5935d17ee0
Introduce first cut of live shifter
Chris Cannam
2023-06-29 16:49:30 +01:00
1eddafd7a7
Update resampler selection text to note that libsamplerate is particularly faster n 32-bit builds
Chris Cannam
2023-06-26 16:42:42 +01:00
df596f472e
Minor threshold tweak
Chris Cannam
2023-06-26 11:38:20 +01:00
82dab93ecf
Merge from branch process-size
Chris Cannam
2023-06-08 09:48:16 +01:00
2949996cf6
Remove redundant test (processOneChunk checks this and returns true if accumulator empty when draining)
Chris Cannam
2023-06-08 09:44:09 +01:00
d97551834f
Make some debug lower priority
Chris Cannam
2023-06-08 09:40:11 +01:00
73f7f7bb0a
Fix handling of oversized process buffers in mid-side mode
Chris Cannam
2023-06-07 17:01:47 +01:00
74395b9976
Fix oversized final block tests for libsamplerate as well
Chris Cannam
2023-06-07 14:37:29 +01:00
fe1aafea28
Further work on end state with long process sizes in R2. This passes tests but is not the minimal necessary change I think - review.
Chris Cannam
2023-06-07 11:16:07 +01:00
b1cd0913e2
Significant further work on internal buffer sizes
Chris Cannam
2023-06-01 14:09:39 +01:00
af6759f74b
Work on proper sizing of internal buffers
Chris Cannam
2023-06-01 10:51:20 +01:00
173c7212f1
Ensure tests actually exercise large input buffers
Chris Cannam
2023-06-01 10:51:09 +01:00
f0b9061028
Use longer time ratio, otherwise we are retrieving fewer samples than we intended to
Chris Cannam
2023-05-31 14:06:15 +01:00
ffaef18a9d
Use parameters more similar to the other tests, as the internal test checking code expects these (otherwise we have spurious failures with some resamplers)
Chris Cannam
2023-05-31 09:35:27 +01:00
49a132b612
Add note about possible repeated reads
Chris Cannam
2023-05-31 09:22:16 +01:00
488de631a0
Add test with longer blocksize in realtime mode
Chris Cannam
2023-05-31 09:10:52 +01:00
7d5b101d27
Merge from branch libatomic-check
Chris Cannam
2023-05-19 12:38:39 +01:00
72cd31eb6c
Fix typo
Chris Cannam
2023-05-19 12:36:05 +01:00
529634134e
Attempt a test for libatomic requirement
Chris Cannam
2023-04-25 17:03:49 +01:00
b04b731a95
Added tag v3.2.1 for changeset 83d096887e1b
Chris Cannam
2023-04-06 11:58:29 +01:00
9e8ef1eeaf
Further slackening necessary for 32-bit builds
Chris Cannam
2023-04-06 11:04:45 +01:00
f6f6bc581b
Update for v3.2.1
Chris Cannam
2023-04-05 14:03:09 +01:00
ab9c0de723
Slacken a test slightly for kissfft builds
Chris Cannam
2023-04-05 12:00:00 +01:00
dfef60c486
Slacken some test expectations appropriately for non-builtin resampler, after reviewing the output
Chris Cannam
2023-04-05 11:51:26 +01:00
b9f2b58e1f
Fix use of final flag more than once at end of processing (causing a test failure with libsamplerate)
Chris Cannam
2023-04-05 11:14:04 +01:00
376bf4dbd6
Update Ubuntu version - seems 20.04 was too old for the version of boost-test we use
Chris Cannam
2023-04-05 11:13:36 +01:00
904e72b10b
Added tag v3.2.0 for changeset 46d8430844d6
Chris Cannam
2023-03-28 16:46:11 +01:00
0804473c63
Merge from branch 3.1-stable
Chris Cannam
2023-03-28 14:41:58 +01:00
63867dc11c
Remove JNI version requirement - I am seeing variants with versions like 1.8.0 as well as with > 7.0.0
Chris Cannam
2023-03-28 14:41:45 +01:00
6b6c1d0ea0
Correct description in changelog
Chris Cannam
2023-03-28 12:26:33 +01:00
908239462b
Adjust versions for 3.2.0
Chris Cannam
2023-03-27 16:29:54 +01:00
daaa47fab2
Merge from branch 3.1-stable
Chris Cannam
2023-03-27 16:26:33 +01:00
179dbd31c9
Update for 3.1.3
Chris Cannam
2023-03-27 08:38:12 +01:00
bbed05108c
This branch will be 3.1.3
Chris Cannam
2023-03-27 08:35:16 +01:00
b39ba8d518
Update CHANGELOG for 3.1.3
Chris Cannam
