📷
AvailableUnderstand the scene
Use the shared camera core and local vision infrastructure to help interpret what is in front of the user, with local fallback when cloud intelligence is not used.
BASIRAAI vision companionBASIRA
AI vision that listens, understands and assists.
A blind-first Android companion that combines on-device assistance with optional cloud AI — and a roadmap toward wearable vision, navigation and spatial memory.
Local-firstCloud only when neededNo continuous cloud videoNo ads
Built today on the smartphone. Developing toward a natural wearable visual companion.
BASIRA TODAY
A real product foundation — not just a concept.
The current Android production release combines a stable camera core, local vision infrastructure, BASIRA AI, voice interaction, privacy boundaries, subscriptions and Google Play distribution.
📄
AvailableRead and access information
Capture text and documents, process them through the reading pipeline and deliver information through speech.
🎙️
AvailableTalk naturally
Tap-to-talk, hold-to-talk and a user-started continuous microphone session support a more natural relationship with BASIRA AI.
🧠
AvailableBASIRA AI
Firebase AI / Gemini adds optional cloud intelligence for richer questions and visual understanding while reusing the same camera core.
🛡️
AvailablePrivacy boundaries
Cloud camera use is based on still images when needed. Microphone use is user-initiated and separately consented. BASIRA does not stream continuous camera video to the cloud.
♿
AvailableBlind-first interaction
Few controls, clear speech, large targets, TalkBack-aware interaction and a goal that core actions remain usable without relying on visual UI.
TECHNICAL FOUNDATION
Local first. Cloud when it adds value.
The architecture is intentionally split so fast, private and essential assistance can remain on-device while advanced cloud reasoning is optional.
LOCAL
On the phone
- Single shared CameraController and camera lifecycle.
- Local vision stack including TensorFlow Lite / Task Vision, ML Kit, MediaPipe and related models.
- Local fallback for independence, latency and privacy.
- Camera is activity-owned and does not remain a camera foreground service in the background.
BASIRA AI
Optional cloud intelligence
- Firebase / Firebase AI Logic / Gemini.
- Still image capture when a cloud visual request requires it.
- App Check, Play Integrity and cloud entitlement controls.
- User-controlled microphone modes with foreground service only for an active continuous audio session.
SUSTAINABLE ACCESS
Keep local help free. Use support to fund the intelligence layer.
Our current model keeps local functions free and uses a monthly BASIRA AI subscription for cloud costs. Supporter, Patron and Sponsor levels are designed to provide the same BASIRA AI capabilities while allowing those who can contribute more to help the project continue.
✓
Same BASIRA AI capabilities across support tiers.
✓
No advertising model.
✓
A future goal is to use community support to help users who cannot afford cloud AI.
OUR DEVELOPMENT VISION
From camera assistant to spatial companion.
These capabilities are our development roadmap. They are shown as planned or R&D — not as completed safety guarantees.
Wearable camera support
Move from always holding the phone toward a small camera on glasses, head or chest, with open-ear audio and the phone acting as the compute and connectivity hub.
Object Finder + tracking
Ask for keys, a chair or another everyday object, scan naturally, track detections and guide the user toward what was found.
Virtual safety corridor
Combine detection, tracking, segmentation/depth, motion and risk scoring to identify obstacles entering the user’s likely path.
Depth and distance
Research ARCore Depth, monocular depth, motion/parallax and sensor fusion to improve practical distance estimates on ordinary phones.
Spatial & place memory
Help BASIRA remember meaningful entrances, rooms, objects and landmarks so future questions can refer to a place instead of a single image.
On-device multimodal AI
Explore smaller local visual-language models for lower latency, lower cloud cost and greater privacy, with cloud fallback for harder questions.
PlannedGOOGLE MAPS NAVIGATION
Where to go — plus what is around you.
We plan to integrate walking navigation powered by Google Maps with BASIRA’s local perception. Google routing can guide the destination and turns; BASIRA can focus on nearby visual context, obstacles and the questions the user asks.
1
Walking route and turn-by-turn voice guidance inside BASIRA.
2
Save frequent destinations, places and useful routes.
3
Keep local scene perception independent from cloud route guidance where possible.
4
Design navigation for audio-first interaction rather than a visually dense map screen.
Walking-route data and computer vision can be incomplete. BASIRA is an assistive tool and must not be presented as a replacement for a white cane, guide dog, orientation & mobility skills, or the user’s own judgment.
DEVELOPMENT ROADMAP
A focused path after the first release.
We do not want to open twenty features at once. The roadmap prioritizes real blind-user feedback before heavier spatial research.
P0
Maintain the public Google Play release, privacy, Data Safety, store listing and production monitoring.
P1
Launch the BASIRA pilot with blind users, associations, feedback and respectful crash analytics.
P2
Improve daily conversation, response length, context, “What is ahead?” and reading flows.
P3–P4
Object Finder + tracking, then safety corridor, motion, depth and risk research.
P5
Spatial memory — the step from camera assistant toward a true assistive spatial companion.
BUILD BASIRA WITH US
We do not want to guess what blind users need.
The next stage is co-design: real use, real feedback, and priorities chosen with the people BASIRA is meant to assist.
Blind & low-vision users
Test BASIRA, tell us what helps, what interrupts you, what is confusing and what problem you want solved next.
Associations & centers
Host a small pilot group, help us reach diverse users and turn recurring feedback into measurable priorities.
Accessibility & research partners
Collaborate on usability, orientation & mobility, computer vision, wearable interaction and responsible evaluation.
Supporters & sponsors
Help cover cloud AI, test devices, user research and development so the project can keep improving.
Help shape the next BASIRA release.
Join the pilot, propose a feature, connect us with a blind association, or discuss a technical and research partnership.
BASIRA GLOBAL COLLABORATION
BASIRA Global Collaboration
Join as a technical or research collaborator, a tester or an institutional partner. Remote contributions help us build with blind and low-vision users.
PRIVACY & TRUST
Assistive technology should respect the person using it.
BASIRA separates local and cloud processing, does not require continuous cloud video, and uses explicit boundaries around camera and microphone access.
Contact
511 Technology Company
support@511.jo
eng.ismael@511.jo
+962 78 700 4585
+962 78 815 6156
BASIRA is available on Google Play: https://play.google.com/store/apps/details?id=com.basira.basira