I want to display captions using the HTML5 Text Tracks API (WebVTT) using the native browser implementations -- without loading audio/video.
How can that be achieved ?
There is no way using the native implementation.
However, the Firefox implementation of WebVTT, vtt.js, is publicly available.
All you need to do is attach a timer while a video element is playing (or just to any custom event without a video) and apply vtt.js functions to the video / captions container.
Since the Firefox VTT implementation is derived from it, this is the closest you can come to native without using the default way of attaching to a video.