x
Hello World 6 July 2011

Has anyone ever run into this problem?...

You have a full mix and you need to remove one track?
You have the separate track, but the files need to be perfectly synchronized.

I have run into this problem so many times.
So my question is 2-fold:

1. Does there exist any software that can completely sync the full mix with the separate track already on the market?

2. Should I bother to create the program that can do this? In other words, would any of you find this useful if it was an automated process?

Thanks in advance for ANY replies or advice you can give me. I can write the code. I work in the IT industry for a living and also write programs and dabble in specific need-based software like this.

By the way, the only software that I know of that can do this easily for the end-user is video-based, I forget the name of it right now but basically it takes the crappy onboard mic from high quality video cameras when you have more than one source and time-synchronizes them, if they are not properly time-coded.

I know exactly how to write this program -- but it's going to take me a lot of time and if there exists a program (or plugin ) that already does this I don't want to waste my time or yours :)

Topic Tags

Comments

Hello World Thu, 07/07/2011 - 12:59

audiokid,

Imagine this only: You have been provided a single mixed down .wav file for mastering. The vox is way too low to work with. So you are only provided with the vox separate track. So now you have exactly 2 .wav files in your possesion only. But the only problem is that the vox track is out of sync with the full mixdown. Now comes the tedious and time consuming task of synchronizing the 2 files you have until they are perfect down to the sample level. Hasn't this ever happened to you?

This is just one example I can think of but by far not the only one. Sometimes I get mixing projects and the tracks are all out of sync. I have a reference mixdown from the client that is a rough mix only. Everytime this happens to me, which is almost always, I think to myself I should just write an application that would auto-sync all the tracks to the reference rough mixdown.

Sorry this is difficult for me to explain, but the idea is to synchronize tracks that are out of sync. Does this clear things up at all? Thanks in advance for your reply.

Hello World Thu, 07/07/2011 - 17:42

Cool -- I'm glad you get was I was trying to say -- Overlapping the exact vox track actually does not cause null and/or weird phasing problems as long as you get it perfectly synched down to the sample believe it or not, in fact that's the whole beauty of it -- you can even remove the track altogether if you want by inverting it (assuming that you can also get the amplitude perfect also.) There's a ton of other uses for this. It's just that I have been doing it manually for years and it's time I write the app that can do it for me. So can I assume that there is not something almost identical already on the market? And do you think this is something that other people would even want, or is it just me?

dvdhawk Thu, 07/07/2011 - 20:54

If I sent you a completed mix and sent you any alternate track(s) a year later both sound files would start at 0:00:00 in the DAW and export the exact same length audio file. If nobody truncates the beginning of either of these files, you can drag the new track into the timeline starting at 0:00:00 and it's done.

I just contributed a part to a friends recording project via the internet. He sent me an mp3 of the rough mix of his latest song. I imported his entire mp3 file into my different brand DAW and played along with it without cutting and intro or silence at the begging. I tracked my part and emailed him the uncompressed WAV file of my part starting from the beginning. He slid my track into his DAW in a few seconds. Both tracks were the same length and started at 0:00:00 - done.

If your clients are exporting just the new vocal track minus the time that elapses before the vocals come in.... tell them to export the files from the beginning of the song 0:00:00.

I might be missing something, but this just seems like a procedural problem to me.

If this is happening 'almost always', it's time to insist the people sending songs do a better job from a procedural point of view. Or provide a combination of pulses at the beginning of the recording. Much like the test tones we used to have to put on our final mixes back when dinosaurs roamed the earth and tape machines had to be calibrated every time you gave your tape to someone.

Best of luck to you.

bouldersound Thu, 07/07/2011 - 22:27

I agree that it's a procedural problem as dvdhawk said. That said, invert the polarity on the vocal-only track and slide it around until you cancel out the one in the mixed track. You should be able to get the alignment perfect as long as there is no difference in eq applied to the two tracks. Then flip the polarity back.

Hello World Sat, 07/09/2011 - 06:58

I completely agree with the last two posts. It is a procedural problem for sure. But, it's difficult to change people. By the way, mp3 adds padding silence which does in fact throw off the sync. We've all dealt with sync problems caused by many different reasons. I just want to write an app that synchronizes the tracks automatically and quickly. If anyone wants to alpha test it, feel free to let me know. Thanks everyone!

dvdhawk Sat, 07/09/2011 - 12:12

You are absolutely right about people being hard to change. Billing them by the hour (and telling them what portion of that was avoidable) usually keeps them from wasting your valuable time and their own money doing something they could correct.

What program are you using that adds silence to the front of your mp3's? (just curious, I've not had that issue.) Good luck with the app though - I'm sure there are others with a need.

bouldersound Sun, 07/10/2011 - 09:21

Okay, wait a second. First you say:

Hello World, post: 373734 wrote: So now you have exactly 2 .wav files in your possesion only.

Then without explanation:

Hello World, post: 373804 wrote: By the way, mp3 adds padding silence which does in fact throw off the sync.

