Hi everyone I've been following this site for a while. This is my first post.

My research area is speech synthesis. To be more specific, speech analysis-synthesis, or vocoding. In the past few years I wrote a few speech processing libraries and made them open source on Github. You'll probably find some of them useful if you're doing speech, audio or other 1 dimensional signal processing.

• ciglet (in C, BSD): DSP code snippets - https://github.com/Sleepwalking/ciglet
• libgvps (in C, BSD): various versions of Viterbi algorithm - https://github.com/Sleepwalking/libgvps
• libpyin (in C, BSD): implementation of probabilistic YIN fundamental frequency tracking algorithm - https://github.com/Sleepwalking/libpyin
• libllsm (in C, GPL): an extensively enhanced harmonic+noise model based vocoder with pseudo-source-tract-decomposition - https://github.com/Sleepwalking/libllsm
• liblrhsmm (in C, GPL): implementation of left-to-right hidden semi-Markov model inference and training algorithms - https://github.com/Sleepwalking/liblrhsmm
• SHIRO (in C and Lua, GPL): toolkit for automatic phoneme alignment/segmentation, based on liblrhsmm - https://github.com/Sleepwalking/SHIRO

I'd be glad if these would help!

Kanru Hua

University of Illinois at Urbana-Champaign

Email: k.hua.kanru (at) ieee (dot) org