How can I match afreetime recording to another freetime part

Cubase
#1

Can someone describe the process of how to match a free time event
to another freetime event?

I have a feeling this could be possible with the features that cubase 8 have, but I don’t know them all well enough
to do it yet.

to explain more what I want to accomplish:

I have a live recording with drums and a guitar. I can use tempo detection to find the tempo of the recording.
the guitar is played in sync with the drums of course, but the drums were recorded without a click track and is fluctuating in time a bit. However, I want to copy the guitar on verse 1 and paste on verse 2, and thus being able to pan the guitars. So cubase have to do some timestretch and know both the old tempo changes, and adapt that to the new tempo changes that might exist on verse 2. Obviously I want to repeat the same process all over the song
so that I get 2 guitars from that one take, for example take choŕous one and paste on chorous 2, and take chorous 2 and paste on chorous 1 and so on…

would really appreciate help, if you know how to solve this, and if you know about youtube videos
that desribe the process, or even just parts of it, that would also help tremedously!

#2

Hi

Once you do the tempo detection, you can select all your events and use the Audio/Advanced/Set definition from tempo.
This will embed the tempo information (where the beats are) into the events. And if they are in musical mode, they should follow the tempo track even when you copy and paste them.

Check this video. You can skip forward to about 58 min.
https://www.youtube.com/watch?v=kjAjI-Xd2_M

#3

Here’s my suggested procedure:

  1. Detect tempo on the drum tracks
  2. Refine the tempo using Time-Warp to fine tune barlines in the project to match with the drums (I quite often skip detect tempo because its not that reliable, so I use Time-Warp entirely and it only takes about twice as long as the length of the song to match up)
  3. Set Definition From Tempo on all tracks (as mentioned above, this embeds the tempo map into the events
  4. This also puts the events into Musical Mode
  5. Cut and paste guitar around, it should now fit nicely
  6. I often delete all the tempo events so the song is at a fixed tempo
  7. Or I selectively edit the tempo map to reduce the number of changes so there are simple speed changes or say V3 = V2 = V1 speed, Ch2 = Ch1 speed etc.

It can be a tricky process sometimes, and certainly takes a bit of learning at the start, but I’m sure there’s a few videos which can help you get further into those details.

Mike.

#4

Thanks for the excellent step-by-step, Mike!

I was wondering - if I had a variable tempo project, and wanted to bring in a vocal from another project that was recorded at a different tempo - is there a way to use “Set Definition from Tempo” like you wrote above?

Would I apply that process after I dragged it into the current project?

Now what I do is drag it in and use VariAudio to adjust the start and stop points. But it’d be nice to not have to do it that way!

Thanks!

#5

Hi,

TBH I’m not sure of the answer 100%, you might have to try a few things to see which works.

Perhaps the best thing to do would be to open both projects, do the SDFT on the vocal event in question and drag it to the other project (or cut/paste). Hopefully that’ll come over with the warp points in tact and it’ll then conform to the new tempo.

Mike.

#6

Yes, this is correct :slight_smile:
Do make sure that, when you “Set Definition from Tempo”, in the sources Project, that you use the option “Write Definition to Audio File”. You can then import the audio file anywhere you want, and it will adapt to the destination tempo.

#7

Great, thank you!

#8

From my tests ages ago I found that when you ‘save tempo to audio file’ that it doesn’t take all the tempo changes, it just write one single tempo into the file… That’s why I suggest using the drag n drop method because this takes over the warp points, which means all the tempo changes are used.

Mike.

#9

Sorry, I’m not understanding…
Why is Drag ‘n’ Drop better than importing? If the embedded warp points are correct, it should’nt make any difference how you then get the file into its destination.
(and, before posting, I did try the method, and the warp points were correct…reimported perfectly, and adapted to the variable tempo map I had created in the destination Project… admittedly, my test example audio file had very obvious warp points :wink: ).
(Btw, this doesn’t embed tempi into the audio file, just the warp points. :wink: )

#10

Just wanna say, thanks for all the responses on this. Very excited to try it out! which I unfortunately can’t do until a week from now, so I might come back with some follow-up questions then, But thanks everyone!

also, when I create the new track which are going to contain the pasted guitar parts, this track has to be put into musical mode right? other than that I don’t have to do anything? (after having used the set definition from tempo)

#11

Welcome to “Everyone confuses Musical Timebase vs Musical Mode”! :smiley:
A Track can be set to either Musical or Linear Timebase… meaning that events (that means… Parts, MIDI notes, audio event start positions) that are placed in the track lock to either bars/beats or time position when the tempo is changed. But the actual content of an audio clip remains unchanged (only its start position moves, if set to Musical Timebase)
To have the content of an audio file change according to tempo, the file has to be set to Musical Mode (either in the Audio Pool or from the Sample Editor).
… but, to answer your actual question :wink:… once you have your embedded audio, it should already be in Musical Mode (you’ll see a small icon at the bottom-right of the clip :wink: ).

#12

Ahh! thanks for the explaination, I had to do some youtube searching to find info about
musical mode. its a bit clearer now

If I got it right then I dont need to set the track to anything? and the “embedded audio” that you mention is
the term you use to describe that the function SDFT has embedded information into the audiofile about tempo changes, am I right?

regarding this aspect:
“Do make sure that, when you “Set Definition from Tempo”, in the sources Project, that you use the option “Write Definition to Audio File”. You can then import the audio file anywhere you want, and it will adapt to the destination tempo.”

