We imported some music into a 48khz pro tools project to record some vocals, the music file was at 44khz and we didn't SRC convert it to 48khz, therefore the music was slower than the original. The whole session is out of time and out of tune.

How can someone convert this back to its original? What is the process for repairing this?

3 Answers 3


Only one thing I can think of without telling you to redo everything, and this is gonna be dicey. I don't if the results will actually be any good.

The fastest way will be to re-import the music file (as a second file, different file name) and allow SRC to properly convert it. You should then have both your buggered music file and the new version that is at the correct speed and pitch.

First consolidate your vocal track into one file that is the exact length of the music region it was recorded against. Find the exact length of the new (correct) region (you may want to view in sample), and varispeed time-compress the incorrect music file to match. Play back and A/B between the two music regions to make sure that it worked properly. Now, time-compress the vocal track (NO VARISPEED this time) to match the exact length of your music files, and you should be there.

Again, I don't know how good the results will be...but it's worth a shot.

  • Sounds like it should work, will pass it on to try! Commented Nov 30, 2011 at 14:27

I don't know what you can do with pro tools. I write to you what I would do on my Nuendo, maybe you can do the same on pro tools.

Make new copy of the project. Brutally switch the project settings from 48 to 44. Nuendo ask to you if you would convert or not the audio files to a new rate. You do answer “NO”. Nuendo ask to you if you want to keep the events at their sample position. You do answer “NO” By this way the events remain at their starting sample, perfectly in sync with your project and the music files are like the original 44 without any SRC. Now your voice is slow than before but is in tune. You need simply time compress it by 8,8354% with the best algorithm you have (for nuendo the ratio is 108,8354%). Time compression with best algorithm on voice is normally without artifact. That's it.


Can you just reimport the music with the right SRC for the project then re-pitch the recorded vocal to that?

  • @Andy Unfortunately it needs time stretching as well as re-pitching, is there a way to calculate the percentage to speed/pitch it up by to bring the length back? The original music was 1min05 at 44.1k, after importing it to a 48k without ticking the SRC box on import audio, it's now 1min10 Commented Nov 30, 2011 at 12:54
  • @edmatthews82 Are you sure you're telling this the right way around? If you play a 44.1k file at 48k, it should be faster and shorter. Sorry if I add to the confusion.
    – EMV
    Commented Dec 1, 2011 at 6:15

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.