Commit Graph

  • e3c817b8cf Make this terminology a bit more consistent Chris Cannam 2022-09-29 08:27:29 +01:00
  • fa92a924f6 Mention command-line utility change Chris Cannam 2022-09-27 10:33:00 +01:00
  • 80eb4a74f2 Minor clarification Chris Cannam 2022-09-27 08:54:07 +01:00
  • 6a45cce987 Update CHANGELOG for v3.1 Chris Cannam 2022-09-27 08:48:09 +01:00
  • cc212f0e4a Avoid naming output "test" - on some dev machines I have a directory called that Chris Cannam 2022-09-27 08:47:59 +01:00
  • 1dc6abdb96 Remove libresample support. It hasn't been exercised in years and nobody should be using it in preference to the other options Chris Cannam 2022-09-27 08:47:33 +01:00
  • 163ac1e485 Ah, I was forgetting the existence of sf_command. Use that to identify file type where possible Chris Cannam 2022-09-26 18:12:04 +01:00
  • f8d94082e0 If the output file extension differs from the input file extension and is one of the readily identified strings wav, w64, or ogg, then open the output file with that format; otherwise use the same format as the input (as we previously did always). See https://todo.sr.ht/~breakfastquay/rubberband/20 Chris Cannam 2022-09-26 17:56:29 +01:00
  • 9e423cdd8c Handle non-ideal sample rates in R3; document sample rate parameter Chris Cannam 2022-09-26 16:02:13 +01:00
  • f586fe601a Try to avoid situation in which JNI autodetection crashes out the build because javac is not found Chris Cannam 2022-09-26 12:48:44 +01:00
  • e5f756d16d Fix use of variable I had tidied out Chris Cannam 2022-09-26 10:20:42 +01:00
  • 57d59b4c7a Eliminate the mlock logic entirely. It's caused a couple of issues lately and it turns out it wasn't even used! Chris Cannam 2022-09-26 10:11:16 +01:00
  • 0eaca1711c Prefer jni as a dependency rather than manually seeking it. This is unlikely to work in Meson < 0.62, and Meson >= 0.62 gives a warning that the feature is new - but it seems to me that older versions should just do a standard dep lookup and fail, so we can fall back on the same logic we were using anyway (i.e. manually supplied include path) Chris Cannam 2022-09-26 09:39:10 +01:00
  • 020702e7ea Fix include path. Very poor that this wasn't exercised by my build or CI runs Chris Cannam 2022-09-26 09:37:55 +01:00
  • 56058cc288 Fix plainly dumb error Chris Cannam 2022-09-20 10:27:16 +01:00
  • a5e2bfa444 Fix some (benign) shadowing warnings Chris Cannam 2022-09-20 09:30:48 +01:00
  • e90f377600 Don't use drand48, it doesn't always exist and since we don't care about the quality of output, it's not worth testing for Chris Cannam 2022-09-20 09:30:34 +01:00
  • 2eb9a20052 There's no longer any good reason to do this; remove it Chris Cannam 2022-09-09 11:22:53 +01:00
  • 214ddd6899 Build fix Chris Cannam 2022-09-02 16:40:52 +01:00
  • 9001b5459c Merge from branch performance Chris Cannam 2022-09-02 16:35:46 +01:00
  • f4115e63fa Update docs Chris Cannam 2022-09-02 16:35:35 +01:00
  • 8ee381efe6 Update docs for library support Chris Cannam 2022-09-02 16:20:17 +01:00
  • 45e8bef4cf Slightly nicer output Chris Cannam 2022-09-02 15:58:08 +01:00
  • 92362bdba0 Resample in chunks as well, to avoid overrunning resample buffer Chris Cannam 2022-09-02 15:08:29 +01:00
  • 021de9d51c Add resample-before mode Chris Cannam 2022-09-02 14:49:34 +01:00
  • deb84a5f78 Update build files Chris Cannam 2022-09-02 13:11:15 +01:00
  • 8c817265d5 Pick more rational rational values Chris Cannam 2022-09-02 12:07:41 +01:00
  • aa1865d22b Add mathmisc.cpp Chris Cannam 2022-09-02 12:07:29 +01:00
  • 2025c51d64 Use extracted pickNearestRational; ensure the max rational is actually the max for both num and denom Chris Cannam 2022-09-02 12:06:45 +01:00
  • a525654d6a Pull out nearest-rational into mathmisc so it can be used in other places than BQResampler Chris Cannam 2022-09-02 12:06:11 +01:00
  • ee3091ab75 Merge from branch speexdsp Chris Cannam 2022-09-02 11:12:35 +01:00
  • e0a6fc686d Add support for external libspeexdsp Chris Cannam 2022-09-02 11:10:44 +01:00
