This is my first post and have withstood the urge of creating an account for audacity for the longest while. First go-to is usually reddit or 4chan. But since they couldnt help me and because I predominately use Audacity it made sense in the end to come here.
I am self taught. Started beginning of this year. I have dabbled in spectral analysis and OpenVinoAI etc. My recording device is a Tascam which has alot of internal electronic noise, but does its job fairly well. I have a few recordings which I need cleaned up and would like help if you could clean this 10 second sample I’m about to give and tell me your cleaning process step-by-step so I can apply the same changes to the entire 2 hour audio.
Overview:
It is a female commentator speaking. Language does not need to be disclosed. I would just love her voice to shine through the background noise and have her words be clear as possible.
So are you telling me that the entire audio is useless and nothing can be salvaged OR are you telling me to provide a lengthier sample? How much clear vocal audio is needed for someone to be able to identify/ rescue the other vocals in the same noisy audio?
So there is no way of reducing that noise?
4.8 - 6.2s is where the vocals are. She is saying “Er isch peder”
Language is a mix of Swiss German and slavic.
What he said. That’s unrecognizable as a voice track.
There is a Hollywood version of this. If the original track is too badly damaged, we get an actor with a similar voice to perform the same words in sync in a quiet, echo-free studio. If you can’t write down what the words were from the original performance, then you have identified the problem.
Okay so providing you the noise profile of my device wouldnt help in this situation either… is that what you are saying?
I just uploaded a pure noise profile my device makes when recording. Same link
