Add ability to speak prompts

Audio Semantic Search is slower than Text Semantic Search, which runs off textual input. In order to support Voice, Voice to Text functionality needs to be added

In order to not overwhelm myself as a solo dev, I put focus entirely on text.

When the program is finished and working wonderfully, Voice to Text support should be added.