Given a full audio source and a full (non-timed) transcript, is there any free software to automatically generate a timed transcript (like a SubRip file) using voice recognition? It would listen for the words in the transcript and assign times to them. This should be an easier problem to solve (programmatically speaking) than regular voice recognition.

Examples for audio: https://media.GRC.com/sn/SN-635.mp3 and transcript: https://www.GRC.com/sn/SN-635.txt