SPTK

This repository hosts a modified version of Speech Signal Processing Toolkit (SPTK) to provide a SPTK library rather than CLI tools so that it can easily be used from FFI interfaces (e.g., cython or cffi in Python).

This version of SPTK is used by the following bindings:

• r9y9/SPTK.jl
• r9y9/pysptk
• r9y9/gossp

Note that if you want the command line programs of SPTK, please check the original one instead.

Supported Platforms

• Linux
• Mac OS X
• Windows

Changes

The important changes from the original SPTK are summerized as follows:

• Integrate waf build tool
• Add swipe inteface to SPTK.h SPTK.h#L382
• Windows 32/64 bit support. DLLs are avilable on the release page (https://github.com/r9y9/SPTK/releases)
• Avoid exit inside a function, as possible.
• Remove wavsplit and wavjoin from target sources to compile for cross-platform compilation ability, since original wavsplit and wavjoin use direct.h that only exists in POSIX enviroments. (#8)
• Add DLLEXPORT macro to expose API functions explicitly in MSVC environments (NOTE: this is not fully used for now) (#8)
• Ensure c89 compatibility
• Export excite function (#11)
• Cleanup and export RAPT function (#12)

Installation

Type the following commands from terminal:

./waf configure
./waf
sudo ./waf install

License

Modified BSD