jkoseattle
...I would try zooming in on the audio track to see where your syllables come before the barlines and try sliding them to the right ever so slightly....
How many takes did you do for these vocals, and how small are you slicing up the takes (meaning, are you singing the whole song beginning to end or recording small bits at a time)? I find that by recording just a line or two at a time and doing a dozen takes or more in succession, I find little vocal inflection ideas that I would not have come up with otherwise. Plus then I'm sure to get a take I really like. (Of course, this means a ton of work auditioning all the takes and also in making sure the lines sound blended together naturally.)
I use both these techniques all the time. I usually record at least six or seven takes of the entire song, and some tough sections...I had a four-bar phrase in one of my tunes that I had 30 takes! I am a baaaaaad singer
. And by bad, I don't mean good!
Often, I only sing alternate phrases each take. This allows me to take a deep breath and lay down a good, strong take. For instance, Take One might be:
"Row, row, row your boat"
<not sung>
"Merrily, merrily, merrily, merrily"
<not sung>
And Take two is:
<not sung>
"Gently down the stream."
<not sung>
"Life is but a dream."
Before even listening to my takes, as a rule I split them all at all possible split-able spots...Ends of phrases, any breath...any place the wave is flat-lined. I then audition, and comp to a new audio track (I usually name the composite track "voice_aggregate".) Then I bounce the aggregate track to a single clip and export to Melodyne. I fix some things there, re-import the melodyned track back into Sonar to a new track, then listen for problems with timing. If there are problems, I once again split the clip at split-able spots and nudge left or right to get the timing as close to what I want as possible.
Simple, eh?