A Voice Assistant in python which scrapes the web using BeautifulSoup to answer user's questions related to COVID-19. Website being scraped for data: www.worldometers.info/coronavirus/#countries
The following python packages need to be installed on your machine/environment in order to run the program:
- beautifulsoup4
- requests
- lxml
- pyttsx3
- SpeechRecognition
- pyaudio
All the above mentioned packages can be simply installed by pip install <package_name> where package names are case sensitive. However, to install pyaudio on OSX you first need to install homebrew and then run brew install portaudio before the pip command.
Once all packages are installed, from project directory: python3 voice_assistant.py
When the program is started, Listening... will be printed on the console once the assistant has fetched the data and is ready to answer questions. The user should start speaking only once Listening... is displayed (preferebly after a slight pause because some delay is possible). The program will print what the user spoke and the assistant's answer on the console. Once the assistant answers a question, it will recalibrate after which Listening... will be displayed again which is when the user can ask another question.
- How many COVID cases are there worldwide?
- How many people have died because of coronavirus in India?
- How many people have recovered from COVID in the UK?
- How many COVID deaths have been conducted in Australia?
- How many active cases are there in the USA?
- How many cases are critical in France?
- To end the program say 'stop' or 'exit'
- To update the data while program is running, say 'update' or 'update data'
- Implement GUI with a speak button which the user can press to talk
- Add feature to locate nearest COVID Clinic to user
- Convert into a Web App using recently acquired skills