Sylvan
...
 
I wonder if doing some de-essing would help that. I will try it later.
 
... 

 
I've always found my best results after cleaning up the takes. I'll do some general EQing, de-essing, and maybe noise/reverb removal. Then I go through with a sync pass, to tighten everything up. 

Seth Kellogg [Cakewalk]

Sylvan
...
 
I wonder if doing some de-essing would help that. I will try it later.
 
... 

 
I've always found my best results after cleaning up the takes. I'll do some general EQing, de-essing, and maybe noise/reverb removal. Then I go through with a sync pass, to tighten everything up. 

Yes, that would do it. Perfect. After making the adjustment I have expected results now. The glitch is gone! If I had cleaned up the take before VocalSyncing, everything would have better than perfect. I was just hurrying so I could provide any helpful feedback. Thank you guys so much! 
 
I want to go on a field trip to Boston and see the Cakewalk headquarters someday.

Anderton
 
 
 A little inside baseball...on Sunday, Noel generated four test files to test Melodyne fixes. Around midnight Boston time he posted #4, I tested, and could confirm that it worked. It also tightened up audio in some other respects. Pretty cool.
 

Yes, Noel is amazing. The whole team is amazing. That was really fast. Seth looked into the VocalSync issue right away too. I hope more current and prospective users see the dedication and personal pride this team has. You can't go wrong by boarding this ship.

Excellent!

Noel Borthwick [Cakewalk]
NOTE: To change this go to Preferences | Configuration file and edit the value there and save it. Then you must click Reload configuration to load the settings in SONAR. Please also note that the change will NOT apply to clips that have already been vocalsynced, only to new vocal sync clips. To make it apply to an existing VS clip, first remove vocal sync and then Vocal Sync it again and then render.
 
I updated my original post with this. For a baritone speaking voice we didnt get artifacts with the bass version. In a future update we may consider adding the stretch modes to the UI. The default works well for most normal cases.

Great...

Sylvan
Can anyone reproduce this? http://youtu.be/gBkqV1F0dzI

Wow.
 
That sounds awful.
 
I guess it's back to VocAlign or Revoice Pro if you want great results with out having to break out Notepad every few minutes.
 
It's sickening to imagine that this was based on the same algorithms as AudioSnap and that GUI access to choice of algorithm wasn't included as an easy to select option for the very same reason that the algorithm choice is available in the AudioSnap panel.
 
It's another Half Baked feature roll out. Yuck.  
 
Thanks for sharing.
 
 
 
edit grammar

Are you experiencing the problem Mike? There's an easy fix and it really only affects those deeper baritone or bass voices.
Sylvan's voice is clearly very deep and mine is nearly at that level so I had to make the change but earlier testing with all sorts of other voices didn't show any signs of the problem. I don't think I'll be breaking out a 3 or 4 click edit every few minutes.

mike_mccue
It's sickening to imagine that this was based on the same algorithms as AudioSnap and that GUI access to choice of algorithm wasn't included as an easy to select option for the very same reason that the algorithm choice is available in the AudioSnap panel.
 
It's another Half Baked feature roll out. Yuck.  
 
Thanks for sharing.

 
You are jumping to false conclusions without understanding or even trying the feature. The stretching aspect of VS is only a small part of what it does. We wanted a feature that worked for the most common cases out of the box without a million options to tweak. Its most definitely not half baked and we went through many months of iteration on this. The most important part of the DSP in such a feature is the part that analyzes and detects the onsets for the voice and determines how the alignment should be done rather than the stretching which is the last step. Vocal analysis is a very deep field and the Vocalign product has evolved over over 15 years. That product is also has a lot of other supplementary features to clean up the audio besides the analysis and they have built in handling for various edge cases. If you want a dedicated product for this area and are willing to pay the price for it that might be a better solution for you.
 
Vocalsync on the other hand does a remarkable job for what its designed to do and the integration in SONAR couldn't be simpler. Its the first version of this feature and it will only get better.

Sylvan

Seth Kellogg [Cakewalk]

Sylvan
...
 
I wonder if doing some de-essing would help that. I will try it later.
 
... 

 
I've always found my best results after cleaning up the takes. I'll do some general EQing, de-essing, and maybe noise/reverb removal. Then I go through with a sync pass, to tighten everything up. 

Yes, that would do it. Perfect. After making the adjustment I have expected results now. The glitch is gone! If I had cleaned up the take before VocalSyncing, everything would have better than perfect. I was just hurrying so I could provide any helpful feedback. Thank you guys so much! 
 
I want to go on a field trip to Boston and see the Cakewalk headquarters someday.

Hi Charles,
 
Yes as Seth indicated you might require some preprocessing of the audio depending on the use case. For example if syncing vocals that were recorded in a noisy environment (as is the case when doing dialog replacement) you may need to do some noise filtering as a prior step. We figured that since you have several tools in SONAR initially we wouldn't attempt to do this automatically (anything automatic can be a can of worms in itself). The first cut of this feature is give you the an integrated tool to align vocals in a simple intuitive way. 
We look forward to hearing more from you once you use it for post stuff. With constructive feedback like yours we hope to improve the feature and workflow in SONAR in the upcoming months.
 

Noel Borthwick [Cakewalk]

mike_mccue
 
It's another Half Baked feature roll out.

 
You are jumping to false conclusions without understanding or even trying the feature.

Jeez that's not like Mike at all. Notice how he didn't let the fact that the OP is completely satisfied with the fix and with the performance of the feature dissuade him from the truth.