Complex and non-linear transformations of speech signal when we change our speaking style - need to speak faster or communicate in a loud environment - present a serious challenge for theoretical speech science as well as speech technology. The project addresses this difficult research problem by exploiting synergies between phonetics and speech synthesis. A combination of state-of-the-art optimization-based modeling, hierarchical signal analysis and statistical parametric synthesis tools will be adapted and developed. The project builds on the ground-breaking research methodology originated by the Phonetics and Speech Synthesis Group of the University of Helsinki and combining theoretical and technological aspects of speech research. The effort will lead to a platform providing theoretical account of how humans vary and modify their speaking style. In addition, the research program has a potential to prepare the way for novel applications in speech technology and industry.

2017

2022