Bilingual video with backtrack in sidechain, issue

Get answers to your questions about color grading, editing and finishing with DaVinci Resolve.
  • Author
  • Message
Offline

BartReynaard

  • Posts: 454
  • Joined: Thu Jul 08, 2021 8:31 pm
  • Real Name: Bart

Bilingual video with backtrack in sidechain, issue

PostThu Sep 16, 2021 8:25 pm

I have this issue to solve and I need your help

I'm making bilingual videos. I mean one export is one language (English)
The second export is second language, Dutch
Each language has a compressor with SideChain SEND activated, for ducking

I have a music backtrack common to both languages, with a compressor, with also sidechain RECEIVE activated, so it can duck

My configuration is now:

Fairlight Timeline:
A1 (English Dialogue) > Bus 1 (English Mix)
A2 (Dutch Dialogue) > Bus 2 (Dutch Mix)
A3 (Music) > Bus 1 (English Mix)
A3 (Music) > Bus 2 (Dutch Mix)

Deliver page:
Render Settings A: Audio tab > Output Track = Bus 1 (English Mix)
Render Settings B: Audio tab > Output Track = Bus 2 (Dutch Mix)

I received as help in tis topic: viewtopic.php?f=21&t=145096#p777827

Now my problem is: the ducking is affected by both languages , not by
1st export English only
2nd export by Dutch only

I tried (as naif) to send the same track to 2 different busses, then to instantiate a compressor on each bus to be supplied by 2 different chains.
But:
• I discovered sidechain is only one. It seems there is not any way to have two independent chains.
• Compressors on busses, have not any sidechain feature

In conclusion: how can I achieve what I need?

Should I just mute the track I don't need at each export?
Or there is a way to do it, that's more efficient?
Bart
________
DaVinci Resolve Studio 18.6.5 on Windows 11 Pro 23H2

Prod. Machine: Ryzen 7 64GB 3.54 GHz
Video: MSI GeForce 4060Ti 16GB
Monitors: 1 x 3840x1440 widescreen, 2 x 1920 x 1080 16:9
Audio: Behringer UMC1820
Blackmagic Speed Editor
Offline

Peter Chamberlain

Blackmagic Design

  • Posts: 13875
  • Joined: Wed Aug 22, 2012 7:08 am

Re: Bilingual video with backtrack in sidechain, issue

PostFri Sep 17, 2021 4:30 am

moved to resolve forum
DaVinci Resolve Product Manager
Offline

Reynaud Venter

  • Posts: 4910
  • Joined: Sun May 28, 2017 9:34 am

Re: Bilingual video with backtrack in sidechain, issue

PostFri Sep 17, 2021 6:03 am

Automate the ducking riding the fader.

Alternatively, use Preview automation.

Preview is both enablement and isolation for scene-based workflows, where the control (e.g. fader) is not controlled by the automation system but remains under manual control while Preview is active.

The current value may then be committed to the defined Range, or controls enter Write or Trim as the automation system goes in to record.

Fairlight menu > Automation > Fill Range
Writes the current parameter's value (in this case fader level) across the range specified.

Fairlight menu > Automation > Glide Range
Glides from the automation data value at beginning of the specified range and glides to the current control's value at the end of the range - useful for creating smooth transitions between levels.

Suggest assigning keybindings to these functions as none are assigned by default.
Offline

BartReynaard

  • Posts: 454
  • Joined: Thu Jul 08, 2021 8:31 pm
  • Real Name: Bart

Re: Bilingual video with backtrack in sidechain, issue

PostFri Sep 17, 2021 8:29 am

Reynaud Venter wrote:Automate the ducking riding the fader.

Alternatively, use Preview automation.

Preview is both enablement and isolation for scene-based workflows, where the control (e.g. fader) is not controlled by the automation system but remains under manual control while Preview is active.

The current value may then be committed to the defined Range, or controls enter Write or Trim as the automation system goes in to record.

Fairlight menu > Automation > Fill Range
Writes the current parameter's value (in this case fader level) across the range specified.

Fairlight menu > Automation > Glide Range
Glides from the automation data value at beginning of the specified range and glides to the current control's value at the end of the range - useful for creating smooth transitions between levels.

Suggest assigning keybindings to these functions as none are assigned by default.


ok but writing the Fader Automation, it will be the same for both languages. As it is notw with the compressor.
And for instance, when in English I get silence and in Dutch is continuing talking (let's say 1-2 seconds longer than English), then the music present in English version is quieter because Dutch driven.

As it is now.

