7h • General discussion
Suno sonic cliff repair
There's a high frequency (~15KHz) cliff that appears in AI generated music, apparently because the generative engine has been trained on low quality MP3 type audio so hasn't learned anything above that frequency. It becomes part of the fingerprint in AI music detection algorithms and, some people say, it results in AI music all sounding the same. Anyway, I've been messing around to see if it's possible to repair that cliff to some extent. By selective saturation it's possible to add back frequencies that are harmonics of the instrumentation so, in theory at least, something useful should happen. The composite image (Cubase) below shows the results from setting up an Fx bus with high-level tube saturation and a high pass filter. Give it a go and let me know what you think. I'd be interested to hear if anyone has better solutions...other than recreating stems with real music! 🤔
0
4 comments
Suno sonic cliff repair
skool.com/mastering
This is a community for artists, producers & engineers who want to achieve pro quality in their music.
Leaderboard (30-day)
1
+121
2
+88
3
+61
4
+52
5
+47
Powered by