Source code and tools

• HTK Tools: A tool kit for training and testing Hidden Markov Models in speech recognition applications. This package was originally designed and now maintained by Speech Vision and Robotics Group Cambridge University, UK. Source code is also available for this package. For use in this course, this software is also installed on the departmental linux cluster (in /usr/local/htk/bin.linux), along with documentation (in /usr/local/htk/HTKBook).

• Festival Speech Synthesis System: A tool kit for speech synthesis applications. This package was originally developed at the Centre for Speech Technology Research, Cambridge University, UK. Source code and an inventory of examples for English sounds is also available with this package.

• Pipewave: A collection of pipeable modules by John Culling for signal processing and auditory modeling.

• GMTK, using graphical models for speech, language, and time series.

• LutEar: Supposedly an excellent auditory modelling package.

• Misc tools:matlab functions for reading WAV and AU.

Journal Contents/Abstracts

• Publications of Acoustic Society of America.

• John Krantz's list of psychologically-related electronic journals and periodicals etc.

• A journal on music perception.

Other sites

• SineWave Speech Page: An interesting speech synthesis page at Haskins Lab, Yale University. This is the source of the sine wave synthesis that we listened to in class.

• SOCRATES Thematic Network in Phonetics and Speech Communication: It has an inventory of internet resources including useful annotated indexes of linguistic data, speech processing tools, etc.

• WWW Information for Speech/Acoustics Research: This an large index from Nara Institute of Science and Technology.

• Mini tutorial on voice production: from University of Iowa.