The aim of this push-to-talk (PTT) application is to enable a one-way, real time audio streaming from the student’s device to the lecturer’s PC utilising WebSockets from FastAPI. This serves as a proof of concept that it is possible to transmit raw audio data (that may require splitting up into chunks) via websockets and outputting it in a suitable form to be played on a loudspeaker on the lecturer side. At any given time, only one student can send his/her voice to the lecturer – the concurrency is managed by our application.
Note: You will need to install Python 3.10. If not, you will face errors running our code.
python3 -m venv venv
source venv/bin/activate
pip install -r requirements.txt
fastapi dev server.py
If you are unable to run the fastapi command, it is most likely that you have an existing venv. Delete the venv folder and reinstall the packages.
Our code utilizes the pyaudio library. To use it, you have to install the portaudio library.
for windows:
sudo apt-get install portaudio19-dev
for mac:
brew install portaudio
pip freeze > requirements.txt everytime there is updates to the requirements
fastapi dev server.py