Integrate Piper's Text-to-Speech (TTS) Feature into Perplexica

[microsuperman]

I think integrating something like piper to do the TTS. This could allow user to pick the voices they would like. Well also keeping things local as piper doesn't have a large footprint as it was optimized low end hardware like Raspberry Pi 4. This is a quality of life improvement making perplexica more end user friendly.