulum Ⅱ is a cost-effective, maintenance-free, and highly stable 3-qubit desktop NMR quantum computer for quantum computing education, demonstrations, and research. It supports any 3-qubit quantum algorithm, open hardware-level pulse sequence editing, and features classical-quantum hybrid programming.
This article details how the author created a dashboard to manage their self-hosted applications, focusing on the use of Homepage and its benefits for organization and access to service information.
Dozzle is a lightweight, self-hosted solution that provides a real-time look into your container logs, offering an intuitive UI, real-time logging, intelligent search, and support for multiple use cases like home labs and local development.
DispatchMail is an open source locally run (though currently using OpenAI for queries) AI-powered email assistant that helps you manage your inbox. It monitors your email, processes it with an AI agent based on your prompts, and provides a (locally run) web interface for managing drafts/responses, and instructions.
This article details how the author successfully set up a Pi-hole DNS server using a $15 Raspberry Pi Zero 2 W, highlighting its low cost, low power consumption, and surprisingly capable performance for handling DNS requests and ad blocking for a home network.
This article details how to enhance the Paperless-ngx document management system by integrating a local Large Language Model (LLM) like Ollama. It covers the setup process, including installing Docker, Ollama, and configuring Paperless AI, to enable AI-powered features such as improved search and document understanding.
