Well, as you can see, I'm newbie. I'm trying to learn how to remaster videos. I'm using Sony Vegas 12.
I don't know if experts use the expression "remastering" with the same meaning as I do, so what I'm trying to do is synchronizing an dubbed audio with a raw high quality video, which still has its original audio (then I'll make a dual audio video).
As you can see here http://i.imgur.com/2UjWTS8.jpg , the graphs of both bitrates (I think that's the word) are very different from each other. The first (the original) audio is way lower. If they were more similar, I could see their similarities which would allow me to do a very accurate synchronization.
However, I don't know how to make them more similar. Some tutorials suggest you to increase volume, export the original audio and import it again, so you would have a clear bitrate graph. My concern is that this way would re-encode the audio. I would like to not doing that in order to to avoid any quality loss, especially taking into consideration I will fill in some empty parts of the dubbed audio with original audio.
What should I do in order to better compare both audios without any conversion?
Our website is made possible by displaying online advertisements to our visitors.
Please consider supporting us by disabling your ad blocker or buy a VSO converter software :)
Please consider supporting us by disabling your ad blocker or buy a VSO converter software :)
+ Reply to Thread
Results 1 to 8 of 8
In Vegas you can just drag the track to make it larger so you can see the waveforms better. Working in uncompressed .wav files will allow you to manipulate your audio with no inherent loss other than the modifications you choose to make.
You don't state what you want your final output to be or how it will be played, so no one can tell you what your best workflow would be, but I suspect you're adding unneccessary steps.
As far as terminology, "remaster" implies something that you're probably not doing, though it's not absolutely wrong. And you are using "bitrate" to mean waveform (or maybe volume.) We understand what you mean this time, but that may not always be the case.
"Remaster" means to take an original source and clean it up, adjust & repackage to create a new version of a Master. You don't really have the original source (you've got a heavily compressed consumer distribution copy), but you are doing an equivalent process. "Optimized consumer remix" would be more appropriate.
As mentioned, NOTHING you've talked about has anything to do with bitrate. That is a "Waveform" Graph that tracks the volume of the signal (amount up and down) across time (horizontal). You should read you manual! It's all in there.
If it's possible, I would have Normalized the upper track (this Maximizes the signal by raising the whole track by a certain amount so that the highest peak in the track is = 0dBFS). There are inherent costs to doing this (raised noise level, incl. quantization noise), but for you I don't think it would be noticeable. The Normalization should have occurred prior to your adding the 2nd track. Unless you have the tracks ganged together, you still should be able to do it now, though.
Your worry about quality loss is valid, but you are doing a major adjustment to the source anyway, so I'm not sure if you'll notice with all the other stuff going on.
You could, after having synced up your track2 to track1 & the video (assuming you aren't editing those tracks), export ONLY your track 2 to an audio file and then remux it with the original, thus not losing any quality in the vid or track1 audio. Only track2 would have been re-encoded. That part is almost unavoidable, since you are doing stuff to it (though SmartRendering could be your friend here).
In Vegas when working, you can save particular views of the timeline, etc. as Workspace Presets. So, save a default and also save one with expanded waveforms (particularly along the vertical/volume axis). Then switch back & forth between them while working.
Hey, Clever. Welcome.
If you're going to be doing any "remastering" work on the new audio track, I'd suggest doing it in a dedicated audio program. You might want to make equalization changes, adjust dynamic range, reduce noise, etc. If you're already happy with your new audio track, never mind.
To sync up the two tracks, yes, it can be helpful to have a waveform display that fills up more of the track. Vegas calls this display 'peaks'.
First, lay out your existing video with its existing audio on the timeline. You'll have one video track and one audio track.
Add an audio track and drop your new-and-improved audio on that one. You can probably get roughly in sync just by sliding it around and listening, if you're familiar enough with the material.
To get a good, tight, close sync, scrutinize the waveform graphs for a transient sound -- a hand clap, gunshot, door slam or other sound that's sharp and sudden. You can use your mouse wheel to zoom in quite close and line them up the best you can. You might not be able to get it visually perfectly aligned because Vegas will snap the audio to the closest frame, which will probably be 1/30 of a second...if you were just matching two audio tracks in an audio program, you could match even closer, but it probably won't matter.
Once those two are synced up, check later in the file to see if they're still in sync later on. If the recordings come from different sources or different machines, the sync might drift a bit over time. If you can get a perfect sync near the beginning...and find a perfect point to sync TO near the end, Vegas will let you stretch the audio by holding down CTRL and pulling the edge of the event. If you don't go more than a few per cent stretched, it won't sound bad.
