Skip to content

yassin448/WATI

Repository files navigation

WATI Assistant

WATI Assistant (Wise Ai for Textual Interaction) is an AI-powered assistant with a variety of features, including conversational abilities, arithmetic and equation solving, weather updates, YouTube music playback, and more. This assistant uses a combination of pre-trained models, semantic similarity, and fallback mechanisms to provide accurate and helpful responses.

Features

  • Conversational AI: Uses a pre-trained language model for generating responses.
  • Arithmetic and Equation Solving: Solves arithmetic expressions and algebraic equations.
  • Weather Updates: Fetches weather information for specified locations.
  • YouTube Music Playback: Plays, pauses, and stops music from YouTube.
  • Sentiment Analysis: Analyzes the sentiment of user queries and provides emotionally responsive answers.
  • Feedback Mechanism: Allows users to provide feedback and update answers in the dataset.
  • Simulated Typing Effect: Displays responses with a simulated typing effect for a more human-like interaction.
  • User History Navigation: Navigate through user input history with the up arrow key.
  • Dynamic Chat Interface: Clear chat history, exit the application, and send queries through a user-friendly interface.

Prerequisites

Ensure you have the following software installed on your system:

  • Python 3.7 or higher
  • pip (Python package installer)

Required Libraries

The following Python libraries are required for the WATI Assistant to function:

  • tkinter
  • pandas
  • spacy
  • re
  • os
  • math
  • time
  • random
  • sympy
  • concurrent.futures
  • csv
  • pytube
  • youtube_search
  • yt_dlp
  • ffmpeg
  • pyaudio
  • requests
  • threading
  • textblob
  • datetime

You can install these libraries using pip:

pip install pandas spacy sympy pytube youtube-search yt-dlp pyaudio requests textblob
python -m spacy download en_core_web_sm

Running the Application

  1. Clone the Repository (if applicable) or download the script files to your local machine.
  2. Navigate to the Directory containing the script files.
  3. Run the Script:
python3 Ai.py

Usage

Once the application is running, you can interact with WATI Assistant through the provided user interface. Type your questions or commands into the input field and press Enter or click the Send button.

Available Commands:

  • Arithmetic and Equations: Ask math-related questions or provide equations to solve.
    • Example: Solve 2 + 2, Solve x + 2 = 5
  • Weather Updates: Ask for weather information.
    • Example: What is the weather in New York?
  • YouTube Music Playback: Play, pause, resume, or stop music from YouTube.
    • Example: Play music: Despacito, Stop music
  • Clear Chat: Type clear chat to clear the chat history.
    • Aditional command: Type --no dataset at the end of your question to answer without the default dataset.
  • Exit Application: Type exit to close the application.

Feedback Mechanism

(Currently not usable in this version) If the assistant provides an incorrect answer, you can correct it by providing feedback:

  1. Type correct <your query> to initiate the correction process.
  2. Provide the corrected answer when prompted.

Navigation

Use the up arrow key to navigate through your input history.

###Other commands 1.correct the question (corrects the text after the the word correct( 2.tell time (tells you the current date and time)

arguments

  1. --use-core (uses the finetuned version of wati)
  2. --use-groq (uses fallbackai) 3.translate to (language):uses the finetuned wati to translate the question into the desired languages.The currently tested languages are(italian, Spanish,german,french)

Notes

1.this version is still in beta and you may experience incorrect or totally broken answer so please try to use the arguments correctly and follow the guide in this readme for it to work properly 2.to use groq or (the fallbackai) you need to obtain and api key by signing up on this website:https://console.groq.com/playground and after signing in you need to go to the api key tab and generate a new one and open the Aidata file and put it in get api key 3.you will find the finetuned Wati version on the release tab which you need to install and add to the same directory of the ai inside a folder named "WATI_core_dependencies" for all the features to work

License

This project is licensed under the Apache-2.0 License.

Acknowledgements