Do I need an SRT or a TXT file to make the captions?
No!
That's actually why we built Zubtitle in the first place. When you upload a video into Zubtitle, it will automatically "listen" to the video and add captions to the video.
Not only do you not need a transcription file, Zubtitle will generate a .txt along with the final video (which will include the captions/subtitles).