Once your video, old audio and new audio are synced, go ahead and select all three and hit G for Group. It will tie the three items together as one, so if you move one, you move them all.
If the old audio is really so low that you can't see to make a match, go ahead and make a copy that's normalized -- it'll raise the loudest point in the file to 0db. You don't have to use that normalized file in the final product. You can bring it into the timeline and group it with the existing one for reference, then mute it if you want.
Well, I'm trying to synchronize the .aac dubbed audio (which was extracted from a mp4 low quality video) to a .mkv high quality video, which has its .ac3 original audio. As you know, Sony Vegas doesn't directly work with .aac, so I just imported the mp4 file to Sony Vegas and deleted the video track. Sony Vegas doesn't work with .mkv or .ac3 either. In order to work with the .ac3, I extracted it to a sound file using mkvtoolnix and imported to Sony Vegas with the help of the program AC3WAV. For the video, I just converted the .mkv to .mp4 in order to work with synchronization on Sony Vegas. I won't use Sony Vegas to merge audio to video, so the conversion was only for working purposes.
My purpose is synchronizing the dubbed audio (and filling in any empty parts with original audio, since dubbed audio is smaller most of times), and then export it alone (I didn't decide yet the output, but it may be .aac). I'll use mkvtoolnix to merge the worked audio with the raw .mkv file.
As I explained above, I don't really mean to use Sony Vegas to export all audios and video. I'm just using it to work with the dubbed audio in order to export it and merge video with the help of other program, as you said.
Since it's my first time doing all that, it may be natural that my first "project" won't be so good as professional ones. In fact, I can't even notice some things experts do. However, I'm trying to avoid any loss, as you see, I'm trying to follow some standard recommendations in order to at least make fairly good quality works.
Hello, thanks for the answer and thanks for the welcome!
As you see, I'm very newbie. The tutorial I'm using just teaches how to synchronize using Sony Vegas. I'm open to any suggestions, but I'm not sure that, if I used a dedicated audio program, I would be able to check audio and video at the same time. Since I'm a bit more used to Sony Vegas, I'd like to use it, but if you may suggest a good audio program which could do that, I'll thank you again. It may be helpful since afterwards I'll try to learn how to improve quality of dubbed audios in order to make better videos.
Since I'm very newbie, I'm afraid I can't quite understand all the stuff you and Cornucopia described, but I'll definitely try what both of you suggested when I have some time to work that video again. I'll give you some feedback later.
Last edited by Clever Sleazoid; 12th Mar 2014 at 18:40.
Thank you, people, you were very helpful I've finished my first work, and it is very good.
I still have problems dealing with sound files, though, since I'm filling in empty parts of the dubbed audio with parts of the original audio. It's difficult, since both have different qualities, volume, and I still have to pay attention to make all the audio sound like a perfect continuity.
Do you have any advices to make audio that uniform?
Only one more question: when I normalize the audio, am I re-encoding it?
Thank you again!
Last edited by Clever Sleazoid; 13th Mar 2014 at 17:52.
Can't say without hearing them, but it is likely you would NEVER get them to sound even close to identical.
So many variables come into play when creating & mixing sound:
1. Vocal talent
3. Mike type, sensitivity, polar pattern
4. Where mike is positioned
5. Kind of mike pre-amp & A->D converter
There are complex tools that allow one to "Convolve" & "Deconvolve" to get one sound to take on the aspects of another sound, but unless you know and have mastered & articulated & stored those minute features involved in each of those things listed (and more), it would be quite impossible to match.
But good for you for getting as far as you have!
I think it's impossible to take into consideration most of points you mentioned, since I can't change the way the recording was made. I'll have to work using what I have.
Well, you know... I would like to get better audios, but unfortunately all I get is TV ripping. Most of times, stuff I want to edit doesn't even have an official release. It makes any high quality work kinda complicated (but, even so, better than nothing). Even so, some works available on internet are impressing!
My purpose for now is just making the empty parts filled in with original parts sound as similar as possible to the existing audio, including its continuity. In a near future, I'll definitely try to learn more about audio working.
Again, thanks for your help! Since you seem to know a lot about this subject, I'll thank you for any other advice you might give me.
Last edited by Clever Sleazoid; 15th Mar 2014 at 01:13.