The idea is to create an app, which have two parts speaker(host) and listeners. So, whenever the listener needs to ask doubt or interact with host it can raise hand in the app, or use hand gesture to ask doubt.
The app will also serve, the person who are disabled, can’t interact and miss important session, so we can use hand gesture-to-speech, wherein, a speech-impaired person could convey their messages using gestures on their smartphone’s front camera and it’d get converted to voice in real-time.
Moreover the hearing-impaired person can see the text spoken on their mobile screen, using speech to text converter.
It also happens that many a time there is language gap, so we can give a choice either to do translation of words from host language to listener language or to do transliteration, example Hinglish, as most time hindi is difficult to read, but hinglish can be easily read and understand.
- Make a bridge to close the communication gap among us.
- The regular listeners can raise hand button in app, so that they can use their mobile’s microphone to talk too host, and then host can * easily communicate with them.
- Since, lip reading has accuracy of only 50% we will use gesture-to-speech with accuracy of 80%+.
- Use of speech to text converter for hearing-impaired person.
- Use of translation and transliteration, to convert host language to listeners language and vice versa.
These instructions will get you a copy of the project up and running on your local machine for development and testing purposes. See deployment for notes on how to deploy the project on a live system.
- Any operating system (i.e. Linux, Windows, MacOS X)
- A little knowledge of Dart and Flutter. Don't worry if you are new to it, you just need knack to learn.
- Any IDE with Flutter SDK installed (i.e. VSCode, Android Studio, etc)
- Any IDE for Python (i.e. VSCode, Pycharm etc)
Setting up the project
- Download Flutter from the below link
- Install Python
- Installing Pipenv
We are using Pipenv for managing project development in a virtual environment and it is also the package manager here.
- Look over the project and read the goals to be met for project.
- Research more about implementation of Machine learning model, for transliteration.
- Research more about implementation of Machine learning model, for hand gesture-to-speech
- Setting up goals.
- Wirting testcases of for transliteration.
- Testing of transliteration and hand gesture-to-speech
- Completing the project.
- If time permits, we will setup AR subtitles, such that re can get live subtitle in prefered version.
Please raise a feature request or issue before sending PR for the same.
Follow the below guidelines for proper coding practices:
- Always create a new branch for your changes and make PR from it ONLY.
- Write neat code with proper comments.
- Follow PEP8 coding style.
- Write descriptive commit messages. Please read this for more information.
- Write detailed PR messages and include
fixes #ISSUE_NUMBERit if closes an issue, otherwise use
concerns #ISSUE_NUMBERto tag the related issues. Please refer here for more PR guidelines.
- It is recommended to have a single commit for a task.
- Use DRY principles to create maintainable code.
You can contribute in several ways. If you know how to code or are a designer, you are welcome to contribute using pull requests. You can also contribute by opening issues about defects and things that could be improved or request entirely new features that you think would help others. Join the Slack Communication Channel
- Ajinkya Taranekar
- Anand Kumar
- Mahak Mandelecha
- Ritik Nandwal
This project is licensed under the Codefoster and SGSITS.