| Stereo Tool https://forums.stereotool.com/ |
|
| Tape dropout & stereo repair. https://forums.stereotool.com/viewtopic.php?f=16&t=24051 |
Page 1 of 1 |
| Author: | dnz2dbt [ Thu Apr 12, 2018 9:24 am ] |
| Post subject: | Tape dropout & stereo repair. |
When r2r and cassettes get old and worn the tend to occasionally drop the audio in left or right or both, and I have yet to find a plug that listens intensly to the stereo field and corrects it so the worn bits are raised to the desired volume compared to the fully playing side, and you get a restored full stereo playback. Yes, the noise is also raised, but that could be addressed with noise cancelling of some sort, if desired. Sudden drops of all the audio is way too quick for the AGC so it needs to work at another threshold and speed. I suspect the noise floor of the tape would be the reference to where the levels are. That is the only constant as far as I know. I have done this many times manually in wavelab, but is timeconsuming to say the least. Some would say: Press mono. Problem solved. But I kind of enjoy a good stereo recording, even with dropouts. |
|
| Author: | Bojcha [ Thu Apr 12, 2018 5:28 pm ] |
| Post subject: | Re: Tape dropout & stereo repair. |
Just not possible to AUTO do that and preserve original. |
|
| Author: | dnz2dbt [ Thu Apr 12, 2018 7:01 pm ] |
| Post subject: | Re: Tape dropout & stereo repair. |
Well, I'm no programmer but isn't it a matter of thresholds, envelopes and logic statements? Say a source has an average loudness of a certain value above a set level. Also a certain stereo loudness balance over time, and one side drops by half the loudness but material resembles the loud side, stereo balance is off. Logic at that point tries to raise up to desired level, and if no large peaks, it holds it there. When drop ends and all volume comes back the logic releases the volume back to before drop, and stereo pressure evens out. I can see that would be a problem on a lot of music material, but standard mastered pop would be possible to predict to some extent. There is also the easily detected s/n ratio to measure on. I'm not aware of anyone ever attempting to program this. |
|
| Page 1 of 1 | All times are UTC+02:00 |
| Powered by phpBB® Forum Software © phpBB Limited https://www.phpbb.com/ |
|