In other words, I can't load in the renderer queue the 2 instances and render them at once.
I have always to render one language first with the second one muted, then mute the first one, and render the second one.

it would be nice to avoid this manual step but if I have not an alternative, it' shte only oen way to go.
Bart
________
DaVinci Resolve Studio 18.6.5 on Windows 11 Pro 23H2

Prod. Machine: Ryzen 7 64GB 3.54 GHz
Video: MSI GeForce 4060Ti 16GB
Monitors: 1 x 3840x1440 widescreen, 2 x 1920 x 1080 16:9
Audio: Behringer UMC1820
Blackmagic Speed Editor
Offline

Reynaud Venter

  • Posts: 4910
  • Joined: Sun May 28, 2017 9:34 am

Re: Bilingual video with backtrack in sidechain, issue

PostFri Sep 17, 2021 12:21 pm

BartReynaard wrote:ok but writing the Fader Automation, it will be the same for both languages. As it is notw with the compressor.
And for instance, when in English I get silence and in Dutch is continuing talking (let's say 1-2 seconds longer than English), then the music present in English version is quieter because Dutch driven.
With a correctly configured template automating a dipped M&E is very straightforward, while still delivering an undipped M&E within the same pass for reversioning, along with the various programme feeds.

Very basic example illustrating the concept:
Offline

BartReynaard

  • Posts: 454
  • Joined: Thu Jul 08, 2021 8:31 pm
  • Real Name: Bart

Re: Bilingual video with backtrack in sidechain, issue

PostFri Sep 17, 2021 1:30 pm

Reynaud Venter wrote:
BartReynaard wrote:ok but writing the Fader Automation, it will be the same for both languages. As it is notw with the compressor.
And for instance, when in English I get silence and in Dutch is continuing talking (let's say 1-2 seconds longer than English), then the music present in English version is quieter because Dutch driven.
With a correctly configured template automating a dipped M&E is very straightforward, while still delivering an undipped M&E within the same pass for reversioning, along with the various programme feeds.

Very basic example illustrating the concept:


I would like to be sure we are talking about the same thing. Because I've difficulties to understand why you sent a video over Pro Tools, when I use Da Vinci for that purpose. Da Vinci should deliver a multilingual audio with independent backtrack ducking.

If I have to use the DAW to do this, I can but it's a further passage more because documentaries are synced into Da Vinci. And each language should duck the Backtrack independently.

Please here an example of 2 spoken tracks and a backtrack (as I said to be sure we are aligned on the concept)

DaVinci Duck.JPG
DaVinci Duck.JPG (70.93 KiB) Viewed 703 times


Because both spoek tracks (above) are ducking the backtrack, by automating the fader OR with the compressor, as result, into the highlighted moment, in the export of the video 2 (audio language 2 (second spoken track), I get quieter backtrack because driven but the audio 1

Hence I have to crate to fully separated exports:

One with the language 2 muted, a second one with the language 1 muted.

I can't automate this process in the queue and then launch both queued videos for rendering.

I need to care it manually one by one.

P.S. I'm not sure if I was able to explain myself univocally. Sorry if it's not.
Bart
________
DaVinci Resolve Studio 18.6.5 on Windows 11 Pro 23H2

Prod. Machine: Ryzen 7 64GB 3.54 GHz
Video: MSI GeForce 4060Ti 16GB
Monitors: 1 x 3840x1440 widescreen, 2 x 1920 x 1080 16:9
Audio: Behringer UMC1820
Blackmagic Speed Editor
Offline

Reynaud Venter

  • Posts: 4910
  • Joined: Sun May 28, 2017 9:34 am

Re: Bilingual video with backtrack in sidechain, issue

PostFri Sep 17, 2021 1:40 pm

BartReynaard wrote:I would like to be sure we are talking about the same thing. Because I've difficulties to understand why you sent a video over Pro Tools, when I use Da Vinci for that purpose. Da Vinci should deliver a multilingual audio with independent backtrack ducking.
Watch the video and follow the concepts involved - they apply to any audio workstation.

You are learning a new skill, so take the time and understand the concepts first.
Then work through and replicate the workflow in Resolve once you have understood.

This will stand you in good stead going forward and this type of workflow will become second nature.
Offline

BartReynaard

  • Posts: 454
  • Joined: Thu Jul 08, 2021 8:31 pm
  • Real Name: Bart

Re: Bilingual video with backtrack in sidechain, issue

PostFri Sep 17, 2021 3:47 pm

Reynaud Venter wrote:
BartReynaard wrote:I would like to be sure we are talking about the same thing. Because I've difficulties to understand why you sent a video over Pro Tools, when I use Da Vinci for that purpose. Da Vinci should deliver a multilingual audio with independent backtrack ducking.
Watch the video and follow the concepts involved - they apply to any audio workstation.

You are learning a new skill, so take the time and understand the concepts first.
Then work through and replicate the workflow in Resolve once you have understood.

This will stand you in good stead going forward and this type of workflow will become second nature.


Hi Reynard. I apply it with my DAW as well. But with the DAW I have another behaviour.