  • 4336d47a02 With longer process sizes we can end up with insufficient space in the inbuf, but enough in the outbuf following the previous retrieve call for a consume to be possible before writing to the inbuf Chris Cannam 2022-09-01 10:41:25 +01:00
  • 20d22f76d6 Don't pass debug level through to FFT - it doesn't respect the log target Chris Cannam 2022-08-30 18:55:33 +01:00
  • 53eca719fb Apply gradual phase-reset on unity in the R2 stretcher (R3 already does this) Chris Cannam 2022-08-18 16:41:43 +01:00
  • e58c6018c5 This appears benign and should be slightly quicker Chris Cannam 2022-08-18 15:24:24 +01:00
  • c9565266f2 Adjust the limits, with performance in mind - to be auditioned Chris Cannam 2022-08-18 14:27:57 +01:00
  • 515f5e2bd1 Allow pre-configuration if explicitly specified Chris Cannam 2022-08-18 12:03:29 +01:00
  • 1d7a684e1a Looks like that was not the way to comment something out Chris Cannam 2022-08-09 16:46:26 +01:00
  • 147e88ad55 Add SLEEF to README. I was going to add it to the CI, but it looks as if Ubuntu's libsleef-dev doesn't include sleefdft. Chris Cannam 2022-08-09 16:44:56 +01:00
  • 8fee46b704 Add SLEEF FFT support Chris Cannam 2022-08-09 15:50:02 +01:00
  • f81598c166 Backed out changeset c03ab2acf74e No, that worked very badly in practice. Revisit Chris Cannam 2022-08-08 17:02:08 +01:00
  • f027a59fa4 I don't like adjusting the rational approximation - it changes (even if only nominally) the effective ratio based on quality. I think this is more appropriate Chris Cannam 2022-08-08 16:55:53 +01:00
  • 6ecc973d41 Keep a structure with recommended limits for inhop/outhop in current configuration; skip readahead entirely when inhop is greater than a certain limit - allowing us to reduce start delay further in short-window mode Chris Cannam 2022-08-05 14:58:12 +01:00
  • 4aea160ec3 Indentation only Chris Cannam 2022-08-04 16:58:00 +01:00
  • 3eae190a85 This is not actually usually a serious problem, unlike the <1 case which can be genuinely troublesome Chris Cannam 2022-08-04 16:31:23 +01:00
  • 63ebc45107 ... I said, in RatioOftenChanging mode Chris Cannam 2022-08-04 13:34:31 +01:00
  • 76ee5998f1 Permit rational precision to vary by quality setting in RatioOftenChanging mode Chris Cannam 2022-08-04 13:31:33 +01:00
  • 7833e1bff1 Avoid kick analysis in single-window mode; comment update Chris Cannam 2022-08-04 12:08:15 +01:00
  • 5b26e90a58 Fix to windowing offsets Chris Cannam 2022-08-04 11:19:13 +01:00
  • fe9e86bc3c Per-channel window source buffer, & connect it up Chris Cannam 2022-08-04 10:31:36 +01:00
  • d1386b0a0c Provide a separate window source buffer to simplify single/multi window logic Chris Cannam 2022-08-04 10:02:09 +01:00
  • 2fa0e1162e Toward properly expressing the window sizes in Guide when in single-window mode, so as to separate the time-domain frame length from the longest FFT within R3Stretcher and allow us to use a shorter time-domain frame. Not working correctly in single-window mode yet. Chris Cannam 2022-08-03 14:16:17 +01:00
  • 9fff2836c6 Avoid RatioMostlyFixed in RealTime mode, it can lead to unpleasant surprises Chris Cannam 2022-08-02 16:22:24 +01:00
  • e3ae576fcd Avoid including incorrect literal in pc.in, even if it is supposed to be replaced on install Chris Cannam 2022-07-28 17:49:13 +01:00
  • d211121701 Ah, this was still in use in Scavenger. It can stay for now Chris Cannam 2022-07-14 17:49:22 +01:00
  • 511645fed2 Further version numbers Chris Cannam 2022-07-14 14:16:16 +01:00
  • 8e246081b6 Merge Chris Cannam 2022-07-14 14:15:16 +01:00
  • fbcd385c32 We're working toward 3.1.0 here Chris Cannam 2022-07-14 14:15:10 +01:00
  • d0cca91e30 Impose a limit here Chris Cannam 2022-07-14 13:52:04 +01:00
  • e018458736 Make this fully single-windowed rather than just short-windowed Chris Cannam 2022-07-14 13:44:22 +01:00
  • 279c44477e Fix unity handling in short-window mode Chris Cannam 2022-07-14 12:07:43 +01:00
  • 02928a3c86 Experimentally (re-)introduce short window mode Chris Cannam 2022-07-14 11:55:21 +01:00
