ProSoundWeb Community
Sound Reinforcement - Forums for Live Sound Professionals - Your Displayed Name Must Be Your Real Full Name To Post In The Live Sound Forums => Pro AV Forum => Topic started by: Cass Johnson on June 10, 2022, 02:01:35 PM
-
We recently recorded a Live Studio Talk Show (which can't be re-filmed) although we had basic "In Camera" audio and it wasn't that good. Our sound tech recorded the show through two wireless lavaliers into PreSonus "Studio One" on a laptop. When the tracks were provided to me they are 1:55 (one minute-fifty five seconds) longer than the video and in camera audio. HUGE problem - the audio is crystal clear but I have no idea if there is anything I can do in order to reduce the overall timing to fit the video. I really hope someone can direct me to a fix or the entire show is a wash.
-
We recently recorded a Live Studio Talk Show (which can't be re-filmed) although we had basic "In Camera" audio and it wasn't that good. Our sound tech recorded the show through two wireless lavaliers into PreSonus "Studio One" on a laptop. When the tracks were provided to me they are 1:55 (one minute-fifty five seconds) longer than the video and in camera audio. HUGE problem - the audio is crystal clear but I have no idea if there is anything I can do in order to reduce the overall timing to fit the video. I really hope someone can direct me to a fix or the entire show is a wash.
You need to go into an NLE (like Resolve)and do a little editing. Align the audio, then clip off the extra. Or if they are actually in a different time scale, you can stretch or shrink the video or audio track to fit.
-
We recently recorded a Live Studio Talk Show (which can't be re-filmed) although we had basic "In Camera" audio and it wasn't that good. Our sound tech recorded the show through two wireless lavaliers into PreSonus "Studio One" on a laptop. When the tracks were provided to me they are 1:55 (one minute-fifty five seconds) longer than the video and in camera audio. HUGE problem - the audio is crystal clear but I have no idea if there is anything I can do in order to reduce the overall timing to fit the video. I really hope someone can direct me to a fix or the entire show is a wash.
As Dave mentioned, if it's an actual clock drift issue, many modern DAWs will allow you to time stretch audio. I do not know if Studio One does. All allow you to trim/shift to adjust timing.
In Reaper, you can import the video along with every audio track and either stretch (shrink) the audio tracks as a whole or add "stretch markers" at intervals to the out-of-sync audio then manually align it with the audio/video track.
Most video editors have multi-track sync capability but I'm unsure if any stretch automatically. I think most just shift to align.
Reaper, Resolve, Premiere, Final Cut Pro, maybe even iMovie can be used to fix your issue.
-
Could be the difference in frame rates, 30 fps vs 29.97...
-
How long is the total video?
If there are any cut away/b roll shots getting cut into the finished video those are good places
to slip the audio track to adjust the sync.
-
Sounds like to me the sound operator just either started recording earlier than the camera guy, or he stopped recording later than the camera guy. Or perhaps a little of both and that equals the 1 minute and 55-second differential in length between the two tracks. What I understand from the OP is one audio track came from an internal mic on the camera or else an external mic plugged into the camera, and a second track came from two wireless mics directly into basically a DAW. So the "in-camera" audio is already embedded in the video, but then there exists a separate audio track that was from other mics straight to basically a DAW. In that situation I would expect the length of the two tracks to almost never be the same. It would take a small miracle if both recordings started and stopped at the identical same instant.
Compare the lead-in portions of both audio tracks (recorded and in-camera) and see if there is some period of dead time in the recorded track. I'm betting there is. So in a DAW import both tracks and find a convenient "first good sound" point and line the tracks up by shifting the recorded track left or right as necessary. Once you get them properly aligned, you can trim the recorded track to match the length of the in-camera audio track.
-
Compare the lead-in portions of both audio tracks (recorded and in-camera) and see if there is some period of dead time in the recorded track. I'm betting there is. So in a DAW import both tracks and find a convenient "first good sound" point and line the tracks up by shifting the recorded track left or right as necessary. Once you get them properly aligned, you can trim the recorded track to match the length of the in-camera audio track.
Like I said. Resolve (it's free) has a one button audio align tool that is amazing.
-
I use Vegas Pro 18 for video editing and audio recording/mixing. It would be a fairly trivial matter to "resize" the audio to fit the video. It's essentially a resampling process. I would drag the beginning and end of the audio to clear landmarks that I can reference to the camera audio, line the first one up then stretch/squeeze (CTRL-drag to engage Time Stretch) the audio clip so the end landmarks lined up. Then I'd drag the beginning and end back out if needed. This is where having an audio slate at the beginning and end of the take is helpful.
Most likely there was a discrepancy between the clocks. For example, the clock on the Zoom H4 and H4n (non Pro models) is notorious for drifting when set to 44.1 kHz, but usually synced up well with other digital devices at 48 kHz.
