I have started working through a set of projects that were recorded at 44.1kHz/24 bit to convert them to 48kHz/24. The method I use is:
save a fresh clean Backup of the Project to work with
change the sample rate in the Project Setup page
say yes to convert files to match
say no to keep sample positions.
This mostly works - but several vocal audio files have ended up with some individual words dropped in volume dramatically - this happens on (only a few) of the files where I have at some point gone in and raised and lowered individual words in the vocal using the range-tool clip gain adjuster.
I guess this is something to do with the boundaries of samples in the conversion, and something that happens as part of that last step where I say No to keeping sample positions as-is. I said “yes” to that on the first project I tried and some stuff came out off-tempo.
Any one shed any light or suggest tips to do in prepping the file to try to avoid this? (Short of baking in my level-changes before the upsampling).
you can use Bounce Selection.
It might be a good idea to work with duplicated track versions in order to have a safety backup. Now, just use the Range tool > Audio > Bounce Selection. Crossfades, Vari Audio adjustements, etc will be baked into the newly bounced file. This should do the trick.
Glad to hear it works.
Regarding commitment: That’s why I suggested to work with track versions. You can easily go back with just one click, nothing to worry about.
Thanks, I’ve learned a lot this morning about using Track Versions before Rendering in Place. However, I now see that EVERY clip gain adjustment in the audio files goes wrong when you upsample the files. Other things (variaudio, automation) seem to fare better.
So this is a warning: it’s not painless / risk free to just upsample a project and carry on working where you left off (I assume downsampling is similar). I guess it’s not a bug, it’s a limitation.
My new process for these work-in-process files is this:
a) Go through all the audio files that use clip-gain inline adjustments: use the inspector panel to save a new Track Version (arrow down to next track, repeat till done).
b) Bounce all the files in place on the new Track Versions
c) Backup the Project
d) Working in the new Backup Project, I use Project Setup to increase the Sample Rate
e) Say Yes to “convert files to new rate?” and overwrite the older files in the Pool, say No to keeping samples in Place.
f) If I don’t like my bounced files while working with the project, the only option is to go to v1, and Remove Event Audio Curves for the track and start again with the breath-dipping etc.
Just a comment about your wording regarding bouncing.
Are you aware of the difference of the two options you have in Cubase ?
Render in place is different from just bouncing (convert audio to file)
I have learned jus one week ago from a YT video by Dom Sigalas