September – Windows Assistant

September is an open-source Windows personal assistant built-in python. Read How to Setup? for additional information on setting up the application.

? Install

• Download Zip File

? Features

• Minimalistic UI
• Lightweight
• Customizable Wake Word
• Type/Speak commands
• Search Wiki, Google and YouTube
• Open Windows Apps
• Mathematical Calculations
• Human-like Conversations

? Overview

? Code Walkthrough

• main.py contains the functions for initializing Tkinter-based windows and converting speech to text.

  • The listening process and tkinter frame refresh process run concurrently using threads.
  • Each window is started as a separate thread for them to work parallelly.
  • Start and Stop Listening sounds are played using the playsound module.
  • The input from the user is taken either as a text from the command entry box or as audio from mic button
    
  • The input is converted to text using SpeechRecognition module and passed to processtext function of processtext.py

• processtext.py contains the function in which the command processing happens.

  • The input obtained from the user in the main function is passed to the processtext function. It has an if-else ladder to search for specific keywords in the input. If the query does not match with the keywords found in the ladder, it is passed to wolfresponse.py
  • Functions like opening apps, searching web happens here.

• wikiresponse.py and wolfresponse.py are the places where the respective APIs are accessed to process the output of the unmatched queries from processtext.py

• texttospeech.py uses the pyttsx3 module to convert speech to text.

  • After processing the query, text to speech function is called.
  • Esc Key is added as a keybind to stop text to speech for one iteration by calling texttospeech_stop function.

• config_data.json contains the values that are displayed in the September app settings in JSON format. It is used for storing windows application paths, wake word and the API key.

• requirements.txt contains the list of all the modules used in this program.

? Dependencies

• requirements.txt contains all the python modules required by the program.
• All the assets used by the program are present in assets folder. The application won’t function as intended without these assets.
• Built in Python 3.10.1. Use python 3.0 or greater versions for a better experience.
• Requires active internet connection.
• Make sure that your antivirus doesn’t block the program from uploading or downloading audio stream data. (Disable antivirus :D)
• Additionally, the Pyaudio module must be installed by following the below instructions.

Install Pyaudio

• Find your Python version using in your terminal

python --version

• Find the appropriate .whl (wheel) file at Pythonlibs and download it.
• Go to the folder where it is downloaded and install the .whl file using pip,
  For example, if you download the wheel file for Python 3.7 64-bit, your pip command would be,

pip install PyAudio-0.2.11-cp37-cp37m-win_amd64.whl

• The wheel file for Python 3.10 64-bit is present in dependencies/PyAudio-0.2.11-cp310-cp310-win_amd64.whl
• Visit for more information

? License

MIT License Copyright (c) 2021 Nithin Balaji

GitHub

View Github