2023-03-27 08:34:54 +01:00
ba9fe26dcd
Never phase reset on silence in side channel, now that such resets are carried across
Chris Cannam
2023-03-24 17:51:44 +00:00
21b968c0c5
Update docs for OptionChannelsTogether
Chris Cannam
2023-03-23 17:58:46 +00:00
18b32431b4
Update changelog
Chris Cannam
2023-03-23 17:49:59 +00:00
23cc540533
Merge from branch resetcheck
Chris Cannam
2023-03-23 17:28:29 +00:00
c66d44dbb4
Merge from branch midside3
Chris Cannam
2023-03-23 17:28:17 +00:00
fbb4979f87
Further reset tests
Chris Cannam
2023-03-23 17:26:36 +00:00
c51f6b1ae2
Fix erroneous limits calculations
Chris Cannam
2023-03-23 17:26:28 +00:00
963752951d
Fix erroneous limits calculations
Chris Cannam
2023-03-23 17:26:28 +00:00
adfe1ff2cf
Further resets
Chris Cannam
2023-03-23 17:26:11 +00:00
717f5530eb
Update docs for centre-focus
Chris Cannam
2023-03-23 13:51:36 +00:00
d063f851b4
Further test
Chris Cannam
2023-03-23 13:01:19 +00:00
47f8845486
Remove duplicate tests
Chris Cannam
2023-03-23 10:26:57 +00:00
99d2d4a080
Further adjustments to reset, and ensure initial prev in/out hops are set the same way if ratios set after construction as they are if ratios set on construction. Plus tests
Chris Cannam
2023-03-23 10:25:18 +00:00
6c71159593
More reset tests, tighten up reset logic
Chris Cannam
2023-03-22 13:26:15 +00:00
5f93a90902
Merge from branch midside3
Chris Cannam
2023-03-20 10:08:20 +00:00
2dcb27d3dd
Toward properly testing reset
Chris Cannam
2023-03-20 10:08:01 +00:00
e2611485d8
Use mid-side processing in channels-together mode in R3; for a more stable stereo image, ensure transient bins are reset in mid whenever they are in side
Chris Cannam
2023-03-17 13:19:18 +00:00
d3830870c0
Experiments with mid-side in R3 - in this code always used when channels=2
Chris Cannam
2023-03-15 18:01:21 +00:00
9e5ebdbafb
Update changelog for 3.2 (coming shortly)
Chris Cannam
2023-03-09 13:41:19 +00:00
4ec2dad6e3
Fix hang toward end of processing, when faced with a file (e.g. some mp3s) for which libsndfile reports more frames from the header than are actually retrieved
Chris Cannam
2023-03-08 14:45:47 +00:00
5455bb95ae
Fix hang toward end of processing, when faced with a file (e.g. some mp3s) for which libsndfile reports more frames from the header than are actually retrieved
Chris Cannam
2023-03-08 14:45:47 +00:00
f2c2cb5446
Avoid duplicated header list; run more tests
Chris Cannam
2023-03-08 14:45:08 +00:00
4d5460c3e8
Reporting fix
Chris Cannam
2023-03-08 14:44:52 +00:00
48680627d8
Should be using HighConsistency in plugin that supports realtime parameter changes
Chris Cannam
2023-03-08 14:39:12 +00:00
782f3b5b5e
Merge
Chris Cannam
2023-02-21 11:31:25 +00:00
a4299b0ec6
Expand tests
Chris Cannam
2023-02-21 11:30:04 +00:00
4b4c50b479
Tests for handling of final flag
Chris Cannam
2023-02-21 10:28:57 +00:00
d5dd0e573a
Further timing tests
Chris Cannam
2023-02-20 17:46:10 +00:00
1368ba87ee
Fix latency calculations in resample-before modes. This affects realtime use in the non-recommended realtime pitch modes (i.e. those other than HighConsistency)
Chris Cannam
2023-02-20 15:26:12 +00:00
320fb2f7fe
Ensure we consume when called with final=true and no samples - this was the default behaviour in R2 and also initially in R3 but changed when resample-before modes were added
Chris Cannam
2023-02-20 15:25:27 +00:00
afd1b1f86c
Ensure we consume when called with final=true and no samples - this was the default behaviour in R2 and also initially in R3 but changed when resample-before modes were added
Chris Cannam
2023-02-20 15:25:27 +00:00
0fb8b19ab5
Working on further tests
Chris Cannam
2023-02-20 09:53:20 +00:00
b317d0e81a
