| Tue, 10 Jul 2012 22:16:00 +0100 |
Chris Cannam |
Simplify confidence metric
|
| Mon, 09 Jul 2012 22:34:11 +0100 |
Chris Cannam |
oops! also avoid memory leak
|
| Mon, 09 Jul 2012 22:32:14 +0100 |
Chris Cannam |
Thread safety
|
| Mon, 09 Jul 2012 22:31:43 +0100 |
Chris Cannam |
More tweaking with thresholds and confidences -- but I feel like this is going in the wrong direction
|
| Sun, 08 Jul 2012 10:39:14 +0100 |
Chris Cannam |
Merge
|
| Sat, 07 Jul 2012 21:06:37 +0100 |
Chris Cannam |
Change peak-to-next-peak ratio to a difference (because difference of peaks in log values don't vary with signal amplification)
|
| Fri, 06 Jul 2012 19:50:01 +0100 |
Chris Cannam |
OS/X build bits
|
| Thu, 05 Jul 2012 21:54:50 +0100 |
Chris Cannam |
Clumsily cut-n-paste interpolation into pitch tracker as well
|
| Thu, 05 Jul 2012 20:50:56 +0100 |
Chris Cannam |
Add rather simplistic cubic interpolation for peak values in a cepstrum output (not yet for pitch tracker). Only gains us 1dp
|
| Thu, 05 Jul 2012 08:29:20 +0100 |
Chris Cannam |
Merge from branch "track"
|
| Wed, 04 Jul 2012 22:28:46 +0100 |
Chris Cannam |
Adjust hypothesis tolerances (i.e. note breaking algorithm)
track
|
| Tue, 03 Jul 2012 21:10:56 +0100 |
Chris Cannam |
A first crack at turning frequencies into notes
track
|
| Mon, 02 Jul 2012 21:37:02 +0100 |
Chris Cannam |
Introduce peak-to-second-peak ratio, which looks like a reasonable proxy for harmonic-ness. Use it to ascribe a confidence to estimates in the pitch tracker & rely on that to determine how many similar estimates make a satisfied hypothesis
track
|
| Sun, 01 Jul 2012 11:33:37 +0100 |
Chris Cannam |
Reap expired/rejected hypotheses from list; report an estimate as accepted only if the accepting hypothesis has been satisfied already (otherwise a different provisional hypothesis might become satisfied first)
track
|
| Sat, 30 Jun 2012 10:12:08 +0100 |
Chris Cannam |
Another small fix
track
|
| Sat, 30 Jun 2012 10:10:05 +0100 |
Chris Cannam |
Make it, at least broadly, work
track
|
| Fri, 29 Jun 2012 21:11:37 +0100 |
Chris Cannam |
Make compile; does not yet work
track
|
| Fri, 29 Jun 2012 11:26:44 +0100 |
Chris Cannam |
Closing unsuccessful anonymous branch
track
|
| Fri, 29 Jun 2012 11:25:05 +0100 |
Chris Cannam |
An experiment in mean-filtering proportion values -- not very successful
track
|
| Fri, 29 Jun 2012 10:09:19 +0100 |
Chris Cannam |
Partial updates to new method (but this doesn't currently build)
track
|
| Thu, 28 Jun 2012 12:17:18 +0100 |
Chris Cannam |
Start an alternative idea
track
|
| Tue, 26 Jun 2012 16:06:00 +0100 |
Chris Cannam |
Some work on accepting/rejecting peaks
track
|
| Tue, 26 Jun 2012 15:30:12 +0100 |
Chris Cannam |
Vertical filtering
track
|
| Mon, 25 Jun 2012 16:35:38 +0100 |
Chris Cannam |
Switch to inverse symmetric method. Forward difference worked nicely for some examples but very badly for others (octave errors)
track
|
| Mon, 25 Jun 2012 15:28:51 +0100 |
Chris Cannam |
Experimental tracker (on branch)
track
|
| Mon, 25 Jun 2012 14:32:05 +0100 |
Chris Cannam |
Copyrights, some cleaning up
|
| Mon, 25 Jun 2012 13:39:46 +0100 |
Chris Cannam |
Add total and peak-proportion outputs
|
| Mon, 25 Jun 2012 11:45:33 +0100 |
Chris Cannam |
Change "relative to mean" to "relative to RMS", and add peak-to-RMS output. Add a mean-filter history. Remove essentially useless forward-power method (same as forward-magnitude with 2x factor). Refactor a bit
|
| Fri, 22 Jun 2012 23:56:37 +0100 |
Chris Cannam |
Attempt a complex inverse transform -- doesn't seem to work well?
|
| Fri, 22 Jun 2012 23:01:00 +0100 |
Chris Cannam |
Add choice of cepstrum transform method
|
