0
$\begingroup$

Input: several air audio recordings of a live concert taken from different locations within the venue. Different quality, different noises (e.g. voices of people in the vicinity of the recording device).

Are there tools that can take the input and automatically produce the cleanest possible audio of the concert, minimizing audience noises?

This question comes close to the following ones, but, unlike them, is focused on practical tools which can be used without deep understanding of the underlying theory:

Improve this question
$\endgroup$
3
  • $\begingroup$ There are a lot of factors that affect the quality of the outcome of such an operation. It is not something that can be generalised easily. You could try importing the tracks into Audacity, align the tracks based on a common "event", select the tracks and then "Tracks/Mix and render to new track" and see what you get. You will need 4-5 tracks at least. Expect better results at the lower end of the spectrum. $\endgroup$ Commented Jun 10, 2024 at 8:57
  • $\begingroup$ @A_A Won't that simply mix them together without trying to identify/remove noise? $\endgroup$ Commented Jun 10, 2024 at 19:09
  • $\begingroup$ Yes, but assuming that the crowd produces random sounds that are picked up locally, averaging over a number of recordings will boost the common parts (that is, the sound of the band). That is the theory anyway. In practice, you will get artefacts due to slight temporal mismatches. There are various ML models that can help with audience noise, but you would have to replicate the process across tracks and the result would still depend on conditions. $\endgroup$ Commented Jun 10, 2024 at 19:35

0

Reset to default

Start asking to get answers

Find the answer to your question by asking.

Ask question

Explore related questions

See similar questions with these tags.