The Subvocal Company captures movements of the tongue, lips, and soft palate by measuring how they modulate RF signals transmitted from the eyewear frame. The founders pivoted to this custom antenna sensing stack after determining that initial electromyography prototypes were too sensitive to sweat, placement inconsistencies, and electrical interference. To handle anatomical variations between users, the team gathered machine-learning training data by paying participants to read prompts while testing the hardware.
Key points
Radio frequency sensing allows the glasses to translate silent speech into text without r…: Radio frequency sensing allows the glasses to translate silent speech into text without requiring users to open their mouths.
The company is actively gathering custom training datasets to account for individual diff…: The company is actively gathering custom training datasets to account for individual differences in facial anatomy and edge cases.
Final hardware units from the Y Combinator-backed startup are scheduled to begin shipping…: Final hardware units from the Y Combinator-backed startup are scheduled to begin shipping to customers in the first quarter of 2027.
Sources checked by our analysis pipeline: 1
Hi HN, we're Aarush and Arnav from The Subvocal Company (YC F26) (
https://subvocal.company/). We’re building smart glasses that let you talk to your computer without talking out loud. You can mouth words silently and discreetly and the glasses decode them to text at the speed of speech.
Here's a video of Dave from YC trying it out: https://www.youtube.com/watch?v=9QdX1eObVYI
(Edit: you don't actually have to move your lips or even open your mouth for this to work! Since the RF can see inside your mouth the system can be fine tuned on "internal speech". This isn't the automatic tongue movements that occur when "thinking" words to yourself, but more like mouthing with your lips closed. It's a bit weird to do, but with practice, we have been able to get it down pretty well.)
Arnav and I were studying CS at Georgia Tech at the beginning of the year, with no experience in silent speech or sensing. We were both AI-native builders and power users of Wispr Flow. What bugged us was not being able to use these tools in…
Retailers across seven nations suffered severe food spoilage overnight after hackers manipulated operational controls to spike cooling temperatures up to 50C (122F). European Union officials warned the coordinated disruptions represent targeted efforts to destabilize the region's food supply and erode public confidence. The breach highlights escalating risks to remotely monitored retail infrastructure amid broader security concerns over hybrid warfare in Europe.
The enterprise database deployment allows organizations to execute serverless, mission-critical operations directly within the Databricks Data + AI Platform. By removing standard infrastructure management overhead, teams can power low-latency real-time applications and autonomous AI agents more effectively. The launch directly reduces operational complexity for businesses running mixed transactional and analytical workloads on Google Cloud.
The tech giant is shifting its focus toward mixed reality hardware to prepare its ecosystem for next-generation devices releasing in spring 2027. Despite millions of active users remaining on the discontinued hardware, future development will prioritize newer headsets that offer advanced processing power and shared software kits. This phaseout establishes a technical foundation for the upcoming product lineup while giving current users ample transition time.