CyberLink Community Forum
where the experts meet
| Advanced Search >
synching multiple inputs
Reply to this topic
ClarkNK [Avatar]
Newbie Private Message Joined: Apr 21, 2020 09:28 Messages: 13 Offline
[Post New]
I am trying to create a collage virtual quartet. I sent each person an mp3 piano file to use as accomaniement as they record their individual video.

So now I have 4 videos, each of a person singing along to a piano music file being played. As each video is being recorded by an Iphone, both the piano music being played through speakers, and the singing are being recorded.

So I wind up with four mp4 video files, each containing audio that is a combination of voice and the piano accompaniement..

I then import the four videos, place each in a trrack, highlight the four audo tracks, and click the synch by audio button.

When I tried this with only two tracks it seemed to work, but it seems not to work with four tracks. I might get three of them synched if i am lucky, but not all four. I realize it is asking quite a bit for the software to synch multiple audio tracks when the audio tracks are composed of a combination of the piano music mixed with different voices.

I am new at this What approach should I be taking here? And suggestions much appreciated.

Thanks
Reply
tomasc [Avatar]
Senior Contributor Private Message Joined: Aug 25, 2011 12:33 Messages: 5066 Offline
[Post New]
Quote So I wind up with four mp4 video files, each containing audio that is a combination of voice and the piano accompaniement..

I then import the four videos, place each in a trrack, highlight the four audo tracks, and click the synch by audio button.

When I tried this with only two tracks it seemed to work, but it seems not to work with four tracks. I might get three of them synched if i am lucky, but not all four. I realize it is asking quite a bit for the software to synch multiple audio tracks when the audio tracks are composed of a combination of the piano music mixed with different voices.

Since you sent identical mp3 files to each individual you should line up the beginning piano notes before the singing starts and then let everything fall in place. The piano part should be consistent in time length if you sent them all a constant and not a variable bit rate mp3.

There is no need to use sync by audio since each vocal is different. This is not multiple audio recordings of the same event but each recording is a separate and not identical event.
Reply
ClarkNK [Avatar]
Newbie Private Message Joined: Apr 21, 2020 09:28 Messages: 13 Offline
[Post New]
Quote

Since you sent identical mp3 files to each individual you should line up the beginning piano notes before the singing starts and then let everything fall in place. The piano part should be consistent in time length if you sent them all a constant and not a variable bit rate mp3.

There is no need to use sync by audio since each vocal is different. This is not multiple audio recordings of the same event but each recording is a separate and not identical event.


Yes, thank you. I was thinking along those lines as a fall-back if I could not get the synch by audio to work. To use that method, do you just greatly expand the tracks and then try to match some early peak in the audio waveform?
Reply
[Post New]
Quote


Yes, thank you. I was thinking along those lines as a fall-back if I could not get the synch by audio to work. To use that method, do you just greatly expand the tracks and then try to match some early peak in the audio waveform?


Yes, that would do it, then It's just a trial and error to move the track(s) back or forward a frame at a time to clean up the audio sync.
What a great idea, great fun.

Cheers
Reply
tomasc [Avatar]
Senior Contributor Private Message Joined: Aug 25, 2011 12:33 Messages: 5066 Offline
[Post New]
Quote Yes, thank you. I was thinking along those lines as a fall-back if I could not get the synch by audio to work. To use that method, do you just greatly expand the tracks and then try to match some early peak in the audio waveform?

You could try this: Append a clapperboard sound to the beginning and end of that constant bit rate mp3 and you won’t have to worry about a tedious syncing. This will also give you more knowledge and information when working with the iPhone.

I do recall that the iPhone records in variable frame rates. The lighting can affect those frame rates a lot if each session with a vocalist was recorded with a different lighting.

The reason for adding the clapper sound at the end of the mp3 is to allow you to easily see the video length of each session which can change. The better the equipment you use the better the result. You want everything to be as constant as you can. If you already recorded the four vocals and don’t wish to start again then just do the best you can. Hope this help…

Edit: The forum timed out with 403 Forbidden.
Reply
PowerDirector Moderator
Senior Contributor Private Message Location: New Taipei City, Taiwan Joined: Oct 18, 2016 00:25 Messages: 793 Offline
[Post New]
Quote

I am new at this What approach should I be taking here? And suggestions much appreciated.

Thanks


Hi,
I'd be interested how it turns out this way.

I see a potential problem in that you have 4 recordings of both vocal and backing. So, depending on each member's recording environment, you will be working with 4 different audio "qualities". By that I mean,, room ambience, playback systems, etc. not necessarily technical quality (but that too may play a part).

For vocals only that may well be part of the effect wanted and the vocals can be edited to bring out or exclude varying parameters to your requirements, but by having the combined vocal and backing you have to combine 4 different voices and 4 "different" versions of the same backing. It may turn out to be as required, but it may turn out to be not that way.

A simple example:
If you take a piece of music and play it back and record it in a bathroom, a lounge area and an aircraft hangar you will get 3 very different sounding pieces. If you then combine them into one track it may turn out to be great - or maybe not!

So it will be interesting to see/hear how you get on. Perhaps you could post a link to your work in the Showcase Forum when you are done? https://forum.cyberlink.com/forum/forums/show/183.page

Cheers
PowerDirector Moderator
Reply
ClarkNK [Avatar]
Newbie Private Message Joined: Apr 21, 2020 09:28 Messages: 13 Offline
[Post New]
Quote


Hi,
I'd be interested how it turns out this way.

