| nhibberd |
It's quite simple and once you have done it a few times it's easy. You first need to import the vocal wav into your sound pool. Drag it to the sequence.
You need to resize the audio track to make it bigger, that makes it visualy easyer to find reference points in the waveform. If you are lucky theirs the remnants of a snare or something for you to synchrosize the track with. Otherwise you need to find referenc epoints by ear.
Imagine the music playing over the vocal track and tap with your feet or something to find the rhythm. Cut the vocals at the exact start of a bar an quantize it to the start of a bar on your sequencer. While doing this you will need to use the hotkey 'j' a lot for a good workflow.
Now you have synchronized the Q-point of the vocals to your sequencer and you need to adjust the BPM to match the vocals. Have Cubase play the audio track solo and tap with your feet again and see how far along the track your 5th tap is. If it's further than a bar you adjust the BPM to play slower and visa versa.
After that you need to make it all more acurate. You can loop four bars on the sequencer and try to find a natural sounding loop. If the vacalist seems to start the next bar and then the loop replays and starts again then you need to speed up the BPM.
After that you can add a simple peace of drums to see if it sounds natural. If you want to be more precise you do the same trick with 8 bars, or 16. But a good thing to know is that a lot of tracks have prety straightforward BPM's like 138, also very often they end in .666 or .333. A lot of more classical music styles have standard metronome BPM's. Go and google and find a pic of a metronome and you will see the standard BPM defaults written on it.
Once you have found the right BPM you rightclick the file in the pool(not on the sequencer) and chose the option proces->time stretch. Set the source BPM to the BPM you have found and the destination BPM to whatever you want your track to be. You can also chose what algorythm to use for the timestretch. If your in a hurry you use 'mode 1'. But if your comp is fast or the vocal sample is short you can use 'advanced mode'. Let it proces the time stretch and after that you can reset your sequencer BPM to the BPM you chose to have your track in. I would advise you only timestretch once because you can lose some quality in the sound.
---------
For the pitch I would use a plugin, not a proces because you can tweak it to match in realtime by ear and if it's a few semitones out of tune the right pitch is very hard to find through trial and error. Waves platinum bundle has an excellent pitch shifter which doesn't glitch like most do. It's called 'SoundShifter'.
The problem is that pitch shifter VST's use a lot of CPU and you might considder finding the right pitch with the VST and then pitch shifting with a process after so you can remove the VST and save CPU-space. The pitchshift process can also be found by right clicking on the soundfile in your pool. proces->pitch shift.
Now you have both pitch and tempo right and you can cut and paste in the vocal sample all you like. It might look like a lot of work seeing as I'v almost written a book here. But when you've done it a few times it's no more than 3 minutes work.
kind regard,
Charlie Darwin |
|
|
