Based on a comment in another YC news post, I thought juwo was a tool for automatically transcribing audio to text (is that correct?).

If so, then why not just say that; the demos can be simplified to be the raw audio/video clip with the text, as transcribed by juwo next to/underneath the clips.

no, that's not correct. It does not transcribe automatically. however the labels can have functionality attached. for example, to run the speech recognizer you have installed on your computer, on a voice note.