Fix handling of final block in speex resampler - there is no guarantee a single pad call will return enough data on its own either
Chris Cannam
2023-01-17 12:34:50 +00:00
cc3e265b26
Update copyright dates
Chris Cannam
2023-01-10 11:10:06 +00:00
549906e033
Remove the last occurrences of "using namespace std"
Chris Cannam
2023-01-10 11:04:50 +00:00
e28651acd0
Use "meson setup build", not the shorthand which is now deprecated
Chris Cannam
2023-01-10 11:04:32 +00:00
8edf1723c4
Added tag v3.1.2 for changeset d2aebfc83e21
Chris Cannam
2022-12-02 16:46:43 +00:00
fa8a918d8a
Update CHANGELOG for 3.1.2
Chris Cannam
2022-11-28 11:16:45 +00:00
c871f92b2e
Merge
Chris Cannam
2022-11-16 10:13:20 +00:00
3960bca0ab
Make this exception optional on NO_EXCEPTIONS, as the others are (required for Android)
Chris Cannam
2022-11-16 10:12:41 +00:00
8d4d783155
Correct the list of source files (thanks to jimsimonz) and reorder in the same (alphabetical) order as in the main build file
Chris Cannam
2022-11-16 10:11:34 +00:00
50eb332ec8
Fixes suitable for mingw32 builds
Chris Cannam
2022-11-08 08:32:05 +00:00
3d3cd0454b
Another twist to JNI autodetection. Let's *always* make sure the compiler can find jni.h, even if Meson thinks it has found it - following reports of JNI autodetect succeeding on macOS but thinking jni.h is in some absurd location
Chris Cannam
2022-11-04 10:49:11 +00:00
379bb879f1
Update versions again
Chris Cannam
2022-11-02 15:06:49 +00:00
ea081f509d
Experiment with Meson's built-in features support for optional targets
Chris Cannam
2022-10-31 12:03:46 +00:00
cbc2453743
Added tag v3.1.1 for changeset b7f4072d02d5
Chris Cannam
2022-10-21 15:40:02 +01:00
bbd8de71da
Bump number in Doxyfile
Chris Cannam
2022-10-21 14:51:10 +01:00
f86f31b9f0
Update for 3.1.1
Chris Cannam
2022-10-21 13:57:21 +01:00
1595fcb47e
Now that we have the complete Java language module optionally added (where available) we should be able to use Meson's own jar target rather than manually packaging it. Again thanks to Eli Schwartz here
Chris Cannam
2022-10-13 13:29:18 +01:00
502241ccf5
Ensure lib directory is created
Chris Cannam
2022-10-13 08:24:06 +01:00
64d2989d28
Added tag v3.1.0 for changeset 137ef076894f
Chris Cannam
2022-09-30 14:53:53 +01:00
755ff57fc5
Fix incorrect extension calculation
Chris Cannam
2022-09-30 13:24:12 +01:00
d424616335
Note only in RT mode
Chris Cannam
2022-09-30 11:49:01 +01:00
55c762b726
Update version in docs
Chris Cannam
2022-09-30 11:47:25 +01:00
394a62a655
No longer will it be pre
Chris Cannam
2022-09-30 11:45:49 +01:00
f82daa208c
Let's call this a version
Chris Cannam
2022-09-30 11:45:38 +01:00
52fc576500
Remove obsolete file
Chris Cannam
2022-09-30 10:36:35 +01:00
1bd266cdb6
Defend against NaN/Inf and negative ratios
Chris Cannam
2022-09-30 10:36:29 +01:00
50b231ee23
Merge
Chris Cannam
2022-09-29 16:26:46 +01:00
f26771d077
Add mathmisc to dotnet
Chris Cannam
2022-09-29 16:19:43 +01:00
11d5deff62
Reverting these tweaked values to their 3.0 state - this change wasn't an obvious improvement and just meant that the output no longer matched the last round of listening tests, so revert it
Chris Cannam
2022-09-29 15:57:11 +01:00
f194becadf
Use number of actually-populated bands, not total number
Chris Cannam
2022-09-29 15:45:47 +01:00
de56cd114a
Update the resampler logic to follow that of R2 more - ignore the pitch hq/hs setting in offline mode entirely. The alternative is too tricky to handle elegantly in the command line tool, which I think means too complicated
Chris Cannam
2022-09-29 11:19:21 +01:00
4599b1afd7
Further docs for short window
Chris Cannam
2022-09-29 08:27:50 +01:00
Chris Cannam