| Tue, 28 Jan 2014 00:31:17 +0000 |
Adam |
Added a function to allow updates to the hop and frame size of the beat tracker
develop
|
| Tue, 28 Jan 2014 00:19:21 +0000 |
Adam |
Changed all pointers to arrays in OnsetDetectionFunction into vectors
develop
|
| Mon, 27 Jan 2014 23:54:18 +0000 |
Adam |
Replaced the pointers to arrays in BTrack with vectors
develop
|
| Mon, 27 Jan 2014 23:20:13 +0000 |
Adam |
Added installation instructions for Python module and Vamp plug-in
develop
|
| Mon, 27 Jan 2014 23:11:31 +0000 |
Adam |
Updated README, commented more code, added a Vamp plug-in
develop
|
| Sat, 25 Jan 2014 18:17:51 +0000 |
Adam |
Did more commenting. Added documentation.
develop
|
| Fri, 24 Jan 2014 21:45:55 +0000 |
Adam |
Renamed many variables, functions and arguments so they have more sensible names. Also removed an apparently redundant variable in OnsetDetectionFunction called wframe
develop
|
| Thu, 23 Jan 2014 18:00:53 +0000 |
Adam |
Did more renaming of obscurely named variables and added a couple of accessor methods
develop
|
| Thu, 23 Jan 2014 15:31:11 +0000 |
Adam |
Replaced switch statements in OnsetDetectionFunction with enums. Renamed lots of functions so that they have better names, in camel case. Added some unit tests for initialisation of BTrack.
develop
|
| Thu, 23 Jan 2014 12:17:06 +0000 |
Adam |
Added some simple unit tests. Removed the destructor from the BTrack class as it was unnecessary.
develop
|
| Wed, 22 Jan 2014 18:47:16 +0000 |
Adam |
Changed the interface of the algorithm so that onset detection function samples are calculated internally. This makes the call to the algorithm for most cases much simpler. Also added a static function for calculating beat times in seconds based upon sampling frequency, hop size and the current frame number.
develop
|
| Wed, 22 Jan 2014 02:49:29 +0000 |
Adam |
Before this commit, the OnsetDetectionFunction class used double precision, and the BTrack class used floats. I have now made BTrack use double precision also. This works fine and the only cost of doing this is that we have to perform one array copy into floating point format so that sample rate conversion (which has to be in floating point format) can take place.
develop
|
