@ohwell I can give it a try with the less obvious or distorted parts and see what comes of it.
What I did so far:
[details=Download the video source]* Download the actual source for each video.
Note: Video source URLs can be found below, but this is how you would get them using Google Chrome.
- Developer tools (F12)
- Network Tab (F5)
- Make video play and you will see the .mp4 appear in the list
- Right-click the mp4 and open in new tab > Download the source.[/details]
[details=Strip audio from video]* Get the audio from the video above.
- Media Tab
- Convert / Save … (CTRL+R)
- Select ‘Add…’ and browse for the mp4 (Open)
- Select ‘Convert’ from drop-down under ‘Convert / Save …’
- Under Settings, select the Profile in the drop-down menu
- Select an Audio profile, preferably FLAC (lossless)
- Browse for location to save to and give the file a name
- Click ‘Save’
- Click 'Start’
I now have the audio stripped from the video as a lossless FLAC.[/details]
[details=Create an Audacity Project]* Create a Project
- Open Audacity
- Drag each of the audio files you stripped from the video into Audacity (4 total so far)
Alternatively use ‘File Tab’ > ‘Open’
- Using the ‘Time Shift Tool’ (F5), I synced the audio of the first 3 files.
The first 3 are all of the same length and seem to match each other really well.
The 4th however appears to be quite different.
- Adding labels (CTRL+B)
For each track I then started selecting parts of voice as precise as possible (Zoom) and adding labels. This is really useful and will give you a good overview of each track. I used different formatting to indicate Elizabeth, distorted/other voice and static. Now I can just select labels and selection will easily snap to them as well.
- Select the ‘File’ tab.
- Select 'Save Project As…'
A warning will show that this is only going to be an Audacity project file and not being saved as actual audio.
- Give the project a name and hit ‘Save’ (.aup file)[/details]
[details=All video sources]
The above is what I am working with. Besides syncing, I have not applied any filters or made actual changes to the audio. If I wish to do so, I create a new track (Tracks > Add New) and mess with it there. Alternatively, I could make changes directly, but then just not saving the project, to make sure I have a clean project to work with.
Here is what my project looks like:
Anyways, this last video really confuses me in comparison to what we had received prior. We still have Elizabeth talking, but the ‘Hello’ at the start appears to be the only left-over from previous videos, when it comes to a clear message from Elizabeth. Everything else she has to say seems completely new. Of course we still have the distorted voice saying things that are hard to make out. This distorted voice has so far been in all of the videos and we can recognise a lot of similarities between the new and old videos. The total duration has changed as well, from the 31s they used to be, to a total of 45s now. However, the last part after we have had the ‘clear’ talk, starts with heavy static, to then be followed by ‘already known’ distorted voice with additional ‘echo’ applied. Removing this last section would bring the total duration back to about 30s again. Still something feels off, due to the message being so different from what we have heard and been able to puzzle back together again. This last message just seems … ‘new’.
Others have mentioned before that it seems like multiple messages mixed together, like @KCrosley here: https://forums.etarc.org/t/sigils-videos-elizabeth-video/339/14 and I agree. Maybe we are being side tracked by the more ‘clear’ voice of Elizabeth and we need to focus more on what the distorted voice actually has to say?
Let’s all give it a try and see what we can come up with? We now have so many parts to work with, that applying filters and/or other edits, may possibly give some ‘clear’ results?
Edit: Maybe spectral editing is the way to go, but I am personally reasonably new to this. It might be the ‘better’ tool to use to rid of unwanted parts to ‘clear’ up things.