Audio speed changes — Delivery workflow

Worked example

Name each version after the multiplier, for example dictation-unit4-080x.mp3, instead of keeping only the automatic -slow-down suffix.

Six weeks later you will still know which copy is which, and the untouched original stays in its own folder.

Match the file to where it will be played. MP3 at 128 kbps works on nearly every learning platform and chat app, M4A is smaller at the same quality and plays well on phones, and WAV is only worth sending to an editor who will process it again.

Say you run a weekly class: export 0.8× and 0.9× copies of the same dictation, post both, and let learners choose. Record the multipliers in the lesson notes so next term you can repeat the exact result from the same original.

A common mistake is reusing timestamps from the original. If the transcript says a word starts at 0:30 and you slowed the clip to 0.8×, that word now starts at 0:37.5, so divide original timestamps by the multiplier before copying them into captions or show notes.

Before sending, play the file once from a phone or a different browser and check the length shown by the receiving platform. If a platform trims or re-encodes audio, check its current upload guide for limits on size and duration. That small habit saves a lot of confusion when several people share the same folder.

60 s at 0.8×

75 s theoretical duration

Keep the original, the multiplier and the export settings together; that trio is all you need to reproduce the same file later.