Subvocal launches silent speech recognition glasses
YC-backed startup demos smart glasses that decode mouthed words to text without audible sound, raising questions about accuracy and privacy.
What to know
- Subvocal Company (YC F26) demonstrated smart glasses using RF sensing in the temples to decode silently mouthed speech to text at natural speed, with under 500ms latency via a cloud companion app.
- Community consensus questions whether silent speech is intuitive to use, notes accuracy is undisclosed and insufficient, and doubts the specialized hardware necessity given rapid advances in software-only lip-reading and whisper-detection technologies.
- Privacy concerns surfaced around RF exposure near the brain and potential for surveillance through lip-reading, echoing Google's prior rejection of similar technology for privacy reasons.
- The form factor and usability barriers remain contested, with commenters advocating for earbuds, electrode pads, or attachments to existing eyewear rather than new smart glasses, citing comfort and accessibility issues.
The dispute Whether specialized RF-sensing hardware in glasses is necessary versus inevitable commodity software solutions (lip-reading, whisper detection) will make it obsolete before reaching market. · positions read across 62 posts and comments
Silent speech is a genuinely cool concept with sci-fi potential but not yet practical for real users.
-
“I think this is incredibly cool and straight out of a sci-fi novel! I think this is phenomenal for accessibility and for hands-free computing in general.”
sourpanda · Hacker News ↗
Hardware glasses are unnecessary; software-only lip-reading and whisper-detection will achieve the same goal.
-
“Aren't you worried that video to text (AI lip reading) will be good enough so it can be a software only solution with any webcam?”
eranation · Hacker News ↗
Accuracy is the critical missing piece; without transparency on performance, this cannot be trusted as a product.
-
“Accuracy is the sales objection. Your website does a terrible job addressing that.”
edmundsauto · Hacker News ↗
The glasses form factor creates unnecessary barriers; existing eyewear compatibility or different wearables would be more practical.
-
“If the tech is mostly in the glasses temples, I would prefer to have something that attaches to my existing glasses. I have prescription glasses and a big head.”
growt · Hacker News ↗
Aarush Subvocal Company co-founderArnav Subvocal Company co-founderThe Subvocal Company Startup (YC F26)
How it unfolded 8 developments, newest first · click a bar or a number to jump articlespostscomments
-
8
Skepticism emerges about hardware necessity vs. software-only alternatives
Commenters questioned whether the specialized hardware is necessary given rapid advances in AI lip-reading from standard webcams, and whether future earbuds with audio-based whisper detection might achieve similar goals without new form factors.
“Aren't you worried that video to text (AI lip reading) will be good enough so it can be a software only solution with any webcam?”
— eranation -
I don’t want to talk to any device. Ever. Even if it’s just mouthing words and not speaking out loud. I understand that this can be an accessibility feature that some people might want or need to turn on, but it’s not a useful feature for the general population.Using your mouth to communicate is low bandwidth. I can type way faster than I can…
2 more of the top 3 · 29 posts in this stretch
-
> there's no way to say it nicely, I have a big nose.But there is!"Érase un hombre a una nariz pegado, érase una nariz superlativa, érase una nariz sayón y escriba, érase un peje espada muy barbado" (Francisco de Quevedo, 1580-1645)Or, as translated by Timothy Adès (https://timothyades.com/francisco-de-quevedo-1580-1645-a-una...):"There was a…
-
We’d like to place a pre-order for a couple of sets right now. Could you please answer these questions:- Are you going to offer inference for free forever? Can I consider the $499 price to include this lifetime promise?- If you may charge in the future, what is the likely token cost?- Do you plan to offer a tiny model for local use, while keeping…
-
-
7
Potential customer cites accuracy as the key barrier to purchase
A commenter who considered pre-ordering criticized the FAQ for not providing concrete accuracy numbers, citing it as the main sales objection. They noted the website fails to address this crucial question.
“Accuracy is the sales objection. Your website does a terrible job addressing that.”
— edmundsauto -
Very cool, this is how Ender communicates with his AI in Xenocide and Speaker for the Dead.I would have preordered but your accuracy FAQ is terrible. It amounts to “it’s really bad now so we won’t give a number, we hope to improve it”. If you were 90%+ with the same promise I would have paid earlyAccuracy is the sales objection. Your website does…
2 more of the top 3 · 5 posts in this stretch
-
Why the eyeglass form factor? As people have mentioned, fitting eyewear is no joke. Not everyone has a face for RayBans.Why not simpler, more discreet electrode pads? Get nifty with flex PCB and wearables instead of another "Smart Glasses".Sweatband, tiara, smart crown, halo.. No crowding here.