  • dc74c993cf Add profile points; switch to microseconds Chris Cannam 2022-07-14 10:02:39 +01:00
  • 7179c2626c Update Profiler to use std::chrono Chris Cannam 2022-07-14 09:42:10 +01:00
  • 766cb4b061 Added tag v3.0.0 for changeset 59872cfc6f23 Chris Cannam 2022-07-07 15:59:49 +01:00
  • ae8f238239 Minor wording fix Chris Cannam 2022-07-07 15:59:45 +01:00
  • 9e6e5a4768 Added tag v3.0.0 for changeset f00a6735b95c Chris Cannam 2022-07-07 15:58:01 +01:00
  • 6ec484fc31 Update Mac pack to include -r3 Chris Cannam 2022-07-07 15:44:16 +01:00
  • 1aa241e6d0 Small doc and CHANGELOG updates Chris Cannam 2022-07-07 13:51:41 +01:00
  • 5a67f9fcdf Update .so version Chris Cannam 2022-07-07 13:39:16 +01:00
  • dca4c96834 Wire up setExpectedInputDuration Chris Cannam 2022-07-07 10:54:23 +01:00
  • 263e0cbf70 Ensure output accumulator is properly drained at finish. We rely on the calculated output count (in offline mode) to truncate properly. Chris Cannam 2022-07-07 10:35:01 +01:00
  • a66d2a5140 Fix consumed counter for key-frame mapping Chris Cannam 2022-07-07 09:43:17 +01:00
  • f47f26d351 Next one should be actual 3.0.0 Chris Cannam 2022-07-06 15:38:34 +01:00
  • 4673a663f0 Add the other new 3.0 functions to the C API Chris Cannam 2022-07-06 11:56:01 +01:00
  • 567e8636cd Enable PROCESS_SAMPLE_TYPE for R3 as well, switching from double to process_t throughout Chris Cannam 2022-07-06 10:40:10 +01:00
  • 6e941fa2d3 Update the OptionPitch* options so that they do the expected thing for the R3 engine as well - this speeds up realtime mode when not dynamically pitch-shifting Chris Cannam 2022-07-06 10:22:50 +01:00
  • 6ecd103dd0 Added tag v3.0.0-beta4 for changeset 458a655243ae Chris Cannam 2022-07-05 18:03:41 +01:00
  • 3850b93786 beta4 Chris Cannam 2022-07-05 18:03:40 +01:00
  • cfee8238bf Added tag v3.0.0-beta4 for changeset cc3dd8ef4e14 Chris Cannam 2022-07-05 18:03:35 +01:00
  • 3efb5625d1 Added tag v3.0.0-beta4 for changeset 54e5351b46cf Chris Cannam 2022-07-05 18:00:01 +01:00
  • 72654b04ea Introduce getPreferredStartPad(), deprecate getLatency() and replace with getStartDelay(); document and test both Chris Cannam 2022-07-05 17:53:36 +01:00
  • 31984af826 Add a test case to cover realtime block-by-block processing with latency commpensation Chris Cannam 2022-07-05 13:49:28 +01:00
  • dd7e8f27a0 Wording suggested something more audibly timing-related than it actually is, correct this Chris Cannam 2022-07-04 16:06:49 +01:00
  • 3dac532ba5 Merge from branch r3 Chris Cannam 2022-07-04 14:23:05 +01:00
  • 687e3958a7 Connect up the ChannelsTogether option in R3 as well, to extend the channel lock range upward Chris Cannam 2022-07-04 10:52:50 +01:00
  • 10661f40c6 Merge from branch r3 Chris Cannam 2022-07-01 17:11:29 +01:00
  • 67cb55c375 Added tag v3.0.0-beta3 for changeset 78a701fb6daa Chris Cannam 2022-07-01 13:53:56 +01:00
  • 7bb916e71e Beta3 Chris Cannam 2022-07-01 13:53:51 +01:00
  • cab969f14c Added tag v3.0.0-beta3 for changeset 58b588a580a1 Chris Cannam 2022-07-01 13:52:34 +01:00
  • ee3e7e45ff Fix & test kissfft build Chris Cannam 2022-07-01 13:14:21 +01:00
  • 335aef35c9 The Boost meson module sets this on the command-line, so avoid compiler warning when redefining Chris Cannam 2022-07-01 12:01:06 +01:00
  • 8d4d8a39bc Actually picking up the runtime libraries may be a bigger task Chris Cannam 2022-06-30 15:27:25 +01:00
  • 1afb4acdd5 Update CHANGELOG for 3.0.0 Chris Cannam 2022-06-30 15:21:46 +01:00
  • e9403fb52d Quoting Chris Cannam 2022-06-30 15:21:39 +01:00
  • b6be35158b Experiment with the single-file build in CI Chris Cannam 2022-06-30 15:07:09 +01:00
  • 68a9e4f8d5 Require specific Boost test version, and don't check for it manually as we can't do the version test. (We use quite a recent version of the test macros) Chris Cannam 2022-06-30 15:05:19 +01:00
  • fc5aaad929 Correct (I hope) package name Chris Cannam 2022-06-30 14:48:09 +01:00
  • 4c07eda216 Add libboost for tests Chris Cannam 2022-06-30 14:37:15 +01:00