A python function to extract the LPC coefficients of recorded speech and replace them with obfuscating coefficients. Based on this paper: https://www.fxpal.com/publications/audio-privacy-reducing-speech-intelligibility-while-preserving-environmental-sounds.pdf