-
putting the 'Due Today' price in a larger font and hiding the total smaller, under that price is sneaky.Big companies do it too, but my feedback has little chance of reaching their ears.
-
-
6
Crowdsourcing reveals prior and concurrent research in silent speech
Commenters cited existing projects (LipFlow, HushKey, Chaplin, LipNet.ai) and noted Cornell has had students working on variants for ~5 years. One raised concerns about RF exposure near eyes/brain and battery life.
“I'd be concerned both about battery life, and the long term health consequences of having an RF chip so close to my eyes/brain.”
— randomImmigrant -
> "you don't actually have to move your lips or even open your mouth for this to work ... It's a bit weird to do"In the video Dave just finished the training mode and is silently moving his lips, which seems pretty natural to me. I just spent a minute trying to do 'internal speech' (as many elements of speech as possible with lips closed). It does…
2 more of the top 3 · 8 posts in this stretch
-
Google wanted to put this tech into its phone, but then was convinced it was a bad idea due to privacy invasion. It could for instance correlate sexual arousal with any phone interaction (video chat, youtube, display ads, etc), giving google and advertisers the ability to make custom targeted ads for specific users, as it would "know your…
-
Really cool. I’m especially drawn to this as the time from thought to mouthing the words is so much faster than speaking out loud, for me.But I’d be concerned both about battery life, and the long term health consequences of having an RF chip so close to my eyes/brain. Have you explored these aspects much?
-
-
5
Commenters raise concerns about accuracy, form factor, and usability
Multiple users questioned whether silent speech is intuitive to perform, noted the accuracy disclaimer on the FAQ, and asked why glasses rather than earbuds, electrode pads, or attachments to existing eyewear. One mentioned Google's prior decision to reject similar tech due to privacy concerns.
“I don't think it is an intuitive action at all and I became self-conscious about not making a sound rather than talking like I normally would.”
— 0x4e -
Yep, we have been quite set on nose pads to keep them comfortable. I will keep you updated with the designs!The decoding is done via a companion app on your laptop or phone in the cloud, but end to end latency is still quite fast, under 500ms usually.To begin with we don't plan to have a wake word, but rather a trigger key on your connected device…
2 more of the top 3 · 7 posts in this stretch
-
I think these have a place but I can't say the same for most other new AI communication devices. Creeper glasses, pins, watches, desktop units etc. I want to be able to speak and hear from my agents at all times but why would I want a new device when I already wear earbuds? New products are being designed to pick up low murmuring/whispering, like…
-
One thing you might consider is actually telling people what it is on the homepage. Took me a fair bit of work to understand that this is effectively a pair of smart glasses.Maybe I could have figured that out from the video but for folks who can’t watch the video it’s quite hard to discover what you’re building.
-
-
4
Founders respond to questions about accuracy, processing, and design
Aarush confirmed the team is committed to adjustable nose pads for fit, that decoding occurs via companion app with under 500ms latency, and they plan RF-based activity detection rather than a wake word—similar to Whisper Flow. They also acknowledged ongoing R&D on silent speech activity detection.
“The decoding is done via a companion app on your laptop or phone in the cloud, but end to end latency is still quite fast, under 500ms usually.”
— aarushyadav -
Just preordered -- glasses request: there's no way to say it nicely, I have a big nose. With a high bridge. Many glasses are really tough to get properly placed; may I request you not give up on the pads w/metal adjusters that you have in the product photos? Don't let a designer take them away from you, us folks not one standard deviation from…
2 more of the top 3 · 3 posts in this stretch
-
For the ones thinking this is great, have you actually tried to talk without making sounds?I don't think it is an intuitive action at all and I became self-conscious about not making a sound rather than talking like I normally would.I suppose it would require some "getting used to."Regardless, I think it is a cool idea and a real problem…
-
If the tech is mostly in the glasses temples, I would prefer to have something that attaches to my existing glasses. I have prescription glasses and a big head, tried the Even G2 smart glasses and the available frames set just too small for me
-
-
3
Users discuss sci-fi applications and real-world use cases
Commenters drew parallels to science fiction (the 'whispering earring' concept) and identified practical scenarios like speaking in loud environments during yard work or using it with encrypted messaging platforms. Others noted it could work across language barriers.
