What kind of process should be at work to do the job. Clearly Noise Reduction may help slightly; what else? Is the sound heavily “compressed”? Perhaps some kind of decompression might be needed?
What kind of process should be at work to do the job.
Reshoot it?
If you’re close enough, pay to have it retransferred. I think can still get access to the prints.
It’s not compression or noise. It’s a bad transfer from old optical track and I’d be shocked if the fidelity made it much over 5KHz, so it’s always going to sound like AM radio. There’s holes in the sound I’m guessing a non-studio transfer of the print—or worse, a copy of a bad transfer. So tiny parts of the sound just aren’t there.
And the whole thing sluiced through YouTube compression.
So no. This isn’t “clean up.” This is disaster recovery and I would be reshooting it.
Many thanks for your help.
I notice the irritation seems to be affecting the actors / narrators voices, whereas the background music sounds reasonably good. Might there be some way to separate the voices from the music; use graphic equaliser to put more ‘timber’ back into the voice and then remarry the two separate tracks back together, (prior to dubbing the audio back to the video)?
You must forgive my ignorance; I am new to Audacity and audio restoration. I only really know two rituals: How to remove hiss and how to (erase) Silence parts from the audio.
I would like to learn a lot more, but finding just what Audacity is actually capable of doing is no easy feat!
On another film I worked on, I cleaned the hiss away and then as a test, I tried Compression. I compared the compressed track with an uncompressed version and was surprised to find there was no audible difference! I always thought Compression pulled loud sounds down and pushed weak sounds up but this was not obvious with my test.
I am always keen to learn, from a layman’s point of view; nothing too technical.
Does anyone have an answer to my last request, please?
I am wanting to separate the speech from the music, so that I have two copies of the sound; one with speech only, the other with music only.
Next: I want to try and better the speech track perhaps using the graphics equaliser.
Then play both speech track and music track back, in sync / unison in order to match them back together and render them as one wav file. Hopefully the youtube trailer I provided a link for, (see above), will sound a lot more coherent and better(?)
Can anyone describe the ritual or routine to actually do this, please? I need step by step instructions, how to.
Thank you.