Voice isolation and leveler usage questions

Since the voice isolation and dialogue leveler don't work in a batch, what is the correct process for using it?

Here is an example of a project I am working on at the moment

Bob recorded in his bedroom - Soft and echo
Cloe recorded in furniture stores - Soft and noisy
Voice-over guy recorded - Level sound. Maximum volume, -2db
Customer interviews with a webcam - Soft and echo and muffled

What I had to do was to reduce the volume of the voice over guy clips to a volume of -4
I then increased the volume of my other clips between 6 and 10
I added voice isolation and leveler on each clip. I probably have 50 clips
I tried to get the sound to -5db throughout the video

This can't be the right process that I am following. Please give your input

Sound waveforms
https://imgur.com/a/TwytS7j

Dynamics
https://imgur.com/a/Xxfs7DN

Here is a quick video of the project. I jump between clips on the timeline and in the sound mixer to show my levels

Are the indicated deciBel values peak level?

Based on the screenshot peak levels are extremely high and signal is over-modulated.

No Clip Gain line is indicated - is Fairlight menu > Show Clip Gain Line disabled? Are those source levels?

Dialogue Leveller was designed to simplify the process of level adjustment without requiring Clip Gain, Automation, or Dynamics processing. With appropriate levels the Dialogue Leveller is not required.

The linked video is set to private.

What is the intended consumption platform for the content? Broadcast, socials?

I changed the video to unlisted.

The video will be used internal to a company for training.

I understand almost nothing what you are asking. Sorry I am still very dumb with fixing the sound in Davinci. I usually don't need to worry too much about my sound because I control the source.

Since the voice isolation and dialogue leveler don't work in a batch, what is the correct process for using it?

Here is an example of a project I am working on at the moment

Bob recorded in his bedroom - Soft and echo
Cloe recorded in furniture stores - Soft and noisy
Voice-over guy recorded - Level sound. Maximum volume, -2db
Customer interviews with a webcam - Soft and echo and muffled
...

A common approach is to 'Checkerboard' your audio by creating a separate track for each talent (speaker). Then you can adjust individual clips as necessary to get them into the ballpark of being acceptable and apply (optionally) dialog isolation and dialog leveller at the track level rather than clip level.

Thank you. I didn't know I could use that function on a track. I will look into it.

I didn't even think of moving the sound to different tracks. Thank you for that idea also.