{"slug": "hotword-detection-in-fhe-on-gpu-in-350ms", "title": "Hotword detection in FHE on GPU in 350ms", "summary": "Belfort Labs launched a live demo at hotword.belfortlabs.com running a fully homomorphic encryption (FHE) hotword detector on its proprietary GPU platform, with server-side evaluation completing in 350ms, according to HEIR compiler creator Jeremy Kun. The round trip breaks down as 15ms for client-side encryption, 350ms for server-side evaluation, and 600ms for network latency, totaling under one second, though one-time setup encryption keys still take about a minute to generate in the browser and upload. The demo, compiled with the HEIR homomorphic encryption compiler, encrypts spoken audio client-side and returns an encrypted result that the client decrypts to update a word-connection game, and Belfort has opened an early access program for its FHE-based cloud platform.", "body_md": "A quick update on our homomorphic encryption\ncompiler [HEIR](https://heir.dev): our partners at Belfort\nhave a hotword detector in FHE that runs in 350ms.\n\nMore specifically, they launched a live demo at [hotword.belfortlabs.com](https://hotword.belfortlabs.com/),\nthat has a little game inspired by the NY Times Connections puzzle.\nInstead of typing in your guess for the word\nthat connects a group of clues, you say it out loud.\n\nYour audio is encrypted under FHE, transitted to their server, and a hotword detector runs on it to determine which word you said (still encrypted). The client receives the encrypted result, decrypts it, and updates the game state.\n\nNot only was this demo compiled with HEIR to target Belfort’s (proprietary) GPU platform, but the inference step runs in 350ms.\n\nMore specfically, the breakdown of runtime is: 15ms for client-side encryption, 350ms for server-side evaluation, 600ms for network latency (my Portland-based laptop to Belfort’s servers which are likely in Belgium) for a total round-trip of less than one second.\n\nThe one caveat is that they pre-load the one-time setup encryption keys, which currently take a minute or so to generate in the browser and upload to the server. So if you want to play the game securely, you need that extra step.\n\nOf course, this could all be done locally, but it’s an impressive showcase of how fast FHE has become. And to my delight, how useful our compiler has become for compiling FHE workloads!\n\nThere’s still a lot more work to do, but this milestone makes me proud of the community of academics and practitioners we’ve grown around this compiler. Hooray for collaboration!\n\nBelfort also has an early access program for their FHE-based cloud platform\nwhich you can sign up for [here](http://eml.belfortlabs.com/).\nYou can forgive me for the advertisement,\nbut they intend to use HEIR as a primary tool.\nSo if you use their platform, you’ll be helping find bugs and improvemenets in the HEIR compiler!\n\nWant to respond? [Send me an email](mailto:mathintersectprogramming@gmail.com),\n[post a webmention](https://webmention.io/www.jeremykun.com/webmention),\nor find me [elsewhere on the internet](https://www.jeremykun.com/about/).\n\nThis article is syndicated on:", "url": "https://wpnews.pro/news/hotword-detection-in-fhe-on-gpu-in-350ms", "canonical_source": "https://www.jeremykun.com/shortform/2026-10-08-1120/", "published_at": "2026-10-15 14:39:20+00:00", "updated_at": "2026-10-15 14:55:13.166790+00:00", "lang": "en", "topics": ["ai-safety", "ai-infrastructure"], "entities": ["HEIR", "Belfort Labs", "Jeremy Kun", "hotword.belfortlabs.com"], "also_reported_by": [], "alternates": {"html": "https://wpnews.pro/news/hotword-detection-in-fhe-on-gpu-in-350ms", "markdown": "https://wpnews.pro/news/hotword-detection-in-fhe-on-gpu-in-350ms.md", "text": "https://wpnews.pro/news/hotword-detection-in-fhe-on-gpu-in-350ms.txt", "jsonld": "https://wpnews.pro/news/hotword-detection-in-fhe-on-gpu-in-350ms.jsonld"}}