NaomiProject/Naomi

Naomi – Open‑source, always‑on voice assistant

What it is – Naomi is a Python‑based platform that lets you run a continuously listening, voice‑controlled application on modest hardware (Raspberry Pi or any Linux box). It forks the older Jasper project and adds a modular plug‑in system so you can write your own “skills” (e.g., query the web, post to social media, control smart‑home devices).

Key capabilities

• Always listening – microphone input is captured continuously; a wake‑word (or simple sound‑threshold) triggers command parsing.
• Extensible skill framework – developers add Python modules that expose a handle(text, mic, profile) function; Naomi routes spoken phrases to the appropriate skill.
• Hardware‑agnostic – runs on a Raspberry Pi or any Linux machine with a supported microphone.
• Fully open source (MIT) – all code, documentation, and example skills are public, encouraging custom builds.

Typical use‑cases

• Home automation: “Naomi, turn the living‑room lights on.”
• Information queries: “Naomi, what’s the weather today?”
• Social media posting: “Naomi, tweet ‘Hello world!’”
• Personal reminders, timers, and calendar look‑ups.

Getting started

1. Install – download the latest release from the website or clone the repo, then run pip install -r requirements.txt (Python 3.5+).
2. Configure – edit the profile.yml (microphone device, wake‑word, API keys for services you want to use).
3. Run – execute bin/naomi to start the daemon; it will begin listening.
4. Add skills – drop a new Python module into the skills/ directory and follow the simple skill‑API documented on the site.

Documentation & support – Full developer docs are hosted at https://projectnaomi.com/dev/docs, and a Discord community is available for troubleshooting and ideas.

Contributing – The project welcomes pull requests. See the CONTRIBUTING.md guide for coding style, testing, and the process for adding new skills.

License – MIT (permissive, commercial‑friendly).

---

Naomi continues the spirit of the Jasper voice‑assistant project, offering a modern, community‑driven foundation for always‑on speech interfaces.