-
This is an issue when one file is recorded as VBR (variable bitrate) and the other is record as CBR (constant bitrate).
I ran sound for a wedding and recorded the audio from the console on my Tascam DR-40, which records in CBR. My mother-in-law captured the event on her iPhone, which recorded in VBR.
The files end up tracking differently and I have to match them up in post- using time stretching/shrinking.
-
This is an issue when one file is recorded as VBR (variable bitrate) and the other is record as CBR (constant bitrate).
The files end up tracking differently and I have to match them up in post- using time stretching/shrinking.
VBR and CBR encodings using the same type of codec (mp3, aac) should align perfectly if the hardware clocks are in sync. If they do not, something is broken in the encoder or decoder. CBR is just easier for software to determine position and length due to the fixed number of bytes per time period. Some codecs add samples to the beginning of the file, shifting the audio slightly but that shift should be consistent for files encoded with the same codec type.
-
VBR and CBR encodings using the same type of codec (mp3, aac) should align perfectly if the hardware clocks are in sync. If they do not, something is broken in the encoder or decoder. CBR is just easier for software to determine position and length due to the fixed number of bytes per time period. Some codecs add samples to the beginning of the file, shifting the audio slightly but that shift should be consistent for files encoded with the same codec type.
It's quite possibly a frame rate mismatch as has been mentioned.
-
It's quite possibly a frame rate mismatch as has been mentioned.
It wouldn't be a frame rate mismatch unless one was being played back at a different frame rate that it was recorded in. The issue is how long a particular clock thinks a second is, not how many frames or samples it puts in that second. If one clock measures a second a little differently than another clock, the recordings will drift.
-
It wouldn't be a frame rate mismatch unless one was being played back at a different frame rate that it was recorded in. The issue is how long a particular clock thinks a second is, not how many frames or samples it puts in that second. If one clock measures a second a little differently than another clock, the recordings will drift.
But if the NLE thinks they're the same rate, then the sound will be off.
-
How long was the track? All I'm thinking is maybe it's a 44.1 vs 48K mismatch - not always evident on speech?
If you are stuck, then simply stretch/shrink the track in a DAW till it fits
-
But if the NLE thinks they're the same rate, then the sound will be off.
The laptop audio was off from the camera audio/video. That's the problem. A slight mismatch between the laptop clock (assuming it was an audio interface) and the camera's internal clock would explain it. They're recorded at slightly different rates then played back at the same rate.
-
This is an issue when one file is recorded as VBR (variable bitrate) and the other is record as CBR (constant bitrate).
I ran sound for a wedding and recorded the audio from the console on my Tascam DR-40, which records in CBR. My mother-in-law captured the event on her iPhone, which recorded in VBR.
The files end up tracking differently and I have to match them up in post- using time stretching/shrinking.
I actually believe it's a "Clock Drifting issue." My audio guy ran our wireless packs into a laptop and recorded with Studio One Artisan - The true sync version is from one of the cameras.
My software is Davinci Resolve Studio 17 and Studio One Artisan - My only option is to try to use studio one to match the clean audio (from the computer) to the audio captured with the camera.
I just don't know how to do that.
-
I use a product called PluralEyes which makes things like this a snap.
-
I actually believe it's a "Clock Drifting issue." My audio guy ran our wireless packs into a laptop and recorded with Studio One Artisan - The true sync version is from one of the cameras.
My software is Davinci Resolve Studio 17 and Studio One Artisan - My only option is to try to use studio one to match the clean audio (from the computer) to the audio captured with the camera.
I just don't know how to do that.
You should be able to do it in Resolve. You can adjust either the audio or the video.
-
You should be able to do it in Resolve. You can adjust either the audio or the video.
I'd do the audio. If you stretch/compress the video you either end up with double frames (if you shorten the video) or you have to interpolate frames (if you lengthen the video). If you do audio it's essentially just a sample rate conversion, which is fairly benign these days if you just do it once.
-
We recently recorded a Live Studio Talk Show (which can't be re-filmed) although we had basic "In Camera" audio and it wasn't that good. Our sound tech recorded the show through two wireless lavaliers into PreSonus "Studio One" on a laptop. When the tracks were provided to me they are 1:55 (one minute-fifty five seconds) longer than the video and in camera audio. HUGE problem - the audio is crystal clear but I have no idea if there is anything I can do in order to reduce the overall timing to fit the video. I really hope someone can direct me to a fix or the entire show is a wash.
I had that issue once
get the video edited and note exact length and start & stop points times
take audio and align correct start point with start of video start time
'stretch' audio to align stop points ending time
add audio to video
they should sync well enough that you cannot tell