Skip to content
Projects

dindin0497/SeeIt has 1,539 GitHub stars and 192 forks. GitHot measured 0 stars in 24 hours. Last checked 2026-11-08 04:18 UTC. Last push 2025-09-18.

How these counts are measured

https://github.com/user-attachments/assets/c31b4838-6ffb-413a-a6cd-38d3aea3a1af

What you can do

  • ⚡ Partial results (near word-by-word feel)
  • 🎯 Candidate ranking (choose the best sign/video)
  • ⚡ Real-time detection (CameraX + ML Kit/TFLite)
  • 🔕 Debounced announcements (avoid spam)

Measured star growth

in 24h
0
in 7 days
Update delayed
in 15 days
Update delayed
in 1 month
Update delayed

Net star changes over overlapping periods, not daily totals. Missing measurements are left blank. Stars measure attention, not software quality.

README.md

Original

SeeIt

Inclusive Android app with two core capabilities:

  1. ✋ ASL Mode – Speech → Text → ASL Video
  2. 👁️ Object Detection Mode – Camera → Detect → Speak

✋ 1. ASL Mode

What It Does

  • 🎙️ Listens from mic → converts speech to text
  • 🔤 Maps text to ASL signs
  • ▶️ Fetches ASL videos from the web and plays them in-app

Why It Matters for the Deaf Community

  • 🤟 Many Deaf and Hard of Hearing (HoH) individuals use ASL as their primary language, not written/spoken English.
  • 🗣️ This feature bridges the gap by translating spoken words into sign language videos, enabling more natural understanding.
  • 🧑‍🤝‍🧑 It promotes inclusion in everyday scenarios such as conversations, announcements, and meetings.
  • 🌍 Designed with accessibility in mind: large buttons and high contrast.

Key Features

  • ⚡ Partial results (near word-by-word feel)
  • 🎯 Candidate ranking (choose the best sign/video)

Demo

https://github.com/user-attachments/assets/c31b4838-6ffb-413a-a6cd-38d3aea3a1af

👁️ 2. Object Detection Mode

What It Does

  • 📷 Uses camera to detect objects in real time
  • 🗣️ Speaks object names via Text-to-Speech (TTS)
  • 🏷️ Optionally shows detected labels on screen

Why It Matters for People with Visual Impairments

  • 👓 Helps users with eye problems or low vision identify objects in their surroundings.
  • 🗣️ Provides audio feedback so users can hear what’s in front of them without needing to see it.
  • 🧭 Improves independence by enabling safer navigation and better awareness of the environment.
  • 🖼️ Optional on-screen labels can support users with partial vision, combining text with speech.

Key Features

  • ⚡ Real-time detection (CameraX + ML Kit/TFLite)
  • 🔕 Debounced announcements (avoid spam)
  • 📊 Confidence threshold + label coalescing
  • 📦 Works offline using local models

Demo

https://github.com/user-attachments/assets/88cdca3c-0b67-4257-b9bf-bb1751122672

Explore related projects