File: 8 hours long.

Task: Let's say I recorded my training shift at work. I want to remove anything but human speech from 8-hour long file, in order to create checklists/flow diagrams based on what was said in my training.

Output: I'd like the output to be a shorter audio file, which contains only speech (some/normal level of background noise is not a problem).

Question: What's the most efficient/fastest way to do that?

Potentially (not a necessity) it would be useful if I could plug the output into speech-to-text software if that's supported by audacity.

  • Show us you at least did some research into this & where you failed to achieve your goal.
    – Tetsujin
    May 27, 2019 at 17:22

1 Answer 1


Protools has a "Split Silence" function which can be used to seperate out all elements where the audio level does not reach a particular threshold. Assuming that most of the audio is actually speech, then this is where I would start. Once you have all the speech segments, you can then 'shuffle' them down to a single region which you can then export.

Yes, there will be some manual work, but there isn't a plugin or app that will do this automatically that I am aware of.

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.