Audio in. Segment times out.

Speech, with the clock still attached.

Drop a WAV or MP3. Cue returns the words and the start and end of every phrase, ready for captions, search, or the next system in line.