Drishya Sahayak combines advanced computer vision with intuitive audio feedback to give you complete independence.
Advanced computer vision meets intuitive audial feedback.
Identify people, text, and obstacles instantly. Our custom YOLO architecture processes visual data in milliseconds.
Natural language interaction allows you to ask "What's in front of me?" or "Read this menu" effortlessly.
All processing happens locally. No video data is ever sent to the cloud, ensuring your absolute privacy.
Built on a custom raspberry pi architecture with edge-optimized AI models. Powerful vision processing without the need for an internet connection.
Our custom YOLOv12-Nano architecture runs at 30fps directly on device. It identifies specific objects, people, and obstacles in your path with 98% accuracy.
Just ask. The integrated LLM allows natural conversation about your surroundings. "Read this sign for me" or "Is the crosswalk light green?"
Invest in independence. One-time hardware cost with optional premium support.
Everything you need to get started.
The complete experience with accessories.
For peace of mind and constant care.
It uses integrated cameras to capture the environment, processes the visual data using on-board AI, and converts it into descriptive audio feedback transmitted via bone-conduction headphones.
No. Core functions like obstacle detection, people recognition, and basic text reading happen entirely offline on the device for speed and privacy. Only advanced conversational features require a connection if enabled.
The device is designed to last a full day of typical use (8-12 hours). With the power bank included in the Independence Bundle, you can extend this to 20+ hours.
Yes, all devices come with a standard 1-year limited warranty covering manufacturing defects.
We are dedicated to removing barriers for the visually impaired through innovative, affordable, and privacy-focused technology.
Drishya Sahayak started as a research project to solve a simple problem: digital accessibility tools were either too expensive or relied too heavily on cloud connectivity.
Our team of engineers and designers spent 2 years iterating on hardware that balances power, battery life, and comfort. The result is a device that feels less like a gadget and more like an extension of your senses.
Initial research into efficient edge-AI models and hardware selection.
Deployed 50 prototype units to local communities for real-world feedback.
Finalized hardware design and optimized YOLO models for the NPU.
Official release of Drishya Sahayak Gen 1.
Lead Developer & Founder
Have questions? We are here to help. Reach out to our team for support, sales, or partnership inquiries.
Drishya Sahayak Labs
Kannur, Kerala
India
Precision engineered hardware meets state-of-the-art computer vision models. Lightweight, durable, and ready for the real world.
Powered by the Raspberry Pi 5, delivering desktop-class performance in a portable form factor for instantaneous object detection.
High-definition wide-angle camera module optimized for low-light conditions and rapid frame capture.
Subtle vibration motors provide directional cues for navigation without interrupting audio feeds.
5G capable with fallback to LTE/4G ensures you are never disconnected from maps and updates.
| Processor | Quad-core Arm Cortex-A76 @ 2.4GHz |
|---|---|
| Memory | 8GB LPDDR4X-4267 SDRAM |
| Camera | 12MP Wide Angle, f/1.8 aperture |
| Battery | 5000mAh Lithium-Polymer (8h active use) |
| Connectivity | WiFi 6, Bluetooth 5.0, GPS/GLONASS |
| Sensors | Lidar (Optional), Accelerometer, Gyroscope |
| Weight | 180g (Device unit) |
Latest firmware, drivers, and manuals.
Release Date: Feb 15, 2025 | Size: 1.4GB
English | 2.5MB
Specializing in YOLO architectures, Object Detection, and Semantic Segmentation on edge devices.
Full-stack development with a focus on modern, responsive, and accessible UI/UX.
Your privacy is paramount. This policy outlines how we handle data.
We do not collect video or audio recordings from the device. All processing is done locally.
We may collect anonymous usage statistics (e.g., battery cycles, error logs) only if you opt-in.
By using Drishya Sahayak, you agree to these terms.
The device is an aid, not a replacement for a cane, dog, or human guide. Always exercise caution.
We provide a 1-year limited warranty for hardware defects.
Comprehensive guides and API references for developers.
Drishya Sahayak is an open-source AI vision assistant designed to run on edge devices like the Raspberry Pi 5. It leverages optimized YOLOv12-Nano models for real-time object detection.
Clone the repository and install dependencies using our setup script.
git clone https://github.com/dhyandevp/Drishya_Sahayak.git cd Drishya_Sahayak chmod +x setup.sh ./setup.sh
Once installed, you can start the vision system with the following command:
python3 main.py --mode=detection
The system will initialize the camera and begin audio feedback immediately.
The system is built on a modular architecture separating vision processing, audio synthesis, and hardware IO.
Interact with the core system using our Python bindings.
import drishya
# Initialize
assistant = drishya.Core()
# Start Detection
assistant.start_detection(model="yolo-n")
# Get Status callback
@assistant.on_event
def handle_event(event):
print(f"Detected: {event.label}")