panorama
Real-time scene recognition
A short spoken summary of the space ahead: layout, furniture, free paths, open doors, queues and who is nearby.
visibility AI-Powered Vision
Your world, spoken aloud.
Sprhava spectacles see, read and describe — so people who are blind or have low vision move through the day hands-free.
<1sScene narration
49gAll-day frame
20+Voice languages
0Screens to touch
- record_voice_over Natural-speech scene description
- chrome_reader_mode Instant text-to-speech reading
- sensors Head-height obstacle alerts
- wifi_off Offline on-device vision engine
- accessibility_new Designed with low-vision users
What Sprhava is
Unobtrusive access to the visual world
Sprhava is a wearable AI vision aid built into an ordinary-looking pair of spectacles. A discreet camera on the frame captures what is in front of the wearer, on-board computer vision interprets it, and open-ear speakers explain it in clear, conversational speech.
There is no app to stare at, no handset to hold and no headphone blocking the ear canal. The wearer simply asks — “What’s in front of me?”, “Read this”, “Find the exit” — and listens.
- Describes rooms, streets, people and products on request
- Reads menus, labels, signage, mail and screens aloud
- Warns about kerbs, doors, poles and head-height hazards
- Works with a long cane, guide dog or existing mobility training
door_front Doorway 2.4 m ahead
text_fields “Platform 4 — Gate B”
directions_walk Person approaching
Real-time scene recognition · on-frame camera · open-ear audio
Core capabilities
Six everyday tasks, one pair of glasses
Every function was prioritised from interviews with blind and low-vision testers, orientation & mobility specialists and low-vision clinicians — not from a feature checklist.
menu_book
Text-to-speech reading
Point and listen. Printed menus, handwritten notes, medication boxes, receipts and signs are read aloud in natural voice.
inventory_2
Object & product recognition
Distinguishes similar packages, identifies currency, finds a dropped key, confirms the right tin on the shelf.
warning
Obstacle & hazard alerts
Discrete tones and short phrases flag kerbs, steps, glass doors, poles and overhanging objects a cane can miss.
face
Face & contact prompts
Optional, consent-based recognition quietly names saved family, friends or colleagues as they come into view.
explore
Orientation assistance
Landmark cues, room-exit guidance and directional prompts that support — never replace — formal mobility training.
How it works
From light to language in under a second
A four-stage pipeline keeps the experience fast, private and usable even when the network is not.
Capture
A low-distortion camera on the temple records the wearer’s natural field of view at their head angle.
Understand
On-device neural networks segment depth, detect objects and lift text within milliseconds.
Prioritise
Context logic decides what actually matters right now, so the wearer is informed, not flooded.
Speak
Open-ear directional audio delivers short phrases while ambient hearing stays fully intact.
Engineering
Computer vision built for the real world
Pavements are uneven, lighting is poor and coverage drops indoors. Sprhava is engineered for those conditions rather than for a demo table.
memory
Edge-first inference
Detection, depth and OCR models run locally. Frames are processed and discarded on the device; nothing is stored by default.
nights_stay
Low-light robustness
Wide-aperture optics with adaptive exposure keep recognition stable at dusk, in corridors and under mixed artificial light.
hearing
Bone-safe open audio
Directional open-ear drivers preserve traffic noise and speech cues — critical for safe independent travel.
shield_lock
Privacy by architecture
Face profiles stay encrypted on the device, cloud calls are opt-in and every session can be wiped with one voice command.
battery_charging_full
All-day power design
Split battery in both temples with duty-cycled sensors delivers a full outing, plus fast top-up in the charging case.
language
Multilingual voice packs
More than twenty speech languages with adjustable verbosity, speed and terminology for clinical or consumer deployments.
Who it serves
Built for daily life, clinics and programmes
Sprhava is deployed by individuals and by the organisations that support them.
home
Independent living
Cooking, sorting post, choosing clothes and checking appliances without waiting for sighted help.
work
Work & study
Reading handouts, whiteboards, name badges and printed documents during meetings and classes.
local_hospital
Low-vision clinics
A trial-ready device for macular degeneration, glaucoma, diabetic retinopathy and retinitis pigmentosa pathways.
public
Distributors & programmes
Assistive-technology resellers, veterans’ services and accessibility funds with multi-unit rollouts and training needs.
Experience & expertise
Co-designed with the people who wear it
Sprhava’s product decisions come from repeated field testing with blind and low-vision participants across homes, transit hubs, supermarkets, offices and hospitals. Every release is reviewed against three questions: is it safe, is it understandable, and does it reduce effort?
Our team combines computer-vision research, wearable hardware engineering and accessibility practice, and we publish plain-language documentation so clinicians, trainers and procurement teams can evaluate the device honestly — including its limits.
- Iterative testing with users who are blind or have low vision
- Input from orientation & mobility and rehabilitation specialists
- Transparent statements about accuracy, coverage and constraints
- Documentation, training and after-sales support for partners
“The value isn’t the camera — it’s deciding what not to say. Sprhava’s alerts arrive when I need them and stay quiet when I don’t.”
“For our clinic, the deciding factors were open-ear audio, offline reading and clear guidance that it complements cane training rather than replacing it.”
“Reading a medicine box by myself, at the counter, in seconds. That is the part that changed my week.”
At a glance
Sprhava spectacles — key specifications
Indicative figures for evaluation. Final configuration varies by region, lens option and partner firmware.
| Device type | Wearable AI vision aid / electronic orientation and mobility aid |
|---|---|
| Frame weight | Approx. 49 g, balanced temple-weight distribution |
| Camera | Wide-field low-distortion module, adaptive low-light exposure |
| Audio | Dual open-ear directional speakers, ear canal left uncovered |
| Control | Voice commands plus single tactile button — fully hands-free operation |
| Processing | On-device vision engine; optional secure cloud model for richer descriptions |
| Response time | Sub-second obstacle alerts and scene narration |
| Languages | 20+ speech languages, adjustable verbosity and speaking rate |
| Lenses | Prescription-ready; clip-on module variant for existing spectacles |
| Charging | USB-C charging case with fast top-up between outings |
Questions answered
Frequently asked questions
Practical answers for users, clinicians and international buyers evaluating visual assistive technology devices.
The glasses capture what is in front of the wearer, interpret it with computer vision, and speak the result through open-ear audio. Users hear scene descriptions, printed text, product labels, saved faces and obstacle warnings — without holding a phone or looking at a screen.
Yes for core functions. Obstacle detection, text reading and object recognition run on the on-device engine, so they keep working in basements, lifts, rural areas and on flights. Conversational, highly detailed scene descriptions use a secure cloud model when a network is available.
The frame accepts prescription and contrast-filter lenses. A clip-on module version is available for users who prefer to keep their current spectacles, which is common for people with macular degeneration or light sensitivity.
No. Sprhava is designed to work alongside established mobility aids and training. It adds head-height hazard awareness, environmental context and reading ability; it does not detect everything and should never be used as a sole navigation tool.
Frames are analysed and discarded in real time, no continuous recording is stored by default, face profiles are encrypted locally and consent-based, and cloud processing is opt-in. A single voice command clears session data.
Yes. We work with assistive-technology distributors, low-vision clinics, rehabilitation centres and accessibility programmes, offering multi-language voice packs, branded firmware options, clinician training material and regional after-sales support.
Bring AI-powered vision to your market
Talk to the Sprhava team about clinical trials, distribution, OEM configurations and accessibility programmes across your region.