What I do normally in the DAW is:
- Put a native compressor on the backtrack (which does its job)
- Route both audio tracks to the Sidechain of the compressor of that track
- Playback just to check the settings and to tweak parameters (threshold, ratio, attack, release etc)
- Mute the no wanted track Language
- Render > Add to render queue
- Unmute the previous track, mute the second audio Language
- Render > Add to render queue
- Render > Render queue > Render all files

And I get the renderings i needed with the applied ducking as "from the book".
I tried the same in DaVinci, having Fairlight as known DAW, thinking/believing it can do the same.

It doesn't. It doesn't consider the status of the tracks (mute/solo, etc), when you put on Queue

i had to find an alternate way (see the post I linked on my opening thread).
I export 2 videos with 1 click, both with their own language. Fine.
The ducking, is not working as it should, as it does with the DAW.

having said that, later I will watch also the video (it's 47 minutes) and I will see if I can catch what you meant to tell to me.

I just wanted to be sure we are on the same line :-)

I will let you know after the vision.
Bart
________
DaVinci Resolve Studio 18.6.5 on Windows 11 Pro 23H2

Prod. Machine: Ryzen 7 64GB 3.54 GHz
Video: MSI GeForce 4060Ti 16GB
Monitors: 1 x 3840x1440 widescreen, 2 x 1920 x 1080 16:9
Audio: Behringer UMC1820
Blackmagic Speed Editor
Offline

Reynaud Venter

  • Posts: 4910
  • Joined: Sun May 28, 2017 9:34 am

Re: Bilingual video with backtrack in sidechain, issue

PostFri Sep 17, 2021 4:32 pm

BartReynaard wrote:i had to find an alternate way (see the post I linked on my opening thread)
The guidance I provided in that thread offers the option of employing a pre-fader Bus Send with a Bus Output from a single source - the send levels of which are both automatable.

Preview automation streamlines the process for fast turn around workflows and does not require extensive knowledge of the operator where a dedicated audio engineer is not available.
Offline

BartReynaard

  • Posts: 454
  • Joined: Thu Jul 08, 2021 8:31 pm
  • Real Name: Bart

Re: Bilingual video with backtrack in sidechain, issue

PostFri Sep 17, 2021 5:53 pm

Reynaud Venter wrote:
BartReynaard wrote:i had to find an alternate way (see the post I linked on my opening thread)
The guidance I provided in that thread offers the option of employing a pre-fader Bus Send with a Bus Output from a single source - the send levels of which are both automatable.

Preview automation streamlines the process for fast turn around workflows and does not require extensive knowledge of the operator where a dedicated audio engineer is not available.


Indeed your suggestion is valid. I applied it till now for ease
Unfortunately just DaVinci is behaving a bit differently.

If on the BUSes i could use a compressor with sidechain (independent from the previous one): game solved.
Appling your suggestion, was the GO solution forever.
But Fairlight is "limping" ....

The video you posted, told many things I know already about the delivery of documentaries with different languages and so on. Although it's different from my case, and it doesn't offer, to me, the solution I thought it could be possible.
Sure it is an excellent guide for who has that specific needs, but it's bit different from my specific use case.

BTW Prèmiere, I mentioned in the early messages here, is worse. Adobe introduced the Ducking with version 2017 or 2018 (so far I remember), which is barely sufficient. Not what I need, though.

I hope Blackmagic will improve this aspect in next releases, introducing features that will enable this specific use case.

There are 2 ways I can see:
adding compressor that supports Sidechain on busses with more than only one sidechain bus
OR
when you put in mute one track on Fairlight and you put on QUEUE the footage, that track is considered muted (now it is not).


[EDIT]
IFI had compressors with sidechain on busses
AND
IF I had 2 separate sidechain busses, following your suggestion I could:

  • Send the backtrack to 2 busses (Language A and Language B
  • Instantiate Compressor with Sidechain 1 to bus A
  • Instantiate Compressor with Sidechain 2 to bus B
  • Instantiate Compressor to track Language A Sidechain 1
  • Instantiate Compressor to track Language B Sidechain 2
  • Create Template with

So during each Delivery I could limit my actions to (which is your suggestion from the other topic):
  • Get the video preset
  • Select the audio BUS (on Audio Tab): Bus Language A
  • Add to queue video A (Language A)
  • Select the audio BUS (on Audio Tab): Bus Language B
  • Add to queue video B (Language B)
  • Click on RENDER

It could make life and workflow way efficienter than now, and less prone to errors.
Bart
________
DaVinci Resolve Studio 18.6.5 on Windows 11 Pro 23H2

Prod. Machine: Ryzen 7 64GB 3.54 GHz
Video: MSI GeForce 4060Ti 16GB
Monitors: 1 x 3840x1440 widescreen, 2 x 1920 x 1080 16:9
Audio: Behringer UMC1820
Blackmagic Speed Editor

Return to DaVinci Resolve

Who is online

Users browsing this forum: 4EvrYng, allshedreamed, ExzxImages, Google [Bot], mpetech, panos_mts, psh_vt, Robert Niessner, Tom Stites and 165 guests