“Imagine having a full on conversation with an AI via subvocalization and listening with ear bud to guide you throughout the day.”
— howunfortunate -
This is:1. Insanely cool2. Gives me serious "whispering earring" vibes. Imagine having a full on conversation with an AI via subvocalization and listening with ear bud to guide you throughout the day.
-
-
2
Commenters question whether the technology is credible or vaporware
Multiple Hacker News users expressed skepticism, with one noting the website focuses too heavily on the problem without explaining the underlying technology, making it feel deceptive. Another questioned whether a Show HN submission should involve product sales rather than open feedback.
“The issue is that this is contrary to people's experience of what is possible, so without discussing WHY you can do this, it looks like a scam or vaporware.”
— kenferry -
I am so glad to see an unpolished early look and an honest evaluation of about five percent error rate in the YouTube video. The really cool stuff will be the applications on top of `subvocal` though at least in my opinion.So this could take "noise canceling" to a whole new level if you could somehow "speak" across a loud bar or across the world…
2 more of the top 3 · 9 posts in this stretch
-
One of the coolest things I have recently seen but would love some details. It also got me thinking a about this recent Neuralink post.
-
I think this is incredibly cool and straight out of a sci-fi novel! I think this is phenomenal for accessibility and for hands-free computing in general. I also agree with other comments that the website should have more detail on how it works, such as what you included in the post body. Amazing work guys!
-
-
1
Subvocal shows silent speech recognition glasses on Hacker News
Aarush and Arnav from The Subvocal Company (YC F26) announced smart glasses that decode silently mouthed words to text using RF sensing, demonstrating the technology with a YC partner named Dave. The glasses work without requiring visible lip movement and process speech at natural speed.
“You can mouth words silently and discreetly and the glasses decode them to text at the speed of speech.”
— Aarush and Arnav, Subvocal Company founders
Also covered reported alongside — the timeline has no entry for these yet
-
first by HN Best, 8d ago · also HN Frontpage
What people are saying 7 voices from 1 site · best of 62 · verbatim
- What is the actual accuracy rate, and how does it degrade with different speech patterns, accents, or lip movements?
- Does inference work offline, or is cloud processing always required for decoding?
- What are the long-term health and RF exposure implications of wearing an RF chip close to the eyes and brain?
- Nov 4
-
I wonder if this could be implemented with low-cost ESP32s, using the raw IQ capture path which was discovered recently [0]. This setup is capable of receiving with over 40 Mhz of bandwidth in the 2.4Ghz band.[0]
- Nov 3
-
The military was being pitched a pattented tooth clip that detects bone resonance and allows silent communication. Sub-vocal seems to point RF at the face and emit radio signals at all times - is that safe? Have you done long-term testing that it doesn't increase cancer risk or cause hearing other issues?
-
This is absolute mind boggling tech. I didn't read the whole article but just wanted to know if glasses are a necessity of the tech or it can be just earphones or a strap around your neck or anything that can serve the same purposeI don't need glasses and I would never use them, so if there is any other way this can be done there you would have…
-
I am 100% your target demographic. My hands are completely fried and I'm scheduled for double carpal tunnel surgeries. I am also being forced to go back to work in an open landscape office where I can't use voice dictation. A product like this is literally what I have been looking for over the past few years.With that said I live in Europe... Do…
-
Have you considered how this might benefit those who are deaf, or speech impaired? Their lip/mouth movements might be totally different to folks with hearing - but if this can be fine-tuned on an individual's specific speech pattern, then I see a huge opportunity in the accessibility market for this too.I would also attempt to patent this…
- Nov 2
-
Theres a LOT of people working on this.. Cornell have been putting students on this task for about 5 years.. Lots of fun different ways of doing it. Interesting you went the radar from glasses route.. I wish you well. Yet to see a product apart from the odd lip reading ML project that makes it into a product..
-
This is very cool, but oddly, the site is TOO focused on the user problem being solved to the extent that it feels like snake oil. The issue is that this is contrary to people's experience of what is possible, so without discussing WHY you can do this, it looks like a scam or vaporware.