Handsign Recognition for Hindi Gujarati and English with VoiceOvers

This project features a trained model utilizing computer vision to detect hand signs representing English alphabets, Hindi varnmala, and numbers. Each prediction is accompanied by a voiceover that announces the identified character or number.

Reference Video

https://www.youtube.com/watch?v=4Xgwo9dR1k4&t=3s

Images

Installation

1. Clone the repository:

   git clone (https://github.com/VedantR3907/Handsign-recognition-for-HIN-GUJ-ENG-with-voiceovers.git)

2. Install requirements.txt

pip install -r requirements.txt

Dataset

1. Hindi Dataset

https://www.kaggle.com/datasets/vedantrajpurohit/hindi-varnmala-hand-signs-dataset

2. English Dataset

https://www.kaggle.com/datasets/vedantrajpurohit/american-sign-language-dataset

3. Numbers Dataset

https://www.kaggle.com/datasets/vedantrajpurohit/numers-hand-sign-dataset

Usage

To run Project, execute the following command:

cd interface
python main.py

Ensure you have Python install and all dependencies from requirements.txt are installed beforehand.