You drop that "by the way" on us as if you had called it an MP3 from the start, which you didn't. Then you post something (addressed to me) about cancelling a track being a good use of sync but not what you were after when I had said that inverting was a good way of syncing the two files, which is completely different. Then you edit the post to put a link to some software. If you are going to change the facts and your posts every few hours there's no point for us (me, anyway) to waste our time on whatever the question is.

If someone came to me with the original request (using WAV files) I'd be reticent about taking the project. If they had MP3 files I almost certainly wouldn't take the job because the results would almost certainly be disappointing.

hueseph Sun, 07/10/2011 - 11:20

I'm still confused. You still haven't mentioned what DAW you are using. I'm not saying this problem doesn't exist but I personally haven't experienced it. The only issue similar to this would be working in a program that doesn't have ADC. Like ProTools LE. But that is finally a thing of the past. So, what DAW are you using.

IIRs Sun, 07/10/2011 - 12:29

hueseph, post: 373824 wrote: Aren't there already formats within most DAWs that allow you to export a file with it's time stamp in tact? Ogg?

Broadcast wavs (BWAV)

If the files are timestamped they can be automatically moved to their original positions on the timeline, assuming your DAW supports it.

hueseph Sun, 07/10/2011 - 15:04

IIRs, post: 373850 wrote: Broadcast wavs (BWAV)

If the files are timestamped they can be automatically moved to their original positions on the timeline, assuming your DAW supports it.

Well, there you go. Thanks for clearing that up. Most DAWs already support this these days. Even ProTools, though very late in the game:rolleyes:. AFAIK Cubase, and Logic both are capable of BWF.

EDIT: Here's the Wiki just to clarify.

[[url=http://[/URL]="http://en.wikipedia…"]BWF[/]="http://en.wikipedia…"]BWF[/]

lambchop Mon, 07/11/2011 - 08:44

This is a very interesting discussion. I just got Sony Vegas for working with some of my band's videos of performances. We recently performed an outdoor concert and the audio from the video was less than perfect. My bass player also recorded the audio with a portable recorder of which I downloaded multiple instances to my DAW program and manipulated the tracks with narrow band eq and some multiband compression, then rendering it to a new wav file and importing it into Sony Vegas. I must've spent about a half hour aligning the new track to the video's audio. Hopefully there is a function built into the Vegas software to perform exactly what Hello World is referring to and I just didn't find it yet. If not, a software program like he's talking about would be a great thing!

hueseph Mon, 07/11/2011 - 11:21

Lambchop: your problem likely has more to do with the video conversion and framerate thn with the audio stream. You need to ensure that you are recording at 24 bit / 48K and that youhave the framerate in Vegas to 30fps or whatever the standard framerate happens to be in your region.

Sent from my iPhone.

EDIT: I just realized you are talking about something completely different. I thought you were talking about drift but obviously not.

hueseph Mon, 07/11/2011 - 13:31

Ok. Maybe I'm confused. Video is not as standard as audio. Audio is consistent pretty well anywhere you go. Video however has different formats(PAL, NTSC etc.) as well frame rates can be different from region to region. To add to that video is ever so slightly faster than standard audio. We're talking 1 or 2 milliseconds if that. But, that time adds up and after five minutes you're way off sync. Vegas probably uses MTC as a standard clock which I believe is embedded in the audio. Otherwise you could encode with smpte which is a pain IMHO. To sum it up, setting the frame rate and using he appropriate sample rate allows Vegas to compensate. Someone will probably step in to correct me soon.

bouldersound Mon, 07/11/2011 - 16:33

hueseph, post: 373895 wrote: Ok. Maybe I'm confused. Video is not as standard as audio. Audio is consistent pretty well anywhere you go. Video however has different formats(PAL, NTSC etc.) as well frame rates can be different from region to region. To add to that video is ever so slightly faster than standard audio. We're talking 1 or 2 milliseconds if that. But, that time adds up and after five minutes you're way off sync.

Video isn't inherently faster than audio, but if you record it at the standard NTSC rate of 29.97fps and play it back at 30fps it would run fast. Vegas lets you choose all that in the project options and on individual files with varying results. I believe if you choose it at the project level it won't change the playback speed per se, but it will infer additional frames every so often, combining two adjacent frames to create a new one. If you set the frame rate at the file level to something other that its original recorded rate it will play fast or slow.

conleec Wed, 01/11/2012 - 12:57

Okay, that was fast. I'm impressed. I think the software you were referring to earlier on the video side of the spectrum might have been Plural-Eyes.

While I might at some point have a need for your software, what I would love somebody to write is a competitor to Titan. I need a way to replace the 16 bit clips from my Avid AAFs or OMFs and rebuild the sequence with the 24 bit multi-wavs from the production mixer. Anybody know of an alternative to Titan, or is that the only game in town.

I'm using ProTools, by the way.

Chris

x

User login