Novice Seeking Input on Personal Voice To Text Project

Hello everyone! I appreciate you reading this.

I am a complete novice who currently works in manual (black box) QA. I'm looking for some good first steps to get started on this project, such as tools I should look into using and such.

Essentially, I've noticed that sports fans say that sports announcers are biased (the recent MLB NLCS as an example). I would like to gather data on what the announcers are saying by converting what they are saying to text (or some other format a computer can understand), and then note how many times each team is mentioned. This data could then be made public to confirm or deny supposed biases. Eventually, I would like it to be able to use real time video or audio from a streaming service perhaps… as opposed to ones that are uploaded. I realize that part is likely very difficult.

I've noticed that Google can process voice to text through an API… and I would like utilize some common web-based technologies such as Javascript or similar, because that's what the folks I work for use: we a Web Application.

Thank you in advance!