Introducing "Vakta 0.0.1" - an innovative and cutting-edge GUI based project developed for the Hackathon. Vakta is a state-of-the-art, fully voice-activated, and AI-enabled assistant, thoughtfully designed with the primary objective of empowering the visually impaired community. with adaptive learning technology. With an unwavering commitment to providing an inclusive and seamless user experience, Vakta will revolutionize the way blind individuals interact with technology.
"Vakta" is a Sanskrit word for "speaker." In our project, Vakta embodies this essence by becoming a compassionate and intelligent mentor for the visually impaired. Just as ancient speakers conveyed wisdom, Vakta aims to create a future of inclusivity and empowerment, bridging the gap between the visually impaired and knowledge.
-
Voice-activated Information (General Mode):
- Engage in voice-based conversations with OpenAI's LLM, powered by Eleven Lab's voice model and Langchain, utilizing memory to retain context throughout the interaction.
- Allows the individual to ask any question and get the answer through voice. Questions like “ What is the Capital of France? ” or “ Define Osmosis” are some examples.
- To conclude, this feature is simply our favorite ChatGPT in voice form.
-
Listen to your favorite book (Book Mode):
- Download the pdf of the requested book and play it like an audiobook.
- Control to stop and continue listening.
- Powered by various NLP algorithms and search techniques using Google Books API.
-
Know the weather around you (Weather Mode):
- Ask the AI about the weather of a particular city.
- AI responds with accurate temperature, humidity, and wind speed. For example, "What is the weather of Delhi?"
-
Stay Updated with the latest news (News Mode):
- Request for news headlines of a particular category or in general.
- AI responds with the latest headlines from categories like Sports, Technology, Business, etc.
-
Search and Listen to Music or Podcasts (YouTube Mode):
- Empowers the user to search any song or video from YouTube.
- Example: "Simon Sinek: How Great Leaders Inspire Action | TED Talk" or "Believer | Imagine Dragons."
-
Send a message to your contacts (Messaging Mode):
- Makes the user to send message easily to their contacts
- Example: "Send a message to 'Contact Name' "
- General Mode: The AI operates in the general chat mode (Feature 1) if no specific command is used.
- Book Mode: Enable book mode by saying "Download book <book_name>."
- Weather Mode: Enable weather mode by saying "weather <your_city_name>."
- News Mode: Enable news mode by saying "Search News ."
- YouTube Mode: Enable YouTube mode by saying "YouTube" and then provide the video title when prompted.
- Messaging Mode: Enabel this mode by saying "send message" and then provie the message you want to send
- Exit Program: Say "Exit Program" to close the application.
- GUI: KivyMD Framework (Python)
- Clone the repository:
git clone https://github.com/krrish-v/vakta.git && cd vakta - Install python if not installed on your system
- In Windows : python website
- In Linux :
sudo apt install python3
- Create a virtual environment
- If virtualenv is not installed :
pip install virtualenv && virtualenv env - Activate virtual environment
a. In Windows :env/Scripts/activate
b. In Linux or MacOS:source env/bin/activate
- Install dependencies:
pip install --upgrade pip -r requirements.txt pip install https://github.com/yt-dlp/yt-dlp/archive/master.tar.gz- Move to GUI directory and run main.py file
- Make the app available for Android and iOS devices since Kivy is a cross-platform Python framework.
- Enhance the features and make it a fully ready platform for voice-based learning.
- Release the app in the market, possibly on a subscription-based model.
-Currently working only on Linux OS system
React
Typescript
Django
Laravel
Facebook
Microsoft
Google
Alibaba
D3
Tencent