by Davidk » Sat Jun 20, 2020 2:54 am
There is a method to "normalise" audio - make all audio the same level - in rendered outputs, but it's for disks. In the burn module step 3, the render output can be done with normalized audio (a tick box selection). What this does is make all audio in the selections the same level - which nullifies any intended soft whispers or loud verbal emphasis components, as well as variations in individual clip levels.
I can't see that this applies to a streaming video requirement; in fact, any editing would be impossible if it is real-time streaming. Broadcast (delayed streaming by another name) of an edited file is entirely possible.
You could try this approach: create rendered outputs to an iso file using normalised audio and assess it; the iso output file is playable by a number of reputable players (VLC usually recommended here).
Otherwise, use the sound mixer and adjust the sound of each clip to a preferred level, then render.
Alternatively, if you have mulitple camera outputs of the same scenes, you could consider using the the captured camera imagery in a multi-camera editing (MCE - icon option in the timeline tool bar) effort. Using MCE: a specific sound source is identified as the sound sync track, can be a video clip or a specific sound source like microphone on a lectern, and everything in the composition is synchronised to it.. One you have mixed/edited the clips, the sound is the selected one, thus any levels are as desired, and any sound in other clips is muted. There is a tutorial on MCE available.
