The project uses speech recognition in a custom function to do keyword detection. it then records the data and stores it in encrypted form use Text to Speech to store a transcription of the log.
- https://github.com/mozilla/DeepSpeech-examples/tree/r0.9/mic_vad_streaming (using the mozilla speech recognition code for streaming)
- https://www.slanglabs.in/blog/how-to-build-python-transcriber-using-mozilla-deepspeech (tutorial on using mozilla deepspeech)
- https://scgupta.medium.com/how-to-build-python-transcriber-using-mozilla-deepspeech-5485b8d234cf (medium article of the same tutorial ^)
- FOR FIXING THE AUDIO FILE EXPORT ISSUE: https://stackoverflow.com/questions/61499350/combine-audio-files-in-python