Show HN: Privacy-first open-source AI voice recorder as a hardware plugin

openvision.engineering

8 points by danielrvega 4 days ago

Phones can often feel cumbersome when quick action is needed.

This is where a sleek, accessible, and intelligent solution comes in—an open-source, AI-powered pocket recorder. It syncs recordings locally, provides summaries, and runs on low-power ESP32-S3 hardware with Bluetooth and WebSocket support. The device attaches to a phone via MagSafe or works as a standalone unit, effortlessly capturing conversations, meetings, and spontaneous ideas.

The first prototypes were developed in a San Francisco garage, and the project is now being made open-source to let others build and customize it.

Your feedback on the device is highly valued!

The landing page is live: https://www.openvision.engineering/, and you can watch the product video here: https://www.youtube.com/watch?v=XupW9pi11MI.

If you’re in San Francisco, you’re welcome to see it in action! Located in Presidio, the creators are happy to meet in person and demonstrate the product.

evanjrowley 4 days ago

Instabuy. This feature set and open source privacy first mindset is exactly what I want.

nhggfu 2 days ago

"unlimited cloud storage" we've heard that story before...

danielrvega 4 days ago

The team has also implemented state-of-the-art research to improve voice labeling accuracy, using fingerprinting and diarization to create error-free transcripts and crystal-clear audio.

tashapais25 4 days ago

Extra disk space, tactile switch, microphone when you're on calls, extra battery life. Some prefer an app but I see the appeal

  • nlinck 4 days ago

    I dont get why people prefer the app more, it is so much friction to unlock my phone (wait for unlock), open an app (wait for it to load), and then press the record button. I'm very excited for hardware devices like this. Lets see if they can actually make it useful.

    • tashapais25 4 days ago

      Our core belief is to make only a few features that are incredibly accurate. People have tons of complaints on past AI wearables: UI is unclear when its recording, battery life only lasts a day, button is incorrectly placed, transcript fails to detect more than 2 speakers. All things that we've put a lot of thought into and have fixed.

xingwu 3 days ago

I don't get it; I already have a phone that can record.

guglielm0 4 days ago

super cool! excited to see how you can leverage external compute, while maintaining the privacy

  • tashapais25 4 days ago

    We're running models for transcription and voice classification. The data policy for those model providers are SOC2 compliant. On top of that we made the important decision to open source the device firmware and app code that will be public on github.com/open-vision-engineering. The audio files are stored on the standalone device's 64 gb and then is synced to your phone's on-device storage. We use AWS S3 to generate a signed url so only temporary access is enabled.

guglielm0 4 days ago

pty neat! Excited to see how you can leverage apple compute too

  • tashapais25 4 days ago

    Yes the grand vision for personal compute Apple is going for is iPhone running models acting on wearable sensors. Think invisible OS using agents.