nowfound

AI · October 12, 2023

WT

Willow: the fastest and most private open source voice assistant

Hey HN! Voice assistants like Amazon Alexa and Google Home have sold hundreds of millions of devices. Unfortunately, they come with less-than-ideal things like an always on microphone (doing who knows what), data collection, and limited integration options. Echo devices especially also end up being advertising billboards you pay for and install in your own home. No thanks, I've already heard of the Barbie movie... Willow is an open source hardware and supporting software voice assistant platform. For $50 Willow hardware (from Espressif) is competitive with commercial products in every regard…

In plain words

Willow is an open source voice assistant platform built on affordable hardware that runs locally without cloud dependency. Designed for users concerned about privacy and data collection, it offers faster performance than commercial alternatives like Alexa while supporting integrations with home automation systems. The $50 hardware is self-hosted, eliminating always-on microphones and advertising, and provides REST API support for customization and third-party connections.

written from the facts on this page · September 2026

From the sources

In the maker’s words, at launch

Hey HN! Voice assistants like Amazon Alexa and Google Home have sold hundreds of millions of devices. Unfortunately, they come with less-than-ideal things like an always on microphone (doing who knows what), data collection, and limited integration options. Echo devices especially also end up being advertising billboards you pay for and install in your own home. No thanks, I've already heard of the Barbie movie... Willow is an open source hardware and supporting software voice assistant platform. For $50 Willow hardware (from Espressif) is competitive with commercial products in every regard while being open source and completely self-hosted. Did I mention it's at least twice as fast as Alexa? Willow supports integrations to Home Assistant, OpenHAB, and just about anything else with our REST command endpoint support. With our initial release Willow received overwhelming interest here on HN and elsewhere when we first announced it several months ago. Today (to the day) we're celebrating six months of Willow development! I made a (rare) demo video demonstrating our most recent release with features like wake only on device closest to the person talking, speaker voice authentication, dynamic configuration, over the air updates, device location, and our new web interface. All faster than ever. Check out our latest demo video: https://www.youtube.com/watch?v=qlhSEeWJ4gs The team and I really enjoyed the feedback and discussion in our first Show HN post months ago and we'd love to hear what you think about our progress. We also feel that at this point we've optimized accuracy and speed and are interested to hear any ideas for new functionality, etc.

More ai this month

the category →
  • I trained a 125M-parameter transformer to autocomplete piano performances in real time (~108 notes/sec on an iPhone 15). The idea is basically GitHub Copilot or Tabnine, except instead of prompting it with code, you prompt it by playing a few notes on a MIDI piano. The model then continues what you played, entirely on-device. The app is free if anyone wants to try it. Happy to answer questions about the model, training, Core ML, or the many things that didn't work.

    AI · 17d ago · simedw.com

  • Astute585

    Automate your B2B brand going viral, with new media creators

    AI · 18d ago · company-app.joinastute.com

  • Grok Bot547

    AI teammates that you can give real work to

    AI · 25d ago · x.ai

  • Hey HN, Henry from Cactus here! We previously released Cactus Needle, a 14MB agentic LLM for tool call, device use, and structured extraction for phones, wearables, smart homes, small robots and microcontrollers. We got really great feedback here, and have now incorporated the suggestions to release Needle 2. The whole model is a single 14MB binary that runs a full session in 28MB of RAM; 45m parameters at 2bit compression. Needle hits 500 tokens/sec decode speed on a Raspberry Pi 5, sits between 400-1,500 tokens/sec on VR devices like Meta Quest 3S and Apple Vision Pro, and ranges…

    AI · 27d ago · cactuscompute.com

  • Turn website visitors into qualified pipeline

    AI · 19d ago · clarasdr.ai

  • Kane CLI446

    Natural language browser & mobile app tests from terminal

    AI · 24d ago · testmuai.com

Launched alongside, October 2023

the whole month →
  • AI autopilot for content creation & marketing

    AI · 2023 · typeface.ai

  • Nudge 2.01,051

    In-app experiences to activate, retain, & understand users

    Growth · 2023 · nudgenow.com

  • World's first open source AI scheduling assistant

    AI · 2023 · cal.com

  • Unlock AI magic for elevated customer engagement, fast

    AI · 2023 · tiledesk.com

  • TalkNotes854

    Turn your thoughts into actionable notes, fast

    AI · 2023 · talknotes.io

  • OD

    Effortlessly discover API behaviour with a Chrome extension that automatically generates OpenAPI specifications in real time for any app or website.

    Dev tools · 2023 · github.com