Skip to content

KabraKrishna/SignLanguageToText

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

5 Commits
 
 
 
 
 
 
 
 
 
 

Repository files navigation

Sign Language To Speach / Text

A machine learning and computer vision based approach to convert Sign language to Speach / Text

Demo link while testing

https://youtu.be/f-ra7_zMwLg

Dataset for train and test link

https://mavsuta-my.sharepoint.com/:f:/r/personal/vxc0340_mavs_uta_edu/Documents/dataset?csf=1&web=1&e=4XjNXj

Link for the model.h5 file

download and save the model.h5 file from the following link in the same directory before running the code https://mavsuta-my.sharepoint.com/:u:/r/personal/vxc0340_mavs_uta_edu/Documents/model.h5?csf=1&web=1&e=Y2ynAj

Description:

Our aim for developing this research is that it may be deployed as a mobile application in the future for persons who cannot speak, acting as a transulator. since these are the early stages of implementation, yet these are helpful concepts So far, we've improved our accuracy while training. When we tested the program on a simple background, it worked well, but when there is a lot of noise, such as background objects, it takes time and some prediction, and we also have to consider that we have to distinguish 26 different signs, rather than simply 1 or 2 faces in a face recognition thats so challenging.

Dependencies

  • python v3.10.x
  • gTTS v2.2.4
  • keras v2.8.0
  • matplotlib v3.5.3
  • nltk v3.7
  • numpy v1.22.4
  • opencv_python_headless v4.6.0.66
  • pyttsx3 v2.90
  • scikit_learn v1.1.3
  • tensorflow v2.8.2
  • textblob v0.17.1

Steps to run the model

  • before running the code unzip all the compressed files.

  • Install python v3.10.x if not present already.

  • Install dependencies from requirements.txt file using command below:

    pip install requirements.txt

  • Run the jupyter notebook app.ipynb and run it to test the model.

About

No description, website, or topics provided.

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Contributors 2

  •  
  •