I see a potential problem in that you have 4 recordings of both vocal and backing. So, depending on each member's recording environment, you will be working with 4 different audio "qualities". By that I mean,, room ambience, playback systems, etc. not necessarily technical quality (but that too may play a part).

For vocals only that may well be part of the effect wanted and the vocals can be edited to bring out or exclude varying parameters to your requirements, but by having the combined vocal and backing you have to combine 4 different voices and 4 "different" versions of the same backing. It may turn out to be as required, but it may turn out to be not that way.

A simple example:
If you take a piece of music and play it back and record it in a bathroom, a lounge area and an aircraft hangar you will get 3 very different sounding pieces. If you then combine them into one track it may turn out to be great - or maybe not!

So it will be interesting to see/hear how you get on. Perhaps you could post a link to your work in the Showcase Forum when you are done? https://forum.cyberlink.com/forum/forums/show/183.page

Cheers
PowerDirector Moderator



Yes, you have stated the problem quite well. I figured from the beginning that the biggest problem in the whole approach would be whether the software behind the "Synch by Audio" button was sophisticated enough to handle four tracks that were each a mashup of piano and vocal.

I did manage to get the four tracks synched, but it was not a slam-dunk. There was enough difference in the starting times of each track (each person took a slightly different amount of time between when they pushed the button to start their video recording and when they pushed the button to start the piano music file they were going to sing with) that I had to manually trim the tracks to get the start times fairly close to each other. After doing that, the "Synch by Audio" worked. It would be very interesting to me to know what kind of algorithm was used to accomplish that -- it has to be one bit of impressive software!

This was just done by four of us amateurs from my church choir, unexceptional voices, and whatever phones we had to record the videos. Three of us used phones, one used a webcam. And one of the phone recordings was done with the piano music being played from only laptop speakers! I played the piano track on my computer speakers, and I don't know how two of them played the piano music file. So , bottom line, questionable equipment all the way.

I had promised all of them that the result would not see the light of day without their permission. But if they are ok with my offering it up as music for one of our virtual church services, I will share it here.
Reply
ClarkNK [Avatar]
Newbie Private Message Joined: Apr 21, 2020 09:28 Messages: 13 Offline
[Post New]
Quote



Yes, you have stated the problem quite well. I figured from the beginning that the biggest problem in the whole approach would be whether the software behind the "Synch by Audio" button was sophisticated enough to handle four tracks that were each a mashup of piano and vocal.

I did manage to get the four tracks synched, but it was not a slam-dunk. There was enough difference in the starting times of each track (each person took a slightly different amount of time between when they pushed the button to start their video recording and when they pushed the button to start the piano music file they were going to sing with) that I had to manually trim the tracks to get the start times fairly close to each other. After doing that, the "Synch by Audio" worked. It would be very interesting to me to know what kind of algorithm was used to accomplish that -- it has to be one bit of impressive software!

This was just done by four of us amateurs from my church choir, unexceptional voices, and whatever phones we had to record the videos. Three of us used phones, one used a webcam. And one of the phone recordings was done with the piano music being played from only laptop speakers! I played the piano track on my computer speakers, and I don't know how two of them played the piano music file. So , bottom line, questionable equipment all the way.

I had promised all of them that the result would not see the light of day without their permission. But if they are ok with my offering it up as music for one of our virtual church services, I will share it here.


An update for anyone following this. I am up to five individual tracks recorded as above. Because I was worried about the ability to Synch by Audio all those tracks, I had the pianest do a loud hand clap before starting to play, to use as a backup method of synchronization. To make a long story short, I was unab.e to syncn five tracks using Synch by Audio, so I did resort to manually aligining the sharp blips of the handclap in the audio tracks. That did work.

This message was edited 1 time. Last update was at May 22. 2020 22:15

Reply
[Post New]
Quote


An update for anyone following this. I am up to five individual tracks recorded as above. Because I was worried about the ability to Synch by Audio all those tracks, I had the pianest do a loud hand clap before starting to play, to use as a backup method of synchronization. To make a long story short, I was unab.e to syncn five tracks using Synch by Audio, so I did resort to manually aligining the sharp blips of the handclap in the audio tracks. That did work.


Well done on managing to do a manual sync, it's not the easiest way but sometimes you have to resort to this method.

Just a little question... Did you trim ALL the clips to just a few second around the handclap, THEN use PD to sync up. After the sync has been performed ( it's usually quite quick when using this method !) you just pull out he tails of the clips to there full length, it usually stays in sync for the entire length. I've had to use this method for long-ish clips.

thanks for keeping us updated
Reply
ClarkNK [Avatar]
Newbie Private Message Joined: Apr 21, 2020 09:28 Messages: 13 Offline
[Post New]
Quote


Well done on managing to do a manual sync, it's not the easiest way but sometimes you have to resort to this method.

Just a little question... Did you trim ALL the clips to just a few second around the handclap, THEN use PD to sync up. After the sync has been performed ( it's usually quite quick when using this method !) you just pull out he tails of the clips to there full length, it usually stays in sync for the entire length. I've had to use this method for long-ish clips.

thanks for keeping us updated


Not sure I completely understood what you are saying, What I did was expand the timeline to a point where I could identify the clap on each timeline, and then I manually aligned the claps. Having alinged the tracks, I then trimmed the beginning away and moved all the tracks to the start point.

Then I produced each track separately for use in a subsequent collage, now that the individual produced tracks had the same starting point.
Reply
Reply to this topic
Powered by JForum 2.1.8 © JForum Team