Aria Tag mode for Audio Descriptions in CT-Player

Problem
- Instead of speaking out loud using react-speech-kit, use aria tags that can be rendered by a users screen reader

Solution
- use aria-live with assertive mode to signal screen reader that there is a new audio description to read

Start by locating the CT-Player audio-description flow and its use of react-speech-kit. Replace the spoken-output path with an aria-live region using assertive mode, then verify that newly generated audio descriptions are announced by a screen reader and are no longer spoken through react-speech-kit.

Written by the indexing model from the issue text.