Whisper Subtitles

Whisper Subtitles transcribes the audible mix inside the selected timeline region and returns word-level timing for subtitle placement.

Kepler renders the selected region as timeline audio first, so the model hears the same mix represented by export: unmuted audio tracks and audible video clip audio. The generated result becomes ordinary editable text clips on the timeline, which means users can move, trim, restyle, or delete captions after generation.

Use this model when you need fast automatic captions, rough dialogue timing, or editable subtitle drafts. Keep the selected region tight around the speech you want to caption for clearer timing and lower credit use.