Diffsptk Versions Save

What's Changed

• Add histogram by @takenori-y in https://github.com/sp-nitech/diffsptk/pull/70
• Fix bug of magic_intpl by @takenori-y in https://github.com/sp-nitech/diffsptk/pull/72
• Support Python 3.12

What's Changed

• Add functional for ease of use
• Add CQT and ICQT based on librosa's implementation

Minor Breaking Changes

• mode parameter in IIR has been removed
• The order of the arguments of LPC has been changed

What's Changed

• Bug fix by @takenori-y in https://github.com/sp-nitech/diffsptk/pull/67

What's Changed

• Add plp by @takenori-y in https://github.com/sp-nitech/diffsptk/pull/62
• Add alaw by @takenori-y in https://github.com/sp-nitech/diffsptk/pull/63
• Use pad instead of cat by @takenori-y in https://github.com/sp-nitech/diffsptk/pull/64

Minor Changes

• LinearInterpolation module drops nn.ReplicationPad1d module
• Restrict the version of vector-quantize-pytorch

What's Changed

• Add lpc2lsp by @takenori-y in https://github.com/sp-nitech/diffsptk/pull/53 and https://github.com/sp-nitech/diffsptk/pull/59
• Add lsp2sp by @takenori-y in https://github.com/sp-nitech/diffsptk/pull/54
• Change implementation of root_pol by @takenori-y in https://github.com/sp-nitech/diffsptk/pull/55
• Add pol_root by @takenori-y in https://github.com/sp-nitech/diffsptk/pull/56
• Add lsp2lpc by @takenori-y in https://github.com/sp-nitech/diffsptk/pull/57
• Update dfs by @takenori-y in https://github.com/sp-nitech/diffsptk/pull/58
• Add lspcheck by @takenori-y in https://github.com/sp-nitech/diffsptk/pull/60

What's Changed

• Minor bug fix by @takenori-y in https://github.com/sp-nitech/diffsptk/pull/50

What's Changed

• Purge freqt2 from smcep by @takenori-y in https://github.com/sp-nitech/diffsptk/pull/26
• Support zero-phase non-cascade MLSA filter by @takenori-y in https://github.com/sp-nitech/diffsptk/pull/28
• Add yingram by @takenori-y in https://github.com/sp-nitech/diffsptk/pull/29
• Add mlsacheck by @takenori-y in https://github.com/sp-nitech/diffsptk/pull/30
• Expose PQMF parameters by @takenori-y in https://github.com/sp-nitech/diffsptk/pull/33
• Support sawtooth excitation generation by @takenori-y in https://github.com/sp-nitech/diffsptk/pull/37
• Support MLSA filter in the frequency domain by @takenori-y in https://github.com/sp-nitech/diffsptk/pull/36
• Improve linear_intpl by @yoyololicon in https://github.com/sp-nitech/diffsptk/pull/38
• Add gmm by @takenori-y in https://github.com/sp-nitech/diffsptk/pull/39
• Support learnable PQMF by @takenori-y in https://github.com/sp-nitech/diffsptk/pull/41
• Add ap by @takenori-y in https://github.com/sp-nitech/diffsptk/pull/43
• Add poledf by @takenori-y in https://github.com/sp-nitech/diffsptk/pull/44
• Add rlevdur by @takenori-y in https://github.com/sp-nitech/diffsptk/pull/45
• Add magic_intpl by @takenori-y in https://github.com/sp-nitech/diffsptk/pull/47

Major Breaking Changes

• cascade parameter in MLSA has been changed to mode parameter
• Minimal PyTorch version has been changed to 1.11.0

What's Changed

• Support sinusoidal excitation by @takenori-y in https://github.com/sp-nitech/diffsptk/pull/15
• Get alpha from sample rate by @takenori-y in https://github.com/sp-nitech/diffsptk/pull/16
• Add root_pol by @takenori-y in https://github.com/sp-nitech/diffsptk/pull/17
• Add vq by @takenori-y in https://github.com/sp-nitech/diffsptk/pull/19
• Add msvq and imsvq by @takenori-y in https://github.com/sp-nitech/diffsptk/pull/20
• Add unframe by @takenori-y in https://github.com/sp-nitech/diffsptk/pull/21
• Add istft by @takenori-y in https://github.com/sp-nitech/diffsptk/pull/22
• Add lbg by @takenori-y in https://github.com/sp-nitech/diffsptk/pull/23
• Fix slow import by @takenori-y in https://github.com/sp-nitech/diffsptk/pull/24
• Module sopr was stripped

Fifth release

Additional implemented modules are:

• cqt
• entropy
• excite
• imglsadf
• mglsadf
• train

The recursive computation used in c2mpir, mpir2c, mgc2mgc, and mgc2sp has been changed to the FFT-based one for parallel processing.

Fourth release

Additional implemented modules are:

• delay
• lar2par
• mgcep (updated from mcep)
• par2lar
• pca
• pitch
• smcep
• sopr

Note pca requires PyTorch 1.10+. Currently, only 16 kHz is supported in pitch. MLSA filter will be implemented in the next release.