Change log

v0.2.0 <2022-xx-xx>

Newer numpy (>v1.20.0) is required to avoid ABI compatibility issues. Please check the updated installation guide.

  • #92: Support for Python 3.10, rewrite tests with pytest

v0.1.21 <2022-06-05>

Fixes windows build issues caused by intrinsic conflicts

v0.1.20 <2022-01-04>

Fixes for ValueError: numpy.ndarray size changed, may indicate binary incompatibility. Expected 96 from C header, got 88 from PyObject

v0.1.19 <2022-01-04>

  • #87: Add pysen and format all the codes. Appveyor is now disabled. Windows CI is done by Github Actions.

  • #83: pitch: support automatic type converseion for swipe and rapt (API change)

v0.1.18 <2020-03-02>

  • #75: Enable py38 test on travis

  • #72: Fix itype!=0 for mgcep, gcep and related functions

v0.1.17 <2019-05-31>

  • Fixed python 3.7 installation problem

v0.1.16 <2019-02-17>

  • Fixed cython-generated files in the release tarball with latest numpy

v0.1.15 <2019-01-08>

  • #65, #67`_: Should avoid build-time numpy dependency for newer releases

v0.1.14 <2018-11-25>

  • #62, #64: IMPORTANT: Add use_scipy option to levdur and lpc, and set it to True by default to improve numerical stability. The results should be same regardless to the option, but this would be technically a breaking change.

  • #64: Add acorr.

v0.1.13 <2018-11-19>

  • Add All zero synthesis filter.

  • Add levdur.

  • Add tranposed synthesis filters (mlsadft, poledft, mglsadft, glsadft)

  • Add missing high level synthesis filter class GLSADF.

v0.1.12 <2018-10-27>

  • #63: Fix lpc2lsp bug, add lsp2lpc function. Add regression tests for those.

v0.1.11 <2018-02-05>

  • #55: Add numpy implementation of cdist

v0.1.10 <2018-01-02>

  • #54: Changes from SPTK v3.11 release. 6 and 7 pade approximatino in lmadf and mlsadf is now supported,

v0.1.9 <2018-01-01>

  • BUG fix: example_audio_data is now included in the release tar.gz

v0.1.8 <2017-12-25>

  • c2acr: Fix segfaults for small fftsize

v0.1.7 <2017-06-28>

  • Extend vec2vec functions to mat2mat #49

  • Support automatic type conversions #48

v0.1.6 <2017-05-18>

  • Add mcepalpha. #43

  • Add mc2e. #42

  • Add sp2mc and mc2sp. #41

v0.1.5 <2017-04-22>

  • Fix mcep eps check and input length #39

v0.1.4 <2015-11-23>

  • Add developer documentation (#34)

  • Separate cython implementation and interface (#35)

  • Add RAPT (#32)

  • Add excite function (#31) @jfsantos

  • Fix inconsistent docs about normalization flag for window functions

  • Fix test failure in c2dps / ndps2c (#29)

v0.1.3 <2015-10-02>

  • Building binary wheels for Windows using Appveyor (#28)

  • Add Installation guide on windows (#25)

  • Start Windows continuous integration on AppVeyor (#24). As part of the issue, binary dependency was updated so that SPTK library can be compiled on linux, osx and Windows as well.

  • Remove unnecesarry array initialization (#23)

v0.1.2 <2015-09-12>

  • Add pysptk.synthesis package that provides high level interfaces for speech waveform synthesis (#14)

  • Add cross-link to the docs

  • Add pysptk.conversion.mgc2b

  • Add speech analysis and re-synthesis demonstration notebook (#13)

  • Add pysptk.util.example_audio_file

  • Add fftcep (#18)

  • Add mfcc (#21)

  • Cython is now only required to build development versioni of pysptk. (#8)

v0.1.1 <2015-09-05>

  • Include *.c to pypi distribution

v0.1.0 <2015-09-05>

  • Initial release