I don’t feel okey doing this since Im thinking that I am sort of destroying the files a little bit, since the timestreching is likey to affect the quality a bit. I’m thinking that I can back up the project before I do it so that I still have the original intact somewhere. But, am I incorrect? Maybe the files “still knows” their original tempo and plays back in their full quality at these tempos normally, its just when cubase tells it to stretch that it actually stretches, or how does this work? if you happen to know :slight_smile: ?

thanks alot for the help really, getting very excited about the ability to fix these recordings up!

#13

right, and right :wink:

regarding this aspect:
“Do make sure that, when you “Set Definition from Tempo”, in the sources Project, that you use the option “Write Definition to Audio File”. You can then import the audio file anywhere you want, and it will adapt to the destination tempo.”

I don’t feel okey doing this since Im thinking that I am sort of destroying the files a little bit, since the timestreching is likey to affect the quality a bit. I’m thinking that I can back up the project before I do it so that I still have the original intact somewhere. But, am I incorrect? Maybe the files “still knows” their original tempo and plays back in their full quality at these tempos normally, its just when cubase tells it to stretch that it actually stretches, or how does this work? if you happen to know > :slight_smile: > ?

Don’t worry… it isn’t stretching anything… it is merely placing “markers” (think of them as “curtain hooks”… they only make a difference when you hang the curtain onto a different curtain rail :wink: ). But of course, do feel free to duplicate the audio file first, if that makes you feel happier :wink:.

#14

cool so you are saying the old .wav format has room for extra information that cubase stores into it without affecting the actual sound recording, and that I can then move this .wav file into any other project and cubase will always know its tempo ( which in this case is fluctuating throughout the whole file)

this is contray to what mike said: “From my tests ages ago I found that when you ‘save tempo to audio file’ that it doesn’t take all the tempo changes, it just write one single tempo into the file… That’s why I suggest using the drag n drop method because this takes over the warp points, which means all the tempo changes are used.”

Can I just assume that this is wrong, and only applies when there only IS one tempo, compared to when it changes about every ms…


so, to sum it up in a final question, all the info is in the .wav file without any need for any additional file with this extra information that I should also copy along?

wow, I’m learning so much, really appreciate it!

#15

I tried using just the wav file a while back because I wanted to know the answer myself. For me it didn’t work, that could have been in Cb7 I think. But if you really want to know which way it works, why not try it yourself to find out and let us know the result :slight_smile:

Mike.

#16

(Hi Mike,
when you said “one tempo”, I presume you meant the initial tempo (e.g. as seen when you look in Cubase’s Audio Pool)? Adding definitions wouldn’t change that… like I said it is merely adding markers… for Musical Mode when the host tempo is changed… it isn’t inserting any actual tempo data at all (I too was under that misunderstanding at first, expecting the audio file to contain an actual tempo map :wink:. Wouldn’t be any use for adapting the file to other tempos anyways :wink: , merely for writing the tempo map to the corresponding location in the DAW (which, admittedly, could be useful too :wink: )

@Score81
Yep, it just places “hooks” at every tempo change from the host, and locks those hooks to the corresponding bar/beat position in the destination when in Musical Mode, and stretches/shrinks the audio as required.

However, (also in response to you question) I have never tried importing a definition-embedded.wav into a different DAW than Cubase… but I would expect this meta-data to be simply ignored.

#17

Mix part 1 than MIX part 2 and just glue them together on Wavelab (or in the mastering suite in Cubase, starting a new poject)

I dont try to synchronise when its not synchronised

I try to avoid to spend more time on technical issues like that than to write a song

When it become complicated I find ways like that to end up the damn song or music and go to another one

Life is too short

#18

I have used this mental model since I read something you wrote a while back. But I have always wondered, maybe I can ask here:

Let’s say SDFT adds 5 hooks/markers/“definition points” to the audio file in the source project.

But perhaps this doesn’t match up with the destination project, maybe there are 2 or 10 tempo changes in the time frame the audio is pasted to, instead of 5.

I wonder how Cubase decides to map the 5 SDFT hooks/markers/“definition points” in the audio file, to the different number of tempo changes in the destination project? If it does it “wrong” I could easily see where the stretching would not result as intended.

Maybe?

#19

Let’s say SDFT adds 5 hooks/markers/“definition points” to the audio file in the source project.

SDFT “prints the project grid” to an audio file, following the tempo changes, if any.
It’s not 5 hooks or 6 hooks.
I guess you could also do it manually in sample editor definition tab.

Using it after tempo detection, it will mark each beat.

#20

I’ve no idea how the details of that work either :wink:, but I guess it is no different from keeping the audio clip at its original/source position, putting it into Musical mode, then making tempo changes there.
Maybe (thinking about it further, I think this is probably more logical :wink: ), rather than placing markers wherever there is a tempo change, it simply places a marker at every beat, according to the source tempo track, so all it has to do is hook them up to every beat in the destination.

EDIT: misohoza just said the same thing :wink: