| Wed, 23 Jul 2014 18:36:41 +0100 |
Chris Cannam |
Max gain of 10 seems like it should be perfectly adequate! (and tests OK)
|
| Wed, 23 Jul 2014 15:17:47 +0100 |
Chris Cannam |
The complicated long-and-short-term approach is not working any better than a simple long-term window. Revert to that, with some decay.
|
| Wed, 23 Jul 2014 10:29:26 +0100 |
Chris Cannam |
Tidy
|
| Tue, 22 Jul 2014 16:11:02 +0100 |
Chris Cannam |
Make object accessible as a standalone class as well as through LADSPA
|
| Tue, 22 Jul 2014 15:40:45 +0100 |
Chris Cannam |
Comment only
|
| Tue, 22 Jul 2014 13:49:46 +0100 |
Chris Cannam |
Slightly more effective max rms
|
| Tue, 22 Jul 2014 13:36:52 +0100 |
Chris Cannam |
Restore local part to that last one, again without much success
|
| Tue, 22 Jul 2014 13:32:55 +0100 |
Chris Cannam |
Simpler decay
|
| Tue, 22 Jul 2014 13:28:05 +0100 |
Chris Cannam |
Some decay on the long-term-only version
|
| Tue, 22 Jul 2014 13:18:33 +0100 |
Chris Cannam |
Introduce some local variation to the global level, with mixed results
|
| Tue, 22 Jul 2014 11:53:30 +0100 |
Chris Cannam |
Global-only version that performs reasonably well
|
| Tue, 22 Jul 2014 11:27:59 +0100 |
Chris Cannam |
A global/local version that works approximately as well in first tests as the global-only version (r8)
|
| Tue, 22 Jul 2014 08:11:45 +0100 |
Chris Cannam |
Slightly increase RMS max
|
| Mon, 21 Jul 2014 18:01:11 +0100 |
Chris Cannam |
Experiment with permitting some local variation rather than always aiming to flatten
|
| Thu, 17 Jul 2014 16:41:55 +0100 |
Chris Cannam |
Try a lower level
|
