Set up a private voice-controlled home assistant
They ran a local AI on their own hardware and used voice to manage home tasks without sending data to outside services.
- What they did
- They installed a local model with Ollama on their PC, choosing one small enough to fit their GPU. Voice input came through Whisper, and Piper handled spoken replies. That setup let the agent talk to a local Home Assistant system and carry out tasks like managing groceries, controlling lights and AC, checking garden moisture, and helping with language practice. Remote access was added with WireGuard or Tailscale.
- How it went
- The setup worked reasonably well for those jobs, but the model was weaker than cloud-based options. The main benefit was full local control with no outside data sharing.
- Tips
- This kind of build can take advanced setup work at the start. If you want privacy, keep the model, voice tools, and home system running on your own network.
Try it yourself with Hermes Agent
You are helping me build a voice-controlled home setup on my own machine. Use [local model name] through [local runtime] on [my computer specs], and connect it to [home automation platform] with voice input from [speech-to-text tool] and spoken replies from [text-to-speech tool]. I want it to handle [task 1], [task 2], [task 3], and [task 4], plus answer a question like "[example status question]". First, outline the steps and confirm which parts are safe to change. Before you make any irreversible or safety-sensitive change, such as sending a command, deleting data, changing heating, unlocking doors, or buying anything, stop and ask for my approval. After setup, test at least five spoken commands and report what worked, what failed, and any errors clearly.
Fill in the [brackets]. Agents can make mistakes, so review anything before it is sent, paid or deleted.
Get 3 agent uses like this every morning
With the day's top AI news. Free, and one click to unsubscribe.
Related agent uses
Set up MCP access for a household wiki
They attached an MCP server to Wiki.js so an LLM could edit household pages.
Set up a Friday night plan with friends
An Instinct agent picked the menu, movie, timing, and order for a group night without any group chat.
Use an agent to plan a workup for low energy
Noah used an agent to turn vague fatigue into a list of tests and doctors, and it pointed him to the right answer first.
Scan my inbox and calendar for repeats and offer to handle them
An app built on Gmail and Calendar looks for recurring tasks, then asks whether to prep meetings, emails, or bills.
Draft tailored job materials before you apply
A commenter had an agent prepare role-specific CVs and cover letters so they could review and edit before submitting applications.
Set up a nutrition bot for meal choices and calorie logs
Zack made a health agent for his mom that helps pick meals and track calories for weight loss.