Lecture: Creating Jarvis and WhatsApp Bot using Python 🦾📞

Jarvis - Virtual Assistant

Tools and Libraries Used

1. speech_recognition: For speech recognition
2. pyttsx3: For text-to-speech conversion
3. webbrowser: For web actions (e.g., opening Google, YouTube)

Steps and Code Highlights

1. Initialize Recognizer and TTS Engine:

   recognizer = sr.Recognizer()
   engine = pyttsx3.init()
   

2. Define Speak Function: Uses TTS to speak text aloud.

   def speak(text):
       engine.say(text)
       engine.runAndWait()
   

3. Listening for Wake Word: Listens for 'Jarvis'.

   while True:
       with sr.Microphone() as source:
           print('Listening...')
           audio = recognizer.listen(source)
           try:
               command = recognizer.recognize_google(audio).lower()
               if 'jarvis' in command:
                   speak('Yes, how can I help you?')
                   # further code to recognize commands
           except sr.UnknownValueError:
               speak('Sorry, I did not get that.')
   

WhatsApp Bot

Tools and Libraries Used

1. pyautogui: For GUI automation
2. Pillow (PIL): For image processing
3. pytesseract: For OCR (Optical Character Recognition)
4. OpenAI GPT-3 API: For generating responses

Example Setup

1. Install Libraries:

   pip install pyautogui Pillow pytesseract openai
   

2. Dependency Setup (on Windows):

   • Install Tesseract OCR from Tesseract GitHub
   • Add Tesseract to system PATH

Steps and Code Highlights

1. Initialize PyAutoGUI for Screenshot:

   import pyautogui
   screenshot = pyautogui.screenshot()
   

2. Read Chat and Generate Response:

   import pytesseract
   import openai
   pytesseract.pytesseract.tesseract_cmd = r'<path-to-tesseract-executable>'
   openai.api_key = 'YOUR_OPENAI_API_KEY'
   
   # Capture, OCR and ask GPT-3
   chat_text = pytesseract.image_to_string(screenshot)
   response = openai.Completion.create(
       engine='text-davinci-003',
       prompt=chat_text,
       max_tokens=150
   )
   reply = response.choices[0].text.strip()
   

3. Automate Reply in WhatsApp:

   pyautogui.typewrite(reply)
   